1. Say's Treatise: bringing Smith to the Continent
When Jean-Baptiste Say published his Traitéd'économie politique (henceforth Treatise or T) in 1803, economic thinking in England was largely shaped by Smith's Wealth of Nations, while in France and some other parts of Continental Europe physiocratic ideas were still predominant. Say, who admittedly had adopted (and adapted) the economics of Adam Smith for the main parts of his Treatise, can be credited with having established certain Smithian ideas on the Continent.1
However, while Say had indeed borrowed much of his economics from Smith, he had not followed the Scotsman in every respect and had not confined himself to merely popularizing the latter's ideas. First, there were other, mostly French authors, especially Condorcet, Condillac and Turgot (see, for example, T: xxxii-xxxvi), who exerted an important influence on Say's thinking. Secondly, beginning with the second edition of the Treatise Say put forward criticisms of Smith's Wealth of Nations. He stressed five 'principal imperfections' of Smith's doctrine: (i) 'To the labour of man alone he ascribes the power of producing value. This is an error' (T: xl); (ii) 'By the exclusive restriction of the term wealth to values fixed and realized in material substances, Dr. Smith has narrowed the boundary of this science' (ibid., pp. xli-xlii; emphasis in the original);2 (iii) 'on the subject of commercial production [he] presents us with only obscure and indistinct notions' (ibid.: xlii); (iv) 'His work does not furnish a satisfactory or well connected account of the manner in which wealth is distributed in society' (ibid.); and (v) 'by not characterizing the two different kinds of consumption, namely, unproductive and reproductive, he does not satisfactorily demonstrate, that the consumption of values saved and accumulated in order to form capital, is as perfect as the consumption of values which are dissipated' (ibid.: xlii-xliii). Say had thus departed from Smith in several respects of his analysis, and in particular with regard to that part which was at the very centre of economic theory: the theory of value and distribution. Say's conceptualization, although undoubtedly inspired by Smith's, was markedly different from it. With its emphasis on 'utility' and on 'supply and demand', the theory of value expounded by Say stressed some elements of Smith's to the neglect of others, with the result that an entirely different theory emerged. As we shall see below, the same can also be said of Say's explanation of interest and profit.
When Ricardo took up the Smithian heritage in England, he could draw not only on the WN but also on the exposition and elaboration of Smithian economics contained in Say's Treatise. While Ricardo praised certain aspects of Say's analysis, from the beginning he was critical of other aspects and especially of Say's theory of value and distribution. Early on he pointed out difficulties, inconsistencies and contradictions in Say's analysis, which, in Ricardo's judgement, the latter was incapable of resolving in the successive editions of the Treatise.
In this paper we deal with the differences of opinion between Ricardo and Say in the theory of value and distribution – the main field in dispute between them. We focus attention on those aspects which, in our interpretation, concern fundamental issues in controversy between the two authors and set aside other aspects which we consider to be only peripheral to the main theme. In particular, we shall refer to Say's concept of the 'entrepreneur' only in passing (see, therefore, Steiner 1996: 30-35 and 1998b), and we shall not deal with the similar, but by no means identical, views of Say and Ricardo with regard to the 'law of markets'.3 Since during the time of their acquaintance until Ricardo's untimely death in 1823 Say variously expressed his wish to learn from his British counterpart and to absorb his doctrine4, our question will essentially be whether he made progress in this regard. With the possible exception of the law of markets there is indeed no presumption that Say exerted any influence on Ricardo, whilst we have Say's word that things were different the other way round. The question then is to what did this influence amount. Accordingly Ricardo's analysis will serve as a measuring rod of Say's analytical achievements in the theory of value and distribution.5 Whenever possible we shall let the two authors speak for themselves and keep our comments and interpretation to a minimum.
