George Grote’s manuscript essay on “Foreign trade”

Abstract

This paper explains the origin, content and importance of a manuscript essay on “Foreign trade” which George Grote, the historian of Greece, submitted to David Ricardo in March 1819. By relating it to the contributions to foreign trade theory of Ricardo and of James and John Stuart Mill it is shown that Grote anticipated important elements of the “Sraffa-Yukizawa-Parrinello interpretation” of Ricardo’s cloth-and-wine example. By suggesting that comparative advantage-based trade is incompatible with the labour theory of value Grote also pointed to an apparent incongruity in Ricardo’s chapter “On Foreign Trade.” A transcription of Grote’s manuscript essay is provided in the appendix.

Keywords:

• Comparative advantage
• George Grote
• David Ricardo
• James Mill
• John Stuart Mill

JEL classification:

• B12
• B17
• F10

1. Introduction

Among the Ricardo Papers is an unsigned manuscript essay on “Foreign trade” in the hand of George Grote, the historian of Greece.¹ From the diary notes published by his wife we know that Grote worked on this topic in the period from January to March 1819.² On 27 March he noted in his diary: ‘Finished my remarks on Foreign trade, and enclosed them to Ricardo’, followed on 28 March by the entry:

Rose at ½ past 5. Studied Kant until ½ past 8, when I set off to breakfast with Mr. Ricardo. Met Mr. Mill there, and enjoyed some most interesting and instructive discourse with them, indoors and out (walking in Kensington Gardens), until ½ past 3. (Grote 1873, 36–37)

It is very likely, therefore, that David Ricardo and James Mill discussed Grote’s ‘remarks on Foreign trade’ with him. The manuscript is not annotated, however, and of their ‘instructive discourse’ we have no record.

The interest in Grote’s essay derives mainly from the fact that Ricardo’s approach to foreign trade theory was later modified and transformed, first by James Mill in the subsequent editions of his Elements of Political Economy,³ and then more substantially by John Stuart Mill in “Of the Laws of Interchange between Nations” (written in 1829/30 but published only in 1844 in his Essays on Some Unsettled Questions), where Mill introduced the determination of international values by means of reciprocal demand. Important elements of the transformation of Ricardo’s formulation of the comparative cost theory into the so-called “Ricardian trade model” were however present already in the first edition (1821) of his father’s Elements, and because of his close relationship with Ricardo ‘James Mill’s new approach was easily accepted as a faithful interpretation of the doctrine, to the point that subsequent readings of Ricardo’s Chapter 7 were systematically biased and made through Mill’s lens’ (Faccarello 2022, 78). Such a bias is still absent from Grote’s manuscript essay of 1819.

An earlier plan of publishing Grote’s essay was announced by Thweatt (1976, 232n), but not carried out. This plan also included Grote’s earlier manuscript note on “Foreign Trade” (mentioned in footnote 1 and reproduced in Appendix B), from which he appears to have set out when on 23 March 1819 he began to ‘re-arrange’ his ideas before sending them to Ricardo.⁴ Thweatt contended that ‘both essays on trade are similar to Mill’s 1821 presentation in the Elements’ (ibid.). He also referred to George Grote as ‘a prominent member of the “Millian group”’, along with William Ellis and John Stuart Mill, and maintained that ‘the modern theory of international trade is indebted to the “Millian group” in a way which up to now has not been recognised and appreciated’ (1986, 232 and 234). Grote was thus enlisted as a contributor to the line of development which eventually led, via James and John Stuart Mill, Alfred Marshall, and Gottfried Haberler, to the formulation of the so-called “Ricardian trade model” as an integral part of modern neoclassical trade theory.⁵ A close inspection of Grote’s two manuscript essays shows, however, that this view cannot be sustained: Grote rather retained important elements of Ricardo’s original approach, on which he sought to elaborate. In particular, his reading of Ricardo’s chapter agrees with (what I propose to call) the “Sraffa-Yukizawa-Parrinello interpretation”⁶ of the famous cloth-and-wine example by means of which Ricardo showed that a mutually advantageous trade was possible between England and Portugal even though less labour was needed in Portugal for producing both commodities: According to this interpretation, Ricardo’s famous four numbers refer to the amounts of labour needed in the production of the (unspecified) quantities of cloth and wine actually traded. The numerical example therefore refers to an already existing trade situation with an already established exchange ratio, observed ex post.⁷

