Marx versus Walras on Labour Exchange

Motohiro Okada

Abstract: This paper compares Léon Walras’s and Marx’s thoughts on labour exchange, thereby illuminating the latter’s perspective that can lead to a forceful counterargument to the neoclassical principle of labour exchange, for which the former affords a foundation. Both Walras and Marx distinguish between labour ability as a factor of production and labour as its service, but exhibit a striking contrast in their explanations of the distinction. Walras’s distinction between ‘personal faculties’ and labour never attempts to re- veal the peculiarities of the relationship they share. Walras essentially equates the re- lationship between the two with that between non-human factors and their respective services by stripping the former of human elements. This not only allows labour ex- change to be incorporated into Walras’s general equilibrium system but also provides the groundwork for its neoclassical principle, which, on the basis of marginal theory, assumes work conditions to be determinable through the stylised market adjustment of the demand and supply of labour on each entrepreneur’s and worker’s maximisa- tion behaviour. In contrast, especially in his pre-Capital writings, Marx underlines the worker’s subjectivity in deciding her labour performance. This implies that the type and inten- sity of time-unit labour varies depending on the worker’s will and the constraints upon it. Accentuating the particular characteristics of the relationship between labour power and labour in this way, Marx’s arguments lead to the invalidation of the neo- classical principle of labour exchange and rationalise the intervention of socio-politi- cal factors represented by the labour-capital class struggle in the determination of work conditions. Thus, this study focuses on the potential of Marx’s labour power-la- bour distinction independent of his exploitation theory-the basis of a weighty refu- tation of the neoclassical system. JEL classification numbers: B 13, B 14.

Walras’s views on labour exchange, and I Introduction thereby shed light on an aspect of Marx’s The objective of this study is to attempt a thinking that has received scant attention but comparison between Karl Marx’s and Léon can lead to a forceful counterargument to the

The History of Economic Thought, Vol. 52, No. 2, 2011. Ⓒ The Japanese Society for the History of Eco- nomic Thought. Okada: Marx versus Walras on Labour Exchange 47 neoclassical economic doctrine. contemporary forerunners of the neoclassical Marx’s economics was formed through school are required because such investiga- his in-depth study of the classical school. His tions have lagged behind the vast enquiries value and exploitation theory, which consti- into the Marx-classical economist relation- tutes the kernel of his economic thought, is ship. purely a product of his criticisms of and in- On the basis of this viewpoint, this study heritances from the works of David Ricardo investigates the worth of Marx’s views on la- and other classical economists. Although, in bour exchange in comparison with those of contrast, the Marginal Revolution writings of Walras, who played a leading role in the William S. Jevons, Carl Menger, and Walras Marginal Revolution and originated the gen- appeared during his life, Marx died without eral equilibrium theory. I have chosen Wal- having an opportunity to learn neoclassical ras as Marx’s match from among numerous economics, which they pioneered. However, neoclassical economists not only for his im- even if Marx had had this knowledge, it may portance and contemporaneity with Marx but have made little impact on his thinking. It also because the two distinguish between la- seems fairly reasonable to imagine that Marx bour ability as a factor of production and la- would have swept the neoclassical doctrine bour as its service. The distinction between aside as a mere sophisticated version of ‘vul- labour power（ Arbeitskraft） and labour（ Ar- gar’ economics. beit）, with the criticism of the obscurity in However, it is an unshakable fact that ne- classical economists’ notion of this issue, un- oclassical economists advanced analytical derlies Marx’s exploitation theory. Yet the techniques and minuteness significantly from comparison between Marx’s arguments on the pre-marginalism ‘vulgar’ economists labour exchange, especially in his pre-Capi- whom Marx savaged, thus having built the tal writings, and Walras’s demonstrates that predominant theoretical system. So, to re- Marx’s labour power-labour distinction- evaluate Marx’s thought with due considera- quite independently of his concept of exploi- tion of the history of economics after his tation-has the potential for a weighty refu- death, we must investigate whether his argu- tation of the neoclassical rather than the clas- ments contain what may qualify as intrinsic sical system. criticisms of neoclassical economics. Long Disregarding factor-service relations pe- discussions on Marx’s value and exploitation culiar to labour exchange, Walras sees work theory, which has been widely recognised as conditions as determinable through the styl- representing Marxian conflicts with the neo- ised market adjustment of the demand and classical school, suggest that it is inadequate- supply of labour on each entrepreneur’s and ly qualified.1） Thus, comparative studies of worker’s maximisation behaviour. This intro- Marx’s and neoclassical economics from a duces the neoclassical manner of explaining broader perspective may be needed to reap- even labour exchange by the marginalism- praise Marx in the current situation. In his- based market theory. It may be called the torical research, it seems that close and wide- neoclassical principle of labour exchange.2） ranging comparisons between Marx and his Recognising the particular characteristics of 48 経済学史研究 52 巻 2 号