The structure of the paper is the following. In Section 2 we provide a brief account of Say's first encounters with Ricardo and of the latter's criticism of his work in 1814 and 1815. Significant differences in terms of scope, method and content of the analyses of the two economists are pointed out in Section 3, while Section 4 identifies the main areas of disagreement between Ricardo and Say. In Section 5 we document Ricardo's reactions to Say's attempts to come to grips with his criticisms in the successive editions of his Treatise. Section 6 turns to Say's concepts of 'value' and 'riches' which Ricardo considered mistaken. Section 7 deals with the different views of the two authors in the theory of value, the problem of the measure of value and Ricardo's criticism of Say's distinction between 'gross' and 'net revenue'. Section 8 addresses their opposing views in the theory of income distribution, focusing attention on the explanation of rents and profits. Section 9 contains some concluding remarks.
2. The first encounters between Say and Ricardo
Ricardo and Say first met in December 1814,6 when Say visited Ricardo at Gatcombe Park and the two together went to see Bentham and James Mill at Ford Abbey. Their meeting had been arranged by Mill, who had informed Ricardo on 24 November 1814: 'Mons. Say, the author of the excellent book with which you are well acquainted, entitled Economie Politique, is in this island. It would be a thousand pities that you and he should not see one another' (Ricardo, Works, vol. VI: 156-7). Say had been commissioned by the French Government to study economic conditions in England and bring back such information as might find useful application in France.7 Ricardo reported on their meeting in a letter to Malthus of 18 December 1814: 'Monsr Say ... does not appear to me to be ready in conversation on the subject on which he has very ably written, – and indeed in his book there are many points which I think are very far from being satisfactorily established, – yet he is an unaffected agreeable man, and I found him an instructive companion' (ibid.: 161).8 In the same letter Ricardo also expressed his agreement with what was later to become one of Say's best-known propositions: 'Mr. Say, in the new edition of his book [Traité, 2nd ed., 1814], ... supports, I think, very a[bly] the doctrine that demand is regulated by production. Dema[nd] is always an exchange of one commodity for another. ... Accumulation necessarily increases production and as necessarily increases consumption' (ibid.: 163-4).
But on the occasion of this meeting Ricardo (and Mill) must also have raised objections to some of Say's 'definitions', because Ricardo was to write to James Mill on 30 August 1815: 'Have you seen Mr. Say's Catéchisme D'Economie Politique? ... I like it very much though he has not altered the definitions to which you and I objected last year' (ibid.: 264-5; see also Ricardo, Works, vol. VI: 269). What Ricardo had criticized in the first two editions of Say's Treatise becomes clear from a letter to Say of 18 August 1815, in which he observed with regard to Say's Catéchisme:
You have I perceive a little modified the definition of the word value as far as it is dependent on utility, but with great diffidence, I observe, that I do not think you have mastered the difficulties which attach to the explanation of that difficult word. Utility is certainly the foundation of value, but the degree of utility can never be the measure by which to estimate value. A commodity difficult of production will always be more valuable than one which is easily produced although all men should agree that the latter is more useful than the former. A commodity must be useful to have value but the difficulty of its production is the true measure of its value. For this reason Iron though more useful is of less value than gold. (Ricardo, Works, vol. VI: 247-8)
In the same letter Ricardo also criticized Say's confounding of value and riches:
Riches are valuable only as they can procure us enjoyments. That man is most rich, and has most valuables, who can procure in exchange for his commodities, not those things which he himself or the world generally consider as most desireable, because they may possibly be procured at little cost, but those things which are of difficult production, which is always the foundation of great value. It appears to me therefore incorrect to say as you do page 95 that that man is superlatively rich, although he has few valuables, who can procure easily or for nothing those things which he wishes to consume. (ibid.: 248)
Finally, Ricardo commented on a passage in which Say had suggested that an increase of capital could be ascertained by valuing all commodities in the capitalists' inventory at their current prices. As against this Ricardo argued that 'An increase of capital is to be ascertained only by its power of employing more industry and of adding to the produce of the land and labour of the country' (ibid.: 248-9).
Ricardo's early objections thus concerned three main issues: the theory of value, the distinction between value and riches, and the problem of the measure of value.
