Sraffa on Marshall’s theory of value in the Cambridge lectures: achievements in an unfinished criticism

Abstract

In his Cambridge lectures, Sraffa criticised Marshall’s theory on the basis of the recognition of the incompatibility between classical political economy and marginalist economics. Preparing his lectures, Sraffa had also identified the need to determine prices and distribution simultaneously although some implications of this latter result were probably not yet fully evident to him. Sraffa still accepted the Marshallian thesis that classical political economy and marginalist economics identified two alternative “ultimate standards of value”. This position bears witness to Sraffa’s initial adherence to the Marshallian theoretical framework. The road towards Production of Commodities was open, but still unfinished.

Keywords:

• Sraffa
• Marshall
• classical political economy
• marginalist economics

JEL CODES:

• B12
• B13
• B24
• D46

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 Sraffa delivered his lectures on “Advanced theory of value” at Cambridge University over the three successive academic years from 1928–29 to 1930–31. His papers include a long manuscript with the same title catalogued D2/4, consisting of 243 pages and divided into three parts. The first part of the lectures occupies the first 71 pages. The second part addresses the issues discussed in the 1925 and 1926 articles, and the last part deals with imperfect competition, questions regarding utility and demand, and the general equilibrium formulations of marginalist theories.

2 The reconstruction of the evolution of Sraffa’s thinking which the present paper is based on was originally put forward by Garegnani (2004, 2005). It has been developed in Kurz and Salvadori (2005a, 2005b), and Kurz (2006), works which are essential to any complete understanding of the evolution. They indicate the role of the lectures as attesting to the “turning point”, a role also shown in Signorino (2005), the first paper focusing on the lectures. Kurz (1998) is essential for an overview of the unpublished manuscripts of Sraffa.

3 This valuable methodological guidance appears today as unusual as it is necessary. See Trezzini (2018).

4 Petty ([1672] 1691), 181.

5 The three terms theoretical approaches, conceptions and notions are used here to clarify an understandable ambiguity in Sraffa’s terminology. Sraffa also frequently uses conceptions or views to indicate different theoretical approaches to the theory of value.

6 It is worth noting that at the beginning of his work on reconstructing the evolution of the notion of cost, Sraffa saw the analysis of Smith but also those of Ricardo and Marx as a degeneration of the initial conception of cost as an aggregate of material goods developed by Petty and the Physiocrats in that they introduced the notion of labour and, with it, a philosophical element to the analysis. On the initial position, see Signorino (2005), while on the fact that later Sraffa realized that this change did not entail any fundamental inconsistency with the conception of objective physical cost of classical political economy, see Fratini (2018) and Trezzini (2018).

7 According to Senior and Mill, profits are morally justified by the fact that capital accumulation requires abstinence from consumption, no abstinence being instead involved in the possession of land. Sraffa argues, however, that this asymmetry disappears when accumulated capital is inherited from past generations and continues to earn profits. According to Sraffa, it is abstinence from the direct consumption not only of capital, inherited or otherwise, but also of land that should be considered relevant, as direct consumption by the owner is possible in both cases (D2/4/3: 40–47).

8 Moreover, when introducing the Austrian notion of cost, Sraffa argues that the notion developed by Senior and Mill is more in line with the conception of classical political economy than with that of marginalist economics. Like the classical economists, Senior and Mill regard cost as a primary fact whereby means of production and subsistence transfer value to the product whereas for the Austrians and the opportunity cost authors, means of production and labour derive their value solely from the utility of what they produce. See D2/4/3: 48.

9 Sraffa explicitly refers to Menger (1871), Wieser (1884) and Böhm-Bawerk ([1889] 1930).

10 Sraffa refers to Davenport (1908), Henderson (1920) and Wicksteed (1910).

11 Sraffa does not use the term ‘value’ in this passage. It has been inserted for the purposes of symmetry with the other passage on the Austrians quoted above and the sense of the sentence.

12 Marshall ([1890] 1920), 339.

13 This reasoning is made more complex both in Sraffa’s argument and in Davenport’s original analysis by the fact that, as is known, the rent of land is not regarded in Marshall’s analysis as a component of the prices of commodities. With this assumption, a further difficulty in connecting the actual expenses of production and real cost arises from the fact that rents certainly enter into the first but are precluded from the second notion of cost. Examination of the notion of opportunity cost leads Sraffa to identify a weakness in Marshall’s position on rent and to argue that it is inconsistent with marginalist principles.

14 Henderson (1920), 164–165.

15 Davenport (1908), 383.

16 Wicksteed (1910).

17 On this point, see also fn. 31 on Böhm-Bawerk's position.

18 In addition to the preparatory manuscripts for the Lectures, which we will refer to later, manuscript D1/22, dated pre-1928, is entitled “Standard and Cause of value”.