the relationship between labour power and II Walras on Labour Exchange labour in contrast to that between non-human factors and their respective services, Marx’s Walras’s theory of labour exchange is incor- views can reveal the groundlessness of the porated into his general equilibrium system neoclassical principle, and rationalise the in- in Elements of Pure Economics. In Part IV of tervention of socio-political factors repre- the book, Walras describes markets for ‘pro- sented by the labour-capital class struggle in ductive services’ in addition to those for the determination of work conditions. ‘commodities’ （consumption goods） expli- By thus opposing Marx against Walras, cated in the previous part, and presents si- this study, unlike the traditional value and multaneous equations to satisfy the equilibri- exploitation theory-centred comparison be- um of all service and product markets. Wal- tween Marx and neoclassical economists, il- ras begins this part by distinguishing be- luminates a facet of Marx’s thought which tween ‘capital’ and ‘income’ or service.3） Here can be an intrinsic and potent criticism of the he remarks: neoclassical paradigm and thereby justify his socio-economic views. The elementary factors of production are Section II deals with Walras’s treatment three in number. In listing these factors, of labour exchange. It will be shown here most authors employ the terms: land, la- that Walras’s distinction between ‘personal bour and capital. But these terms are not faculties’（ facultés personnelles） and labour sufficiently rigorous to serve as a founda- （travail） never leads him to recognise the tion for rational deduction. Labour is the vital difference between their relationship service of human faculties or of persons. and the nonhuman factor-service relation- We must rank labour, therefore, not with ship. As a result, Walras perceives no partic- land and capital, but with land services ular characteristics of the human labour ex- ［‘rente’］ rendered by land, and with capi- change. This contributes to the formation of tal services［ ‘profit’］ rendered by capital the neoclassical principle of labour exchange goods. （emphasis in original; L. Walras as well as Walras’s general equilibrium sys- ［1926］ 1954, 212） tem. Section III focuses on Marx’s discus- sions that stress the worker’s subjectivity in Thus, Walras stresses the distinction between the decision on her labour performance and human（ or personal） faculties and labour as the resulting peculiarities of the labour pow- their service. This perception is comparable er-labour relationship. It will thus be argued with Marx’s labour power-labour distinction. that, as opposed to Walras’s personal facul- The similarity was previously pointed out by ties-labour distinction, Marx’s labour power- Mark Blaug（ 1996, 258）. Walras mentions labour distinction involves the grasp of the that he follows The Theory of Social Wealth human elements of labour exchange, and this by Auguste Walras, his father, regarding his insight leads to a cogent denial of its neo- definition of capital and income, as referred classical principle. Section IV concludes the to in note 3（ see L. Walras［ 1926］ 1954, paper. 212）. In that book, indeed, Auguste supplied Okada: Marx versus Walras on Labour Exchange 49

the same definition（ see A. Walras［ 1849］ of labour power （see L. Walras ［1926］ 1997, 140）. Furthermore, Auguste distin- 1954, 237-40, 271; Marx ［1867］ 1996, guished between personal faculties and la- 180-83）. bour before his son: ‘Labour, it is the income Walras states that the amount of labour is from personal faculties’（ Le travail, c’est le measurable in terms of labour time per capita revenu des facultés personnelles） （see A. （see L. Walras［ 1926］ 1954, 237）. This im- Walras［ 1849］ 1997, 157）.4） Although Wal- plies that each time-unit use of a personal ras’s personal faculties-labour distinction faculty provides the same service. Hence, the was thus an inheritance from his father, it is variability of the type of labour（ which is noteworthy because Jevons, Menger, and limited by the worker’s abilities but ought to most other early neoclassical economists did be considered different even if assuming one not specify a similar distinction. and the same personal faculty） and its inten- For all this resemblance, Walras’s treat- sity is precluded here. As will be noted in the ment of labour exchange otherwise exhibits next section, Marx perceives this variability a striking contrast with Marx’s. as rooted in the nature of labour power, While Marx maintains that labour per se which is inalienable and so whose perform- cannot be an object of market exchange, ance is contingent on the worker’s will. Wal- ‘services of persons’ as well as personal fac- ras also remarks: ［‘ P］ersonal faculties are ulties are supposed to be marketable in the inalienable capitals’ （les facultés person- Elements. Walras assumes that personal fac- nelles sont des capitaux intransmissibles） ulties are classifiable and quantifiable just （emphasis in original; see L. Walras［ 1877］ like land and capital goods （see Marx 1993, 75）. This observation, however, is ［1867］ 1996, 535-42; L. Walras ［1926］ scarcely reflected in Walras’s theory of la- 1954, 271-72）. In this fashion, Walras rec- bour exchange. Walras assumes that services ognises the diversity of labour ability or skill. from land, personal faculties, and capital As to land and capital goods, they can be goods can be used for personal consumption classified and quantified on the basis of their as well as production（ see L. Walras［ 1926］ physical properties. Yet in what way can per- 1954, 237-38）. Of labour retained or pur- sonal faculties, which are generally irreduci- chased for their own consumption and that ble to material attributes, be classified and sold to others, as Ugo Pagano （1985, quantified? Without giving any account of 100-01） notes, it is only the former that this issue, Walras quantifies each kind of Walras presupposes to directly affect indi- ‘services of persons,’ or labour, and embodies viduals’ welfare. This can be seen from the its quantity in production functions with system of equations of equi-marginal utility fixed input coefficients in the Elements. He （‘rareté’ in Walras’s terminology）-price ra- also explains that the price of each of the tio for all products and services that Walras personal faculties can be estimated by dis- presents as required for each individual’s counting on the price of the corresponding maximum satisfaction. Here, Walras includes labour. This concept differs markedly from the amount of labour retained or purchased Marx’s reproduction cost theory of the value for each individual’s own consumption in the 50 経済学史研究 52 巻 2 号 argument of her utility function, but not that Then, time-capita, chosen as a labour unit in sold to others（ see L. Walras［ 1926］ 1954, the Elements, is definite enough to be recog- 237-38）. Pagano（ 1985, 110-11） interprets nisable for all market participants, and has this as ‘tantamount to assuming that the indeed been generally adopted as a labour workers are indifferent among alternative al- unit by neoclassical economists. Time-capita, location of their manpower in production- however, can only become a labour unit suit- that is, indifferent to the kind of work they ing the above requirement of marginal pro- perform.’ In fact, it can rather be observed ductivity theory if each time-unit labour of that Walras’s theory excludes such workers’ all workers exercising the same personal fac- preference by presuming the identity of serv- ulty, to the exclusion of its variability, is as- ice out of each time-unit use of a personal sumed to provide the same service. Ruling faculty. Yet the preference as to the type and out the worker’s choice of the type and inten- intensity of labour, which undoubtedly af- sity of her own labour entailed by the inal- fects the worker’s welfare, is vital for her as ienability of her personal faculties, this as- a possessor of personal faculties and will. sumption weakens human elements of labour Hence, this omission creates a crucial imper- exchange and, as a result, equates its frame- fection in Walras’s theory of labour ex- work with that of the exchange of land and change. capital goods services, in which each factor It may well be said, however, that the and its service are in unique correspondence. above treatment, as in the Elements, affords a True, the Elements theorises that the price pivotal foundation for the neoclassical prin- （wage rate） and amount of each kind of la- ciple of labour exchange. On the one hand, bour are determinable through the stylised Walras himself did not subsume marginal market adjustment of its demand and supply productivity theory, which underlies the on each agent’s （entrepreneur’s and work- principle, into his general equilibrium sys- er’s） maximisation behaviour, in exactly the tem.5） Marginal productivity theory, on the same way as the price and amount of non- other hand, requires such a metric definition human services and products are determined of each sort of input as can guarantee the （see L. Walras［ 1926］ 1954, 222-26）. identity of service from each input unit of It may be thus concluded that Walras’s the same sort. Additionally, the input unit personal faculties-labour distinction is made must be such that it can be commonly recog- never with the intention of explaining the pe- nised by the demanders and the suppliers in culiarities of labour exchange, but solely as a order for the trade to be conducted in the means of his distinguishing between factors market place. The neoclassical principle of of production（ ‘capitals’ in his phraseology） labour exchange hinges on the finding of a and their services in general. The factor- labour unit which meets these requisites. service demarcation is necessary for the de- Here the measurement of labour by its prod- velopment from the general equilibrium sys- uct contradicts the marginal productivity tem as regards product and service markets theory, whose groundwork lies in the stipula- in Part IV of the Elements into the system tion of the causality from input to output. that adds ‘capital’ markets in Part V. Walras’s Okada: Marx versus Walras on Labour Exchange 51

theoretical contribution in this respect, per la volonté des autres hommes, autrement dit se, should not be underrated. Walras, howev- les rapports de personnes à personnes）-as er, places the relationship of personal facul- determining factors of work conditions（ see ties to their services and that of land and L. Walras 1993, 487）. capital goods to their non-human services in On the other hand, Walras notes that the the same category. Thus formed the new equilibrium of product and service markets ‘Trinity Formula,’ upon which was grounded depicted in the Elements is ‘an ideal and not the neoclassical principle of labour ex- a real state.’ Yet he adds: ［‘ E］quilibrium is change. Philip H. Wicksteed, who advocated the normal state, in the sense that it is the the universality of the marginal productivity- state towards which things spontaneously based distribution rule, expressed the logical tend under a regime of free competition in conclusion of this direction given by Walras exchange and in production’（ see L. Walras to economics forthrightly: ‘The crude divi- ［1926］ 1954, 224）. The labour market is no sion of the factors of production into land, exception here. Furthermore, Walras attaches capital and labour must . . . be abandoned’ ethical value to free competition among （see Wicksteed［ 1894］ 1992, 83）. workers and entrepreneurs. In Economics Walras’s ‘dehumanisation’ of labour ex- and Justice, Walras remarks that workers change may be predictable from his method- and entrepreneurs are placed on the same ology. Walras regards ‘pure economics’ deal- footing（ placés sur le même pied） in free ing with the determination of exchange value competition （see L. Walras ［1860］ 2001, as a natural science （see, for example, L. 186）. In Studies in Applied Economics, Wal- Walras ［1860］ 2001, 151-53; 1993, 487- ras states that a ‘just wage’ is what corre- 88）. In Economics and Justice he argues: sponds to the equilibrium of labour demand ‘Natural facts will be . . . distinguished from and supply as a result of free exchange（ see moral facts in that the former will have their L. Walras［ 1898］ 2005, 221）. While Walras origin in the fatality of natural forces, and the argues for the nationalisation of land, he sees latter in human free will’（ Les faits naturels little need for the regulation of labour ex- se distingueront . . . des faits moraux en ce change. As opposed to Marx, who stresses que les premiers auront leur origine dans la the difference between the value created by fatalité des forces naturelles, les seconds labour and the value of labour power, Walras dans la volonté libre de l’homme）（ see L. asserts the ‘equivalence of labour and wage’ Walras［ 1860］ 2001, 152）. In Walras’s no- （équivalence du travail et du salaire） deter- tion, labour exchange is also subject to ‘the mined under free competition（ see L. Walras fatality of natural forces.’ So, unlike Marx, ［1860］ 2001, 185-86）. Walras has no Marx- Walras negates the role of class struggle and ian-type concept of exploitation. He sees other ‘moral facts’-or ‘those which result capital accumulation as the fruit of labour from human will being exercised on other and savings, and denies the conflicting na- humans’ will; in other words, relations of ture of the labour-capital relationship people to people’（ ceux qui résultent de la claimed by socialists（ see L. Walras［ 1860］ volonté de l’homme s’exerçant à l’endroit de 2001, 522; 1996, 582-87）. 52 経済学史研究 52 巻 2 号