3. On differences in scope, method and content
As the early exchange between Say and Ricardo and its reflection in the correspondence of Ricardo with James Mill, Malthus and McCulloch also indicate, their differences of opinion concerned the scope, method and content of economic analysis. As regards scope, Say opted for a Political Economy that was practically useful and this aim, he thought, necessitated the employment of the inductive method (see also Steiner 1996: 25-8 and 1998b: 202-6). 'The excellence of this method', we read in the Treatise, 'consists in only admitting facts carefully observed, and the consequences rigorously deduced from them' (T: xvii). And: 'Political economy, in the same manner as the exact sciences, is composed of a few fundamental principles, and of a great number of corollaries or conclusions, drawn from these principles. It is essential, therefore, that these principles should be strictly deduced from observation' (ibid.: xxvi). Say was of the opinion that mathematics had no role to play in the study of economic phenomena and that mathematical calculation was 'the most dangerous of all abstractions' (ibid.: xxviii n.). Given his concern with practical matters, the time frame of his analysis was often short term.
Ricardo was also interested in establishing principles that could provide guidance to the statesman. However, contrary to Say and notwithstanding his expertise as a highly successful stock jobber, in the theory of value and distribution Ricardo advocated the deductive method of reasoning. In the introduction to the Treatise Say criticized him for this:
It is, perhaps, a well founded objection to Mr. Ricardo, that he sometimes reasons upon abstract principles to which he gives too great a generalization. When once fixed in an hypothesis which cannot be assailed, from its being founded upon observations not called in question, he pushes his reasonings to their remotest consequences, without comparing their results with those of actual experience. (ibid.: xlvii)
In accordance with his deductive bent in the theory of value and distribution, Ricardo did not share Say's view as regards mathematics. This becomes indirectly clear when Ricardos writes about Malthus: 'Another of his great mistakes is I think this; Political Economy he says is not a strict science like the mathematics' (Ricardo, Works, vol. VIII: 331). At the same time there is evidence indicating that Ricardo distinguished between different spheres of economic analysis and the ability of the theorist to establish economic laws in them. In his letter to Malthus of 9 October 1820 he specified:
Political Economy you think is an enquiry into the nature and causes of wealth – I think it should rather be called an enquiry into the laws which determine the division of the produce of industry amongst the classes who concur in its formation. No law can be laid down respecting quantity, but a tolerably correct one can be laid down respecting proportions. (ibid.: 278)9
As regards the determination of value and distribution in a given time and place, Ricardo implied, a law can be formulated. In such a formulation mathematics was of considerable use, as is discernible especially in the chapter 'On Value': see, in particular, Ricardo's discussion of the inverse relation between the real wage rate and the rate of profits, his so-called 'fundamental theorem of distribution' (see Blaug 1997 : 96); and his treatment of fixed capital which shows that he employed the annuity formula in the numerical examples given (see Ricardo, Works, vol. I: 54-62). Much of Ricardo's analysis is concerned with analysing the properties of long-period positions of an economic system in conditions of free competition, characterised by a uniform rate of profits and uniform rates of remuneration for each kind of primary factor of production.
Say and Ricardo had somewhat different views of society and its stratification which are reflected in their respective theories. There are passages in Say's writings expressing an essentially harmonious view. However, as we shall see below, he was perfectly aware that the relationship between the different classes of society was conflict-ridden, a fact that found expression in recurrent social unrest. Apparently, he was convinced that Political Economy could help to alleviate, and eventually dissolve, such conflicts of interest. In one place he boldly contended that Political Economy 'satisfactorily proves that the interest[s] of the rich and poor, and of different nations, are not opposed to each other, and that all rivalships are mere folly' (T: lix). But his warnings of the selfishness of entrepreneurs and his plea that the state ought to defend the workers against them were more down to earth and reminiscent of Smith's view. In Ricardo, as in Smith before him, social conflicts accounted for a good deal of the thrust of the analysis. The observation that the interests of the different classes of society were not congruent was a common premise of their theoretical edifices and a main explanatory factor of economic and social development. In Ricardo the emphasis was on the conflict of interest between the landed gentry, on the one hand, and the rest of society, on the other, as was reflected, for example, in the controversy about the Corn Laws. He was also concerned, though to a lesser extent, with the conflict between the rising class of industrial, merchant and money capitalists, on the one hand, and workers, on the other.