19 Beyond the extraordinary confusion that Sraffa talks about, an understanding of the subject of the debate on the ultimate standard of value is made clearer, in the eyes of today's reader, by the fact that in Production of Commodities, the term standard commodity is used by Sraffa to indicate a notion that is much more related to the measure than to the cause or origin of value. The origin or the ultimate cause of the value was instead the centre of the debate at the time.

20 Ricardo (1951–1973), vol. I, 10.

21 A similar interpretation of the articles was put forward by Mongiovi (1996), who also discusses an apparently similar position to Samuelson's (1987) on Sraffa’s articles.

22 In the next section, we briefly deal with some different reconstructions.

23 “Suppose now that we found that the price of one commodity is determined exclusively by expenses of production: could we conclude that it is “ultimately” determined only by real costs? Of course not. [….] We could only say that the price of those commodities is only determined by “the conditions of supply”, i.e., by the wages, interests, rents that have to be paid for its production, and that demand price has nothing to do with its determination. But this would not in the least exclude that we might hold a theory of distribution which said that wages, interest, etc. are only determined by the utility of their products (as a whole) or by monopoly, or any other general theory of value. The difference would be that utility would make its influence felt through the conditions of supply of that (or each of all) commodity, and not through its conditions of demand” (D3/12/3. 60–61).

24 The manuscripts in which the two issues are dealt with alternately are: D3/12/11 (November 1927), D3/12/5 (Winter 1927–28), D3/12/10 (December 1927), D3/12/8 (Lent term 1928), D3/12/7 (After 1927), D3/12/9 (May-July 1928) where the indications of the period are written by Sraffa himself. Another manuscript of the same period, D3/12/4 (November 1927), is only about the evolution of the notion of cost while in D3/12/6 (Winter 1927–28) and D3/12/10 (Michaelmas term 1928) we find again equations.

25 In these equations, the quantities produced are taken as data; one distribution variable (the subsistence wage) was given and the other determined simultaneously with prices.

26 In those manuscripts, Sraffa works on the hypothesis of constancy of the relationship between the value of net product and that of capital, as the distribution of income changes, a hypothesis that Bellofiore himself defines as “anti-Sraffian” and that was subsequently abandoned with torment – “disaster of the model” – by Sraffa. This attempt would, however, be indicative of what were the real objectives that Sraffa would then pursue with the elaboration of the standard system. Bellofiore (2014) also argues that in the manuscript entitled “Use of the notion of Surplus” of November 1940, Sraffa studies how Marx identified the origin of surplus-value by studying a situation without surplus and then moving on to a situation in which the surplus emerges as a consequence of the lengthening of the working day. This would support this line of arguing. Carter (2014) discusses, on the other hand, the so-called Majorca Draft of 1955.

27 See Petri (2015) on the New Interpretation of Marx and Levrero (2017) on the contributions in Bellofiore and Carter (2014).

28 In the lectures, Sraffa began to understand much more clearly than his contemporaries that labour, in the classics, was a measure of physical cost and not the origin of value.

29 As already mentioned, the necessary simultaneity of the determination of prices and distribution takes different forms in the two approaches. Prices are determined simultaneously with all the distributive variables in marginalist theories but simultaneously with one distributive variable (the rate of profit or the surplus wage rate) in surplus approach theories while the other (the real wage or profit rate) is taken as given. Rent is determined separately.

30 Böhm-Bawerk (1894) explicitly criticises Marshall and the “English and American colleagues” who supported Marshall’s metaphor of the scissors and counterposes the Austrian approach. He mainly criticises the equality between disutility of work (or abstinence) and the remuneration of factors and thus the conception of cost with negative utility. He discusses numerous differences between reality and the hypotheses needed to ensure that marginal disutility of work is actually equal to wages. However, there are two crucial points in his argument: first, the condition of equality between prices and costs does not determine prices but is only a condition imposed by competition and allows uniformity of prices of the same goods and that of the remunerations of the same factors. Second, cost according to the Austrians – as Sraffa claims in his lectures – is not an autonomous primary fact, but a reflection of the remuneration of productive factors. His position on the question of the ultimate cause of value is based on a general economic equilibrium reasoning in which a single ultimate standard does not exist. The section that Böhm-Bawerk puts in italics is significant:

There is a certain limited quantity of productive power which at any given time, under the conditions set by the technical development of that time, can bring forth only a certain limited quantity of products. These products, through the action of certain leveling influences in the different branches of production, are disposed of in a regular order of succession, in each case, to the best paying purchaser. The satisfaction extends downward in the scale of wants until a certain equalization to the (money) marginal cost of production is attained, and it is this which decides the value of all goods that come under the dominion of that leveling influence. It determines the value of the products as well as the value of the productive power, which is represented by the cost (p. 50).