Hence, Walras is naturally hostile to ideal would be inaccessible not merely be- workers’ industrial actions, in which he in- cause of the lack of workers’ or other agents’ sists that the state should rationally intervene edification, but due to particular characteris- to suppress strikes for the purpose of the bet- tics inherent in labour exchange. Although ter working of product and service markets. he had some knowledge of Marx’s economic Walras observes that mere laissez-faire is not thought, Walras did not pay attention to conducive to the cause （see L. Walras Marx’s labour power-labour distinction, ［1898］ 2005, 223-24, 372; 1987 b, 510）. In which bore a similarity to Walras’s personal addition, Walras recognises that his demon- faculties-labour distinction, but led to a dis- stration of the superiority of free competition parate view on labour exchange by examin- is premised on economic agents’ high capa- ing its essence. Consequently, Walras’s com- bility, ‘the supposition . . . that consumers ments on Marx largely consisted of criti- know their needs and producers their inter- cisms of Marx’s labour theory of value, as est; workers are serious, and sufficiently in- did many other early neoclassical econo- structed and moral; entrepreneurs are truly mists’（ see, for example, L. Walras［ 1936］ enlightened people’（ la supposition . . . que 2010, 149-51）. les consommateurs connaissent leurs besoins, et les producteurs leur intérêt; que les ouvri- III Marx’s Distinction between ers sont sérieux, suffisamment instruits et Labour Power and Labour: moraux; que les entrepreneurs sont des hom- Another Possibility mes véritablement éclairs）（ see L. Walras In his early works, Marx argued that the 1987 a, 221-23）. This seems to suggest that worker-capitalist power relationship or their Walras had already discerned stringent req- class strife is the major determinant of wag- uisites for the ‘perfect competitive market.’ es.7） Marx’s stance on labour exchange was Yet Walras never changed his basic con- thus opposed to Walras’s from the beginning. ception of labour exchange. In Studies in Ap- Marx’s mature system of economic thought, plied Economics, Walras writes: ‘Why are especially his theory of exploitation based on there no strikes between entrepreneurs and the labour theory of value, was sketched out capitalists? It is because the market in fixed in the Grundrisse, or the Economic Manu- capital（ the Bourse） and circulating capital script of 1857-58, and consummated in Cap- （the banks） have been organized somewhat ital. In this sense, as Allen Oakley（ 1979, better than the labour market, which is not 287） puts it, ［‘ The Grundrisse］ provided a organized at all’ （see L. Walras ［1898］ foundation for Marx’s later critico-theoretical 2005, 225）. This implies that the more the development, but no more’ （emphasis in labour market becomes ‘organised,’ the closer original）. Meanwhile, Michael A. Lebowitz it comes to the ideal state described in the （2003） accentuates the ‘one-sidedness’ of Elements. For this object, Walras emphasises Capital. Lebowitz observes ‘Capital is essen- the need for workers’ cultivation（ see, for tially about capital,’ noting that ‘Capital example, L. Walras 1996, 585-86; 1987 a, looks upon the worker from the perspective 223-24）.6） Here, he never perceived that the of capital（ that is, as an object for capital Okada: Marx versus Walras on Labour Exchange 53 rather than as a subject for herself）’ （see tality of the subject itself and manifests it- Lebowitz 2003, ix, 104）. Actually, in this re- self only as his own expression of life, the gard the Grundrisse and other pre-Capital buying of the labour capacity, the appro- writings contain descriptions that attach priation of the title to its use naturally more importance to the worker’s autonomy places the buyer and the seller in the act of than does Capital. In addition, they present its use in another relationship to each other diverse discussions on the distinction be- than that in the buying of objectified la- tween labour power and labour; in Capital, bour existing as an object outside the pro- however, the distinction is exclusively posi- ducer（. Marx 1987, 506） tioned as the groundwork for the exploitation theory. Although those discussions in Marx’s Thus, Marx underlines the difference be- pre-Capital writings are rather fragmentary, tween the purchase of labour capacity and they can be integrated into a forceful coun- that of ‘objected labour,’ or commodities in terargument to the neoclassical principle of general, with regard to the acquisition of labour exchange. their use value, because of the fact that ‘the In the original text of A Contribution to labour capacity exists in the vitality of the the Critique of Political Economy, Marx subject itself and manifests itself only as his writes: own expression of life.’ Marx describes the particular characteristics of labour exchange As use value, the labour capacity is real- which this entails in the Economic Manu- ised only in the activity of labour itself, but script of 1861-63 and the Grundrisse, re- in much the same way as with a bottle of spectively: wine which is bought and whose use value is realised only in the drinking of the wine. ［L］abour is . . . the expression of the Labour itself falls as little within the sim- worker’s own life, the manifestation of his ple circulation process as does the drink- own personal skill and capacity-a mani- ing. The wine as a capacity, dunamei, is festation which depends on his will and is something drinkable, and the buying of the simultaneously an expression of his will. wine is appropriation of the drinkable. So （Marx 1988, 93） is the buying of the labour capacity the ap- propriation of the ability to dispose over What the free worker sells is always only a the labour（. Marx 1987, 506）8） particular, specific measure of the applica- tion of his energy. Above every specific In this manner, Marx recognises that for both application of energy stands labour capaci- labour capacity and non-human commodities ty as a totality. The worker sells the specif- a demarcation must be made between them- ic application of his energy to a specific selves and their respective use value. How- capitalist, whom he confronts independ- ever, he continues: ently as a single individual. Clearly, this is not his［ real］ relationship to the existence Since the labour capacity exists in the vi- of capital as capital, i.e. to the class of cap- 54 経済学史研究 52 巻 2 号