4. Areas of disagreement between Say and Ricardo
If we now turn to the areas in Political Economy with regard to which Say and Ricardo disagreed in their correspondence and publications, we see that these concerned almost exclusively the following:
(i) the distinction between 'value' and 'riches';
(ii) the theory of value, the problem of the measure of value, and the distinction between net and gross revenue; and
(iii) the theory of income distribution, especially the explanation of rent and of profits.
There were a few related themes, in particular the impact of machinery on employment and that of agricultural improvements on rents. A careful inspection shows that it was only with regard to the law of markets, later known as 'Say's law', that the two authors were largely in agreement, at least at first sight. Both insisted that acts of saving do not involve a reduction in effective demand for consumption goods, but rather a change from 'unproductive' to 'reproductive consumption'. Hence they concluded that there cannot be a 'general glut' of commodities as was maintained by authors such as Malthus and Simonde de Sismondi. A closer look would reveal, however, significant differences of opinion of the two economists (see Gehrke and Kurz 2001: section VIII). The view that there cannot be a lack of aggregate effective demand was a consequence of the fact that both authors, like the classical economists in general, tended to identify saving and investment. More precisely, they lacked a proper analysis of the interplay of saving and investment.
As regards areas (i)-(iii), on which we focus attention, Ricardo's disagreement with Say is all too obvious and not even concealed by his polite and diplomatic ways to refer to this in their correspondence and in print. And, as we have seen, the differences were there right at the beginning of their intellectual encounter and did not go away over time, notwithstanding Say's repeated affirmations that he had absorbed Ricardo's doctrine and had made use of it in the numerous changes between the different editions of the Treatise, his magnum opus. While Ricardo took notice of these changes, he felt with an ever increasing astonishment, which was to turn into disenchantment, that the substance of Say's respective arguments, which he thought erroneous or misleading, was basically left untouched. Therefore, his belief in Say's ability to understand his, Ricardo's, doctrine gradually vanished.
Scrutiny shows that the differences of opinion concerned what were then and are today considered to be core fields of Political Economy. Indeed, appearances to the contrary notwithstanding, Ricardo did not endorse the central 'principles' Say advocated, that is, doctrines which the latter had adopted with some adaptations from Smith and which Ricardo had already criticized in Smith. In particular, Ricardo thought that in the theory of value and distribution Say contradicted himself in important respects and that his overall analysis lacked coherence. As to the distinction between net and gross revenue, Ricardo saw Say committing essentially the same error as Smith, an error which derived from the defective theory of value the two authors advocated. According to that theory the price of a commodity equals the added up payments of factor services in its production. This component parts approach to the theory of value misled Smith and Say to contend that all prices would rise or fall as wages rose or fell. The two authors thus missed the constraint binding changes in the distributive variables, given the technical conditions of production, demonstrated by Ricardo. Say emphasized also the importance of utility for prices and argued that the latter were determined by demand and supply. According to Ricardo these propositions were either wrong or vacuous, because the concepts of 'demand' and 'supply' had little or no analytical content. As regards income distribution, Say mixed Smithian ideas such as the one concerning the conflict over wages between master and workmen with his own demand–and–supply approach yielding a compromise which Ricardo found unsound. He criticized Say especially for not having understood the principle of differential rent and, once rent is set aside by focusing attention on marginal (i.e. no-rent) land, that the rate of profits is determined once the real wage rate is known, and that these two distributive variables are inversely related to one another.
All other differences of opinion follow from those rooted in differences in their respective theories of value and distribution. For example, Say very much in the tradition of the physiocrats and Smith maintained that improvements in agriculture would increase rents, whereas Ricardo argued that this was not necessarily the case: it all depended on the particular form of improvements under consideration. According to Say the employment of improved machinery could never cause a problem of unemployment. Ricardo in the third edition of the Principles showed that there are cases in which the introduction of new machinery was injurious to the labouring classes.