The determining causes of the value are, then, the circumstances that are assumed as data in the theory. In the above passage, we find quantities of available factors, consumer tastes and techniques of production. In the last section, Böhm-Bawerk tries to conclude on the ultimate standard by reconciling his position with the terms of the debate:

What then is the “ultimate standard” [….] If we wish to answer this question in a single phrase, then we cannot choose any less general expression than “human well-being.” The ultimate standard for the value of all goods is the degree of well-being which is dependent upon goods in general. If, however, we desire a more concrete standard, one that will give us a more definite idea, just how goods are connected with well-being, then we must take not one but two standards, which though co-ordinate in theory are yet of very unequal practical importance, because of the greater prevalence of the phenomena in which one of them is operative; one is the utility of the good, and the other is the personal sacrifice or disutility involved in the acquisition of the good. The domain of the latter is much more limited than we usually think. In the great majority of cases, even in those in which the so-called law of cost undoubtedly plays a part, the final determination of the value of goods is dependent upon utility (pp. 59–60).

With the expression human well-being, Böhm-Bawerk seems to synthetize the quantities of productive powers available that according to the given techniques can satisfy given needs (tastes) whereas the second part of the passage seems a slightly contradictory concession granted to the terms of the debate.

31 On the contrary, he considered his discovery of the diversity of the two approaches solid and verifiable in the texts. This result was therefore made explicit in the lectures and crucial to his criticism of Marshall.

32 Pages 85–87 of the manuscript D3/12/7 are meaningful because in them Sraffa questions which hypotheses about "variable returns" are to be included in his equations. Besides imagining, as described in the text, different possible roles for the equations and for the Marshallian scheme of supply and demand, he does not understand that effectively no hypothesis regarding returns is necessary when the produced quantities are assumed as given; he does, however, advance the idea of suggesting to a potential reader to assume for simplicity the hypothesis of constant returns, something that he will actually do in Production of commodities.

33 “Man from the moon. The significance of the equations is simply this: that if a man fell from the moon on the earth, and noted the amount of things consumed in each factory and the amount produced by each factory during a year he could deduce at which values the commodities must be sold; if the rate of interest must be uniform and the process of production repeated. In short, the equations show that the conditions of exchange are entirely determined by the conditions of production” (D3/12/7: 87).

34 In classical political economy, there is no analysis of the dependence of cost on the level of production of a commodity. This lack of analysis is not due, however, as interpreted by Marshall and initially by Sraffa, to a hypothesis of constancy of costs, but, as it began to appear to Sraffa in 1928 and is clearly stated in his mature contributions, to the hypothesis of given produced quantities in the determination of prices. Sraffa identified the data of the classical theory as: (i) the size and composition of the social product; (ii) the technical conditions of production; and (iii) the real wage (or the profit rate) (Eatwell 1977; Garegnani 1984).

35 As already mentioned, Sraffa delivered his lectures over three consecutive academic years and added notes on or entirely new pages each year. This assignment cannot be precisely dated. This is unfortunate since this indication of Sraffa’s awareness of the relationship between value and distribution, although limited to the marginalist approach, could be the result of late development that did not, however, prompt him to make any radical change to the lectures explicitly addressing the question. In the case of other issues, such as Marshall’s theory of rent, Sraffa did, on the other hand, revise the lectures between the first academic year and the second so as to include the advances he had probably made in the meantime.

36 As Marcuzzo (2005) recalls, in a letter announcing the appointment (JMK to PS, 31 May 1927; KP: L/S/37 – 9) Keynes suggested a course to Sraffa in three parts, the first on the theory of value, the second on the theory of distribution, and the third on an applied subject. Sraffa (PS to JMK, 5 June 1927; KP: L/S/40-1) replied that he wanted to limit the course to the theory of value with an initial section on the theory of supply and a second on demand-side and exchange questions. He excluded the distribution theory because, as he specified, of his ‘fragmentary, confused ideas on the subject’.

37 “For Marshall, wages, interest and profits, are simply shares in the product; they are coordinate quantities that can be regarded as acting upon the value of the product in the same way. Both are the inducement required to call forth certain sacrifices, which are equally necessary for production, and they are also the reward of those sacrifices. […] Petty and all the classics, on the contrary, take the opposite view. They don’t regard at all wages as an inducement; they regard them as a necessary means of enabling the worker to perform his work” (D2/4/3: 22–3).

38 The note in the margin refers once again to the search for an ultimate standard of value independent of distribution, which – as argued above – is incompatible with any complete understanding of the general connection between value and distribution and the need for simultaneous determination that this entails.