italists. Nevertheless, as far as the individ- and outside the factory,’ in the postindustrial ual, real person is concerned, a wide field economy with the rising dominance of the of choice, caprice and therefore of formal service sector, immaterial labour creating freedom is left to him（. Marx 1986, 392） ‘immaterial products, such as knowledge, in- formation, communication, a relationship, or These excerpts show that in his pre-Capital an emotional response’ achieves hegemony writings Marx pays due attention to the and imposes a tendency even on agricultural worker’s subjectivity regarding her labour and industrial labour; ‘other forms of labor performance. In the second extract, from the and production are adopting the characteris- Grundrisse, Marx connotes that the worker’s tics of immaterial production.’ Here, Hardt subordination to the capitalist class does not and Negri suggest, workers are to incessantly annul her independence against her individu- change the mode of their labour as appropri- al employer. In the Elements, Walras presup- ate to their communication and interaction poses that the worker’s various personal fac- with others, just as the computer can continu- ulties are separable from each other as enti- ally modify its own operation through its ties, and that each time-unit use of each of use. Additionally, ‘the cooperative aspect of them provides the same service. In contrast, immaterial labor is not imposed or organized according to Marx, ‘Above every specific ap- from the outside, as it was in previous forms plication of energy stands labour capacity as of labor, but rather, cooperation is complete- a totality.’ In other words, Marx implies that ly immanent to the laboring activity itself’ labour capacity, a counterpart of Walras’s （emphasis in original; see Hardt and Negri personal faculties, is an indecomposable ele- 2000, 294）. In these circumstances, although ment of human abilities. Thus, unlike in the workers are still or even more subordinate to Elements, the worker originally has discre- their employers in the power relationship, tion over the entire mode of the use of her their subjectivity acquires crucial importance labour capacity as its inalienable owner. as to their labour performance, and so it is ir- Hence, to quote from the above excerpts, the relevant to assume the constancy of the type type and intensity of labour is variable as ‘a and intensity of per time-capita labour as manifestation which depends on his ［the Walras does de facto. worker’s］ will and is simultaneously an ex- Thus is drawn such a conclusion in the pression of his will,’ and so ‘a wide field of following passage from Capital: choice, caprice and therefore of formal free- dom is left to him’ on labour performance. One consequence of the peculiar nature of The above viewpoint, simple in itself, labour power as a commodity is, that its possesses profound significance for the use value does not, on the conclusion of present-day postindustrial economy. Michael the contract between the buyer and seller, Hardt and Antonio Negri（ 2000, 280-303; immediately pass into the hands of the 2004, 103-15） argue that in contrast to the former（. Marx［ 1867］ 1996, 184） modern industrial economy, where ‘workers learned how to act like machines both inside The severance of the buying and selling of Okada: Marx versus Walras on Labour Exchange 55 labour power and the buyer’s acquisition of ent’（ emphasis in original; see Marx 1986, its use value, or labour, mentioned here is, of 109）. course, not solely a matter of time lag. Its Accordingly, what remains as a candidate crux must be that the purchase of labour for the trading unit of the labour service mar- power or the labour contract as such does not ket is labour per se. In the Economic Manu- assure the buyer her actual acquirement of script of 1861-63, Marx argues this issue: any certain labour, owing to ‘the peculiar na- ture of labour power as a commodity,’ that is, Originally, it is true, we were able to meas- its inalienability and the resulting depend- ure labour capacity with money, because it ence of its use on the worker’s will or prefer- was itself already objectified labour, and ence and the constraints upon it; however, the capitalist could therefore buy it; but Capital shuns the accentuation of this feature were unable to measure labour itself di- of labour power. This severance is already rectly, for as bare activity it escaped our articulated in the Grundrisse （see Marx standard of measurement. Now, however, 1986, 204-05）. Thus, unlike Walras’s per- in the measure to which, in the labour sonal faculties-labour distinction, Marx’s process, labour capacity proceeds to its distinction between labour power and labour real manifestation, to labour, the latter is is not a mere factor-service demarcation, but realised, appears itself in the product as defines the particular characteristics of la- objectified labour time. . . . At the end of a bour exchange. certain measure of labour time, e.g. hours, In the Grundrisse, Marx notes: ‘In so far a certain quantity of labour time has been as it［ labour time］ is exchangeable in that objectified in a use value, say twist, and form（ is itself a commodity）, it is not only now exists as the latter’s exchange value. quantitatively but also qualitatively deter- （emphasis in original; Marx 1988, 83） mined and differentiated’（ see Marx 1986, 107）. Now that Marx’s labour power-labour This citation articulates the impracticability distinction implies that time-unit labour, un- of measuring ‘labour as bare activity,’ or la- like in Walras’s model, can provide quantita- bour in a concrete useful form, and suggests tively and qualitatively diverse service-or that abstract human labour alone can be concrete useful labour in Marx’s terms- quantitatively perceived through the ex- contingent on the worker’s preference and change value of the product.9） The denial of the constraints upon it, labour time as such the immediate measurability of labour-true, cannot be an adequate trading unit of the la- it may safely be said that except for very bour service market. Indeed, Marx writes: simple and mechanical work, the quantitative ‘Labour time itself exists as such only sub- perception of concrete useful labour in gen- jectively, only in the form of activity’（ see eral exceeds human capability-involves the Marx 1986, 107）. Furthermore, Marx ex- negation of rendering labour per se the trad- plains that even in the same branch of indus- ing unit of the labour service market.10） try ［‘ t］he labours of individuals . . . are not In consequence, no adequate trading unit only quantitatively but qualitatively differ- of the labour service market can be found in 56 経済学史研究 52 巻 2 号 either labour time or labour per se, and so a tween the working men and the capitalists. requisite for the formation of the market, that This very necessity of general political ac- is, the definition of its appropriate trading tion affords the proof that in its merely unit, proves absent. Thus, Marx’s discussions economic action capital is the stronger lead to the refutation of the neoclassical prin- side.