A further reason for publishing Grote’s essay is that it draws attention to a major distinction between comparative versus absolute advantage-based trade, which perhaps was not sufficiently emphasised by Ricardo. As Grote makes clear, comparative advantage-based exchange is incompatible with the labour embodied rule of relative exchange value unless there exist persistent wage differentials or barriers to the free movement of capital and labour. In Grote’s essay this distinction occupies a (too) prominent place: It led him to maintain, no doubt erroneously, that in all trade based on absolute advantages, irrespective of whether domestic or foreign, relative exchange values are governed by the labour embodied rule, and to assert that trade based on comparative advantages ‘would not naturally exist’, and ‘never could have been permitted to exist, had individual interest exclusively operated on the human mind’ (Appendix A, p. 24). However, with this latter statement Grote pointed to an apparent incongruity in Ricardo’s chapter “On foreign trade”, which seems to have been little noticed. It concerns the compatibility of Ricardo’s claim that relative values in domestic exchanges are governed by the labour embodied rule with the voluntary exchange in the “shoemaker-hatter example” to which he referred in a footnote to chapter 7 of the Principles ([1817] 1951, 136 n). As will be shown below, in order to make Ricardo’s example compatible with his exchange rule, he must be supposed to have implicitly presumed the existence of a wage differential in a well-specified range.

The comments on Grote’s manuscript essay provided in the following sections are intended to explain its origin, content and importance. Grote’s essay shows that Ricardo’s original approach to the explication of the comparative cost theory, prior to its reformulation by James and John Stuart Mill, could still be properly understood by a contemporary reader. This is not to say that Grote’s argument is perfectly consistent with Ricardo’s or that his essay does not contain misunderstandings, inconsistencies and/or erroneous statements – in fact there are many, and attention will be drawn below to the more important ones.

I start, in Section 2, with providing some background information on the origins of Grote’s manuscript essay on “Foreign trade.” In Section 3, the importance of Grote’s essay in the development of the comparative cost theory is then assessed by relating it to Ricardo’s chapter “On foreign trade”, and in particular to the famous cloth-and-wine example. Grote’s insistence on the incompatibility of comparative advantage-based trade with the labour embodied rule provides an inducement for studying more closely the problem of the compatibility of Ricardo’s statement on the different rules governing foreign and domestic trade with his footnote reference to the “shoemaker-hatter example” in chapter 7 of the Principles. Finally, in Section 5 some notes are offered on further issues raised by Grote in his two manuscript essays, such as the effect of taxation on foreign trade and trade-induced changes in the international distribution of the precious metals, followed by some concluding remarks.

2. The origin of Grote’s manuscript essay

George Grote (1794–1871) was a partner in the banking house of Grote and Prescott. He became acquainted with Ricardo in 1817 and, chiefly due to conversations with him, ‘acquired a lively interest’ in political economy.⁸ This led him in 1818 to study carefully Smith’s Wealth, Ricardo’s Principles, Say’s Traité and a number of other works, and then to write up some manuscript notes on selected problems of political economy.⁹ It was in this connection that he also began to set out his ideas on foreign trade, drawing inspiration primarily from his reading of the chapter “On foreign trade” in Ricardo’s Principles.¹⁰ The short manuscript note headed “Foreign Trade” in his papers on political economy (below, Appendix B) can likely be identified as an earlier version of the manuscript essay that he submitted to Ricardo in March 1819.

Already in the early 1820s, however, Grote’s intellectual interests then quickly shifted to other fields, and although he was a founding member of the Political Economy Club, from which he resigned only in 1831, further papers on political economy from him are not known to exist.¹¹ Grote belonged to the Ricardo-Bentham-Mill circle and was strongly influenced in his philosophical and political views by James Mill, who later also urged him to enter parliament. In the early 1820s he was a frequent visitor, together with his wife, Harriet, at Ricardo’s house in London.¹² Grote also belonged to the group of “Younger Utilitarians” around John Stuart Mill, which met in his house in Threadneedle Street, but he did not participate in the study group’s discussions on political economy in 1825, from which J. S. Mill’s novel ‘theory of International Values’ (as well as his ‘modified form of Ricardo’s theory of Profits’) emanated.¹³

A number of books which can still be identified as having once belonged to George Grote are held by the Senate House Library in London,¹⁴ but these would not seem to offer any substantive additional information. Grote’s copy of Ricardo’s Principles has only some minor marginal notes and underlining; the extant copy of James Mill’s Elements from Grote’s library is lightly annotated, but not in the relevant sections on foreign trade;¹⁵ and his copy of John Stuart Mill’s Essays on Some Unsettled Questions is not annotated at all.

3. Grote’s manuscript essay on “Foreign trade”

For a proper assessment of Grote’s essay in the Ricardo Papers it seems useful to begin with some observations on his earlier manuscript note on “Foreign Trade” in the British Library (below, Appendix B), from which in all likelihood he started out when re-arranging his ‘remarks upon foreign trade’ after his meeting with Ricardo and Mill on 23 March 1819 (see footnote 4 above). Its first part is notable chiefly because of its initial error in maintaining that ‘the trade might still continue, though only one of the countries should derive benefit from it’ (Appendix B, p. 22), and then of its subsequent recognition, in the final paragraph, of the impossibility of mutually advantageous trade when the cost advantages possessed by one country in both commodities obtain

in an equal degree; as when Portugal has an advantage over England of 50 percent in producing both wine and cloth; In this case I do not think there can be any trade between the two countries. (Appendix B, p. 23)