5. Insubstantial modifications: Say's difficulties with Ricardo's theory of value and distribution
Ever since they had first met in 1814 Say felt the need to respond to the objections Ricardo had raised against his doctrines. In the successive revisions of his Treatise he explicitly sought to bring his own exposition into conformity with Ricardo's ideas. This process began in 1815, before Ricardo had published his Principles,10 and it continued until Ricardo's death in 1823.
Ricardo reported on Say's reaction to the publication of his Principles in a letter to Trower of 23 August 1817:
Say ... spoke favourably of my book – was quite sure that in a few years there would not be a shadow of difference between us, but he complained that I had made demands too great on the continued exercise of thought on the part of my reader, and had not sufficiently relieved him or assisted him by a few occasional examples, and illustrations, in support of my theory. (Ricardo, Works, vol. VII: 178)
In December 1817 Say informed Ricardo: 'I am still busy correcting my Traité d'Economie and I am making much use of your Principles of Political Economy' (ibid.: 227).(1*) However, instead of a new edition of the Treatise, Ricardo was first to receive the French translation of his Principles by Constancio, with Say's 'explanatory and critical notes', in December 1818.11 He commented on them in a letter to Mill of 22 December 1818: 'M Say does not appear to me to have clearly seen the doctrine which I wish to establish' (ibid.: 371). James Mill, in a letter of 24 December 1818, was rather more outspoken:
I am full of contempt for these notes of Say. ... There is not one of your doctrines, that he has seized, or perceives the force of in any degree. Think of his saying ... that you have assuredly in the text committed a great error, because in talking of exchangeable value, you have not included profits of stock and rent, as constituent parts. This is to declare, as plainly as words can speak, that the man knows not in the smallest degree what your book is about. (ibid.: 375)12
Ricardo's assessment, though expressed less harshly, was substantially the same. In a letter to Mill of 28 December 1818 he said about Say's notes:
Some of them are ingenious, but he does not grapple with the real question in dispute, – he makes a shew of answering it, but he completely evades it. In his note on gross and net revenue [see Say in Ricardo 1819, vol. I: 222-4] – he begs the question; – he first supposes that a part of the revenue received by the labourers is more than their wants require – that is to say is net revenue, and then he says that there is an advantage in increasing the gross revenue altho' you do not thereby increase the net revenue. In what I said on that subject I expressly guarded myself, by saying, that Adam Smith had not argued this question on a supposition that by increasing the number of labourers you were increasing the number of human beings susceptible of and enjoying happiness ...; and yet M Say answers my observations by saying that there would be a greater number of human beings enjoying happiness. (Ricardo, Works, vol. VII: 378-9)
Ricardo had contemplated the idea of including a translation of Say's notes in the second edition of his Principles, together with his comments on them. But, as he informed Mill, he then decided to drop it and added: 'I think of making no other answer to M Say's observations but that of remarking that he has left my main position respecting the regulator of rent unanswered' (ibid.: 379).
In the meantime Say prepared the fourth edition of his Treatise.13 When in October 1819 he sent a copy to Ricardo, he wrote:
You will see by the corrections I have made, in particular in the first chapters of the second book on the Distribution of Wealth, how useful your criticisms have been for me, since they have obliged me to work again on the most delicate parts of my doctrine. I will consider myself very fortunate if these corrections and several others succeeded in winning you over on the few points where I have had the misfortune not to agree with you. (Ricardo, Works, vol. VIII: 136)(2*)
Ricardo did not consider their differences to be only few and insubstantial. His reaction, in a letter of 11 January 1820, must have disappointed Say:
Your chapter on value is, I think, greatly improved; but I cannot yet subscribe to all your doctrines on that most difficult part of the science of Political Economy. In that Chapter you appear to have misapprehended a position of mine. I do not say that it is the value of labour which regulates the value of commodities, for that is an opinion I do all in my power to overthrow; but I say that it is the comparative quantity of labour necessary to the production of commodities, which regulates their relative value. (ibid.: 149)
Not surprisingly, Ricardo was at a loss to understand how Say could possibly attribute to him the old Smithian view, which Ricardo had long shown to be untenable, that value was regulated by the value of labour, that is, wages, rather than by the quantity of labour 'embodied' in a commodity. To McCulloch Ricardo wrote more frankly, on 23 November 1820:
I have looked over carefully all the new matter in his fourth edition without discovering any thing to induce me to alter the opinion which I have given of the confusion of his ideas respecting value. Utility, riches, value, according to him are all the same thing. A commodity is more valuable because it is more useful. A man is rich in proportion as he is possessed of value – of utility, and it makes no difference whether commodities are of a low value or of a high value. Erroneous as I think these views are he has not the merit of uniformly adhering to them, for he often acknowledges that commodities will fall in value if their cost of production be diminished, altho' they preserve the same utility. The book I think is altogether an able one, but I am quite convinced that M Say does not see quite through the subject. (ibid.: 298-9)
Say continued to modify his exposition, and when he presented Ricardo with his Lettres à M. Malthus in August 1820, he wrote:
I strongly desire that the explanations I here give of my doctrine of value will satisfy you better than those found in my previous writings. This doctrine seems to me to be suitable for being adopted and further elaborated by you, and I await with impatience the first writings you are going to publish in order to know what you think of it; because I believe to have shown it to be nothing else but yours only expressed in different words. (ibid.: 280-81)(3*)
Ricardo was not of this opinion. In a letter of 4 September 1820 he wrote to Malthus: 'I have seen Say's letters to you; ... I am not convinced by any thing Say says of me – he does not understand me, and is frequently at variance with himself when value is the subject he treats of' (ibid.: 227-8).14 Malthus for once agreed: 'There are more contradictions in it than those which relate to value, and there are some doctrines, besides those which directly concern me that appear to me to obscure, rather than to throw light on the general subject. I cannot agree with him in making no distinction between services and products, in his strange and useless application of the term utility, in his opinions respecting the immateriality of revenues, and in his mode of reasoning by exclamations which enable him to stop short when he comes to the stress of the argument' (ibid.: 259-60). Ricardo informed Malthus in a letter of 24 November 1820 that he had written some notes on Say's Lettres (which, however, are not extant), 'with which I am by no means pleased. He is very unjust to me, and evidently does not understand my doctrine; and for the opinions which we hold in common, he does not give such satisfactory reasons as might I think be advanced.15 ... In Say's works, generally, there is a great mixture of profound thinking, and of egregious blundering. What can induce him to persevere in representing utility and value as the same thing?' (ibid.: 301-302). And to McCulloch, who had published a review of Prinsep's translation of Say's Traité in the Scotsman, Ricardo wrote in a letter of 25 April 1821: 'The criticism on Say in the Scotsman is I think very just – he is certainly very far behind in his knowledge of the present state of the science' (ibid.: 374).
When Ricardo presented Say with a copy of the third edition of his Principles on 8 May 1821, he pointed out
the particular difference which exists between us, respecting the meaning which should be attached to the word 'value'. You use it in the same sense as 'riches' and as 'utility' and it is this part of your valuable work which I am very anxious should have the benefit of your further consideration. ... Allow me also to remark that your work would be much more valuable if you entered more fully into the laws which regulate rent and profit. (ibid.: 379-80)
Upon returning from his 'Tour on the Continent' in December 1822 Ricardo reported to Malthus: 'I saw Say several times, but our conversation did not turn much on subjects connected with Political Economy – he never led to those subjects, and I always fancied that he did not much like to talk about them. ... Neither ... [Garnier] nor M. Say have succeeded in at all understanding what my opinions are' (Ricardo, Works, vol. IX: 248-9). Ricardo gave a similar account in a letter to Trower, and added: 'Speaking to the Duke of Broglie of M Say he observed that he did not appear to him to have the least notion of the doctrines of the New School, – that his notes in the French translation of my book shewed clearly that he did not know what the subject in dispute was' (ibid.: 244).
While it is true that Say revised his Treatise several times, the modifications he introduced with regard to the central issues that were in dispute between him and Ricardo were insubstantial. A detailed account of these modifications therefore seems dispensable.16 In the following discussion of these controversial issues we will therefore make use of the English translation of the fourth edition of Say's Traité and occasionally draw on the correspondence between the two economists and their other writings.17