（ emphasis in original; Marx［ 1898］ ciple of labour exchange asserting the mar- 1985, 146） ket determinability of work conditions. That, in turn, allows room for the intervention of By this logic, Marx’s labour power-labour ‘extra-economic’ factors, especially the work- distinction-based arguments legitimise his er-employer power struggle, in the settle- other observation, the role of labour-capital ment of work conditions. There is no reason class conflict as a prime determinant of work to deny that the struggle does not remain in- conditions. Note that the whole reasoning dividual, but unfolds socio-politically as la- here stands quite independent of the concept bour-capital class strife. In Value, Price and of exploitation as the main contention of Profit, indeed, Marx states: Marx’s labour power-labour distinction-the difference between the value created by la- The maximum of profit is . . . limited by bour and the value of labour power as the the physical minimum of wages and the source of profit. physical maximum of the working day. It On the other hand, Marx explains how is evident that between the two limits of the development of mechanisation in capital- this maximum rate of profit an immense istic production deprives workers of skills scale of variations is possible. The fixation and resistance, thereby giving capitalists of its actual degree is only settled by the overwhelming dominance（ see, for example, continuous struggle between capital and Marx［ 1867］ 1996, 374-509）. The charac- labour, the capitalist constantly tending to terisation of the worker ‘as an object for cap- reduce wages to their physical minimum, ital’ in Capital arguably reflects the ‘full- and to extend the working day to its physi- fledged’ stage of capitalism in this respect. cal maximum, while the working man con- The adverse effect of mechanisation on the stantly presses in the opposite direction. worker is mentioned also in Marx’s pre-Cap- The matter resolves itself into a question ital writings（ see, for example, Marx 1986, of the respective powers of the combat- 41; 1988, 318-46）. Yet they simultaneously ants. demonstrate the prototype of labour ex- . . . As to the limitation of the working day change, where the worker acts as an inde- in England, as in all other countries, it has pendent agent towards the employer. never been settled except by legislative in- IV Concluding Remarks terference. Without the working men’s continuous pressure from without that in- This study has examined Walras’s and then terference would never have taken place Marx’s thoughts on labour exchange in their ［sic］. But at all events, the result was not comparisons, and illuminated the latter’s dis- to be attained by private settlement be- cussions that can lead to a potent argument Okada: Marx versus Walras on Labour Exchange 57 against the neoclassical principle of labour dation of its neoclassical principle and ra- exchange, for which the former affords a tionalises his other observation-the inter- foundation. As an inheritance from his father, vention of socio-political factors represented Walras distinguishes between personal facul- by the labour-capital class struggle in the ties and labour. This distinction, however, determination of work conditions. never aims to reveal the peculiarities of the In this manner, Marx’s discussions on la- relationship between the two; rather, Walras bour exchange attach importance to the essentially equates it with the relationship worker’s preference regarding the content of between non-human factors and their respec- her labour. Walras and other neoclassical tive services by stripping the former of hu- economists, as utility-oriented theorists, man elements. This treatment not only al- should take due account of this preference, lows labour exchange to be incorporated into but they ignore it. This contrast causes Walras’s general equilibrium system, but also Marx’s thoughts on labour exchange to em- provides the groundwork for the marginal brace forceful and intrinsic criticisms of the principle-based neoclassical labour exchange neoclassical doctrine.11） However, Marx’s fo- doctrine, which deduces the market determi- cus on the worker’s subjectivity nearly disap- nability of work conditions through the con- pears in Capital. Here, in principle, labour ventional adjustment of the demand and sup- power is treated as a commodity, used at the ply of labour on each entrepreneur’s and bidding of the capitalist. Arguably, this worker’s maximisation behaviour. change is attributable to the fact that the la- Although only fragmentarily, in his pre- bour power-labour distinction in Capital is Capital writings Marx underlines the subjec- exclusively identified as the cornerstone of tivity of the worker, qua an inalienable pos- the principle of exploitation based on the la- sessor of labour capacity, in the decision on bour theory of value.12） As a result, pushing her labour performance. As Marx deduces, abstract human labour to the fore, Marx in this implies that unlike in Walras’s model, Capital marginalises concrete useful labour where each time-unit use of a personal facul- as irrelevant to value and exploitation, and ty is assumed to provide the same service, tends to underestimate labour in the service the type and intensity of time-unit labour is sector as producing no surplus value. Yet it variable depending on the worker’s will and is concrete useful labour, not abstract human the constraints upon it. Hence, labour time labour, whose concept is shared by neoclas- cannot be an adequate trading unit of the la- sical economists and which concerns the bour service market. Marx also denies the worker’s preference. Additionally, the service immediate measurability of labour per se. industry has become increasingly dominant Accordingly, in general there exists no ap- in today’s economy. propriate trading unit in terms of which the Thus, while the establishment of Marx’s labour service market is formed. Thus, high- value and exploitation doctrine in Capital lighting the particular characteristics of la- marked the limit of his studies and criticisms bour exchange, Marx’s distinction between of classical economics, this doctrine shoved labour power and labour leads to the invali- aside those viewpoints in his pre-Capital 58 経済学史研究 52 巻 2 号 writings that could lead to convincing coun- in this study will have more impact than a terarguments to the neoclassical theory. Most mere historical retrospect. Marxists up to the present followed the di- Motohiro Okada: Faculty of Economics, rection established in Capital. Historical re- Konan University search on Marx’s economics, too, centred around topics as to the value and exploitation Notes doctrine. The same is true of the comparison 1） As is well known, Eugen v. Böhm- between Marx and neoclassical economists . Bawerk（ ［1896］ 1984） pioneers the debates This made the comparative study of them re- concerning the transformation problem by main secondary to the study of the Marx- indicating the discrepancy between the theo- classical economist relationship, despite the ry of value developed in Volume I of Capital fact that more than a century has passed and that of the price of production in Volume since the neoclassical school obtained he- III. Moreover, also famously, Nobuo Okishio gemony. Consequently, critics on the neo- （1963） and others prove that the so-called classical side asserted that Marx made little fundamental Marxian theorem, or the capi- talist’s exploitation of labour as a requisite positive contribution to the history of eco- for positive profit, holds good independently nomics and so mainstream economists might of the labour theory of value. Having rigor- well neglect his economics（ see, for exam- ously examined exploitation, John E. Roemer ple, Brewer 1995）. （1982） shows that the demonstration of ex- By giving attention to Marx’s discussions ploitation in the Marxian sense becomes im- on the labour power-labour distinction in his possible once the heterogeneity of labour is pre-Capital writings and contrasting them assumed. Samuel Hollander （2008）, with with Walras’s counterparts, this paper revises close reviews of Marx’s writings, concludes the above tendency and shows that Marx’s his theories of value and exploitation to re- sult in their submission to the orthodoxy. thinking contains an element which, inde- 2） This principle is explicitly stated in most pendent of his labour theory of value and modern economics textbooks（ see, for exam- conception of exploitation can be an intrin- , ple, Mankiw 1998, 384-92）. For literature sic and potent criticism of the paradigm of criticising neoclassical concepts of labour neoclassical economics and thereby justify exchange, see, for example, Herbert Gintis his socio-economic views. This also urges （1976） and Geoffrey M. Hodgson（ 1980）. the re-evaluation of the development of 3） Walras states: ‘Since capital, by definition, Marx’s economic thought. outlasts its first use and consequently affords Indeed, given that present-day capitalism a series of successive uses, the flow of uses is characterised by the globalisation of a evidently constitutes a flow of income. . . . In order to bring out the distinction between market-oriented economy, which the neo- capital and income, we shall designate all classical tenet indisputably supports ideolog- those incomes which consist in the uses ically, and that Marxian thinking along the made of capital by the name of services’ traditional line reaches a grave stalemate to- （emphasis in original; see L. Walras［ 1926］ day, the reappraisal of Marx’s pre-Capital 1954, 213）. discussions on labour exchange introduced 4） About Auguste’s influence on his son Okada: Marx versus Walras on Labour Exchange 59

concerning terminology and definitions, see, ence and the constraints upon it also forms a for example, William Jaffé（ 1935, 190-92）. basis of the efficient wage theory（ see, for 5） Walras did not refer to the marginal pro- example, Yellen 1984; Shapiro and Stiglitz ductivity theory in the first three editions of 1984）. Samuel Bowles and H. Gintis（ 1990, the Elements. He did so in Lesson 36 of the 314） remark that the idea of the efficiency fourth edition of the book（ see L. Walras wage theory originated in Marx’s analysis of ［1926］ 1954, 382-92）. Kayoko Misaki the extraction of labour from labour power. （1998, 68-71） emphasises Walras’s inten- Efficiency wage theorists, however, scarcely tion to introduce marginal productivity theo- offer a specific explanation as to the sub- ry as an analytical tool of the determination stance of labour, or what they call effort. of minimum average cost-bringing coeffi- Nevertheless, they quantify effort and, more- cients of production rather than as distribu- over, de facto presuppose that both the work- tion doctrine. er and the employer have the ability to grasp 6） Walras complains: ‘Workers have no ex- effort level. This also holds for Bowles and act notion of the mechanism of the determi- Gintis’s ‘neo-Marxian’ contested exchange nation of prices of products and productive model, which embraces the efficiency wage services through free competition’ （Les concept. The model assumes that the em- ouvriers n’ont point une notion exacte du ployer sets the maximum profit-bringing mécanisme de la détermination des prix des wage level based on her knowledge of the produits et des services producreurs par la li- worker’s effort response to wages, or the ‘la- bre concurrence）（ see L. Walras 1996, 581）. bor extraction function’ （see Bowles and 7） ‘Wages are determined through the antago- Gintis 1990, 179）. If so, then there will be no nistic struggle between capitalist and worker’ denying that bargaining can be conducted in （emphasis in original; Marx 1975, 235）. ‘The terms of effort, which is supposed to be size of wages is determined at the beginning quantitatively perceived by both the worker by free agreement between the free worker and the employer. Contrary to Bowles and and the free capitalist. Later it turns out that Gintis’s intention, this admits the buying and the worker is compelled to allow the capital- selling of labour per se negated by Marx, and ist to determine it, just as the capitalist is results in turning back the contested ex- compelled to fix it as low as possible. Free- change model to the conventional labour dom of the contracting parties has been sup- market model（ see Bowles and Gintis 1990, planted by compulsion’（ emphasis in origi- 180-82）. Criticising efficiency wage theo- nal; Engels and Marx［ 1845］ 1975, 32-33）. rists and others’ notion that effort is cardinal- 8） As in this quoted passage, Marx, in pre- ly measurable, Martin Currie and Ian Steed- Capital writings, generally used the word ‘la- man（ 1993, 136） write: ‘What are the units bour capacity’（ Arbeitsfähigkeit or Arbeits- in terms of which care, attentiveness and ini- vermögen） in lieu of labour power. tiative are to be measured?’ （emphasis in 9） ‘As an intrinsic element, each concrete original）. In all truth, ‘care, attentiveness and useful labor has a specific form and aim; it is initiative’ are important work categories in not measurable in quantity, and is distin- today’s service production-dominated econo- guishable only in qualitative terms’ （Park my. 2003, 163）. 11） Pagano （1985, 50-59, 114-15） argues 10） The observation that labour intensity is persuasively that Marx considered commu- variable depending on the worker’s prefer- nist society to be a world where the worker’s 60 経済学史研究 52 巻 2 号

welfare obtained from the content of work, and the Close of His System. In Karl Marx to which she is originally sensitive but which and the Close of His System by Eugen von is undermined in capitalist and even early Böhm-Bawerk & Böhm-Bawerk’s Criticism socialist society, can be sought to the fullest of Marx by Rudolf Hilferding, edited by P. extent, whereas Walras supposed the work- Sweezy. Philadelphia, PA: Orion Editions: er’s indifference to the content of work to be 9-118. primordial（ about this matter, see, in particu- Bowles, S. and H. Gintis. 1990. Contested Ex- lar, Marx 1989, 85-87）. Hodgson （1980, change: New Microfoundations for the Po- 265） reveals the contradiction inherent in litical Economy of Capitalism. Politics & neoclassical economics that, while assuming Society 18（ 2）: 165-222. the worker’s ‘free’ will in the labour contract, Brewer, A. 1995. A Minor Post-Ricardian? neoclassical economists necessarily negate Marx as an Economist. History of Political that will in order to admit the appropriation Economy 27（ 1）: 111-45. of labour like that of a machine’s services. At Currie, M. and I. Steedman. 1993. Taking Effort least this criticism may hold for Walras’s Seriously. Metroeconomica 44（ 2）: 134-35. thought. See also David A. Spencer（ 2009, Engels, F. and K. Marx［. 1845］ 1975. The Holy 69-93） as an insightful investigation into the Family. In The Collected Works of Karl difference among neoclassical economists in Marx and Frederick Engels, Vol. 4. New their treatment of work. York: International Publishers: 5-211. 12） One of the most critical problems with Gintis, H. 1976. The Nature of Labor Exchange Marx’s ‘commoditisation’ of labour power and the Theory of Capitalist Production. Re- arises when he supposes that the value and view of Radical Political Economics 8（2）: price law regulating commodities in general 36-54. also holds for labour power（ see, for exam- Hardt, M. and A. Negri. 2000. Empire. Cam- ple, Marx［ 1898］ 1985, 144;［ 1867］ 1996, bridge, MA: Harvard Univ. Press. 180-81）. Marx does not specify how this no- -. 2004. Multitude: War and Democracy in tion can harmonise with the following view the Age of Empire. New York: Penguin in Capital: ‘In contradistinction . . . to the Books. case of other commodities, there enters into Hodgson, G. M. 1980. A Theory of Exploitation the determination of the value of labour without the Labour Theory of Value. Science power a historical and moral element’（ see & Society 44（ 3）: 257-73. Marx［ 1867］ 1996, 181）. Blaug（ 1996, 258） Hollander, S. 2008. The Economics of Karl gives a pertinent comment on this passage: Marx: Analysis and Application. New York: ［‘ Marx］ fails to point out that competition Cambridge Univ. Press. provides no mechanism to reduce the “mar- Jaffé, W. 1935. Unpublished Papers and Letters ket price” of labour power to its “natural of Léon Walras. Journal of Political Econo- price.” The labour theory of value as such my 43（ 2）: 187-207. does not guarantee that labour power will Lebowitz, M. A. 2003. Beyond Capital. New sell at its（ labour） value.’ York: Palgrave-Macmillan. Mankiw, N. G. 1998. Principles of Economics. References Fort Worth, TX: Dryden Press. Blaug, M. 1996. Economic Theory in Retrospect. Marx, K.［ 1867］ 1996. Capital Vol. I. In The Cambridge, UK: Cambridge Univ. Press. Collected Works of Karl Marx and Freder- Böhm-Bawerk, E. v.［ 1896］ 1984. Karl Marx ick Engels, Vol. 35. New York: International Okada: Marx versus Walras on Labour Exchange 61

Publishers. ments. Science & Society 67（ 2）: 160-72. -.［ 1898］ 1985. Value, Price and Profit. In Roemer, J. E. 1982. A General Theory of Exploi- The Collected Works of Karl Marx and Fre- tation and Class. Cambridge, MA: Harvard derick Engels Vol. 20. New York: Interna- Univ. Press. tional Publishers: 101-49. Shapiro, C. and J. E. Stiglitz. 1984. Equilibrium -. 1975. Economic and Philosophic Manu- Unemployment as a Worker Discipline De- scripts of 1844. In The Collected Works of vice. American Economic Review 74 （3）: Karl Marx and Frederick Engels, Vol. 3. 433-44. New York: International Publishers: Spencer, D. A. 2009. The Political Economy of 229-346. Work. Abingdon: Routledge. -. 1986. Economic Manuscript of 1857-58. Walras, A［. 1849］ 1997. Théorie de la richesse In The Collected Works of Karl Marx and sociale（ The Theory of Social Wealth）. In Frederick Engels, Vol. 28. New York: Inter- Auguste et Léon Walras, Œuvres économiqu- national Publishers. es complètes, II. Paris: Economica: 107-70. -. 1987. The Original Text of the Second and Walras, L［. 1860］ 2001. L’économie politique et the Beginning of the Third Chapter of A la justice（ Economics and Justice）. In Au- Contribution to the Critique of Political guste et Léon Walras, Œuvres économiques Economy. In The Collected Works of Karl complètes, V. Paris: Economica: 77-313. Marx and Frederick Engels, Vol. 29. New -［. 1877］ 1993. Théorie mathématique de la York: International Publishers: 430-507. richesse sociale（ The Mathematical Theory -. 1988. Economic Manuscript of 1861-63. of Social Wealth）. In Auguste et Léon Wal- In The Collected Works of Karl Marx and ras, Œuvres économiques complètes, XI. Frederick Engels, Vol. 30. New York: Inter- Paris: Economica: 1-309. national Publishers. -［. 1898］ 2005. Studies in Applied Econom- -. 1989. Critique of the Gotha Programme. ics, Vol. II. Abingdon: Routledge. In The Collected Works of Karl Marx and -［. 1926］ 1954. Elements of Pure Econom- Frederick Engels, Vol. 24. New York: Inter- ics. London: George Allen & Unwin. national Publishers: 75-99. -.［ 1936］ 2010. Studies in Social Econom- Misaki, K. 1998. Walras no Keizai Shiso-: Ippan ics. Abingdon: Routledge. Kinko- Riron no Shakai Vision （Walras’s -. 1987 a. La loi fédérale sur le travail dans Economic Thought: The Social Vision of His les fabriques（ The federal Law on Factory General Equilibrium Theory）. Nagoya: Labour）. In Auguste et Léon Walras, Œu- Univ. of Nagoya Press. vres économiques complètes, VII. Paris: Oakley, A. 1979. Aspects of Marx’s Grundrisse Economica: 219-29. as Intellectual Foundations for a Major -. 1987 b. Ruchonnet et le socialisme scienti- Theme of Capital. History of Political Econ- fique（ Ruchonnet and Scientific Socialism）. omy 11（ 2）: 286-302. In Auguste et Léon Walras, Œuvres écono- Okishio, N. 1963. A Mathematical Note on miques complètes, VII. Paris: Economica: Marxian Theorems. Wirtschaftliches Archiv 504-14. 91:287-99. -. 1993. Analyse des éléments d’économie Pagano, U. 1985. Work and Welfare in Econom- politique pure（ The analysis of Elements of ic Theory. Oxford: Basil Blackwell. Pure Economics）. In Auguste et Léon Wal- Park, C.-S. 2003. On Replacing Labor as the ras, Œuvres économiques complètes, VII. Substance of Value: Early and Recent Argu- Paris: Economica: 487-501. 62 経済学史研究 52 巻 2 号

-. 1996. Cours d’économie politique appli- tion of the Laws of Distribution. Hants: Ed- quée（ Courses in Applied Economics）. In ward Elgar. Auguste et Léon Walras, Œuvres économiqu- Yellen, J. L. 1984. Efficiency Wage Models of es complètes, XII. Paris: Economica: 383- Unemployment. American Economic Review 714. 74（ 2）: 200-05. Wicksteed, P. H.［ 1894］ 1992. The Co-ordina-