Grote clearly recognised that mutually advantageous trade can only exist when either each country ‘has an advantage peculiar to itself in the manufacture of a particular commodity’, or when the cost advantages possessed by one country are ‘not in an equal degree; as when Portugal has an advantage of 50 perCent over England in producing wine, but only of 30 perCent in producing cloth’ (Appendix B, p. 23). Grote thus spelt out the comparative advantage principle in the “comparison of ratios form”, which was not present in Ricardo’s exposition of the cloth-and-wine example,¹⁶ and which had been first stated by James Mill in his Encyclopaedia Britannica article “Colony” (1818, 26).¹⁷ However, there is no indication that Grote’s argument was derived from a study of Mill’s article; it seems more likely that he had picked up the clue from Ricardo’s footnote in chapter 7 (1951–1973, I: 136n). Be that as it may, it should in any case be noted that this is a remarkably clear and concise statement of the principle of comparative advantage, which could lead us to suppose that Grote had accurately grasped the meaning and scope of application of the comparative advantage principle. Unfortunately, however, some of his statements in the manuscript essay that he sent to Ricardo cast into doubt his proper understanding of this principle, and of Ricardo’s explication of it.

What initially prompted Grote to study the problem of international values and the associated distribution of the trade gains was his attempt to ascertain the ‘Effect of Taxation on Foreign Trade’ (Appendix B, p. 23), that is, his attempt to determine how an already existing trade situation was affected by a commodity tax. Grote noted explicitly that in his numerical examples the amounts of labour needed were meant to refer to the ‘quantities of cloth and wine which were exchanged one against the other’ (Appendix B, p. 23). Like Ricardo, he thus adopted an ex post perspective, contemplating a given trade situation, with given (but unspecified) quantities of commodities exchanged at given prices, so that the gains from trade (prior to the imposition of the tax) are also known. Grote omitted to set out explicitly the calculation of each country’s trade gains in terms of the amounts of labour saved by importing instead of producing domestically, but his reasoning clearly shows that he was aware of it. With $L_j^i,$ $\left(i=E,P;j=c,w\right),$ giving the amounts of labour needed for producing the quantities $Q_c$ of English cloth and $Q_w$ of Portuguese wine exchanged between England (E) and Portugal (P), England’s trade gain amounts to $L_w^E-L_c^E$ and Portugal’s to $L_c^P-L_w^P.$ A further noteworthy feature of Grote’s discussion of the effect of taxation is that the quantities demanded of the two commodities are treated as given and constant (after the imposition of the tax), and that the new trade situation is analysed from the point of view of the (English and Portuguese) merchants, similar to the approach adopted by Ricardo in his analysis of the effects of improvements in chapter 7 of the Principles.¹⁸

Some, but not all of those features were retained in the manuscript essay that Grote sent to Ricardo, which he now started off with the bold statement:

As it is acknowledged that the laws which regulate that species of exchange which takes place between two countries are not universally the same as those which regulate other species of exchange, it forms a proper subject of inquiry, to ascertain the grounds of this difference, to determine whether it is found in all cases, and to measure the extent to which it may ever be carried. (Appendix A, p. 18)

The main purpose of Grote’s essay thus was to explain when and why international exchange ratios are not governed by relative labour values (which, following Ricardo, he supposed to govern domestic exchanges) and to determine the ‘extent’ to which international values could possibly deviate from the latter.¹⁹ Unlike John Stuart Mill ([1829–30] 1844, 2), Grote did not state explicitly that Ricardo had left international values and the associated distribution of the gains from trade indeterminate. However, his attempt ‘to ascertain the proportion according to which the labour of two countries exchanges one against the other’ was, of course, an attempt to determine that distribution, merely phrased differently. Grote thus clearly departed from Ricardo’s approach to foreign trade theory, in which the exchange ratios of the traded commodities were not sought to be determined (nor left indeterminate), but were simply taken as given. While Grote started out from a correct reading of Ricardo’s four numbers, and indeed followed him in this regard by also setting out his own examples in terms of the total labour requirements for producing the quantities of the two commodities actually exchanged, he nevertheless saw a need for ascertaining the rule that governs relative exchange values in international trade. And according to Grote, international trade is indeed governed by two different rules, depending on whether such trade is based on the presence of absolute or of comparative advantages.

Because in Grote’s argument the (in)validity of the labour embodied rule of relative exchange values plays an important role, it needs to be clarified that this rule is strictly valid only under very restrictive technological assumptions, as Ricardo had shown in Chapter 1 of the Principles. Having clearly recognised that price ratios will normally deviate from the corresponding ratios of embodied labour, Ricardo nevertheless assumed in the remainder of his treatise, including Chapter 7, that domestic exchange values are governed by relative labour values (1951–73, I, 36–37). Ricardo’s simplifying assumption can be depicted most conveniently by supposing commodities to be produced by “unassisted labour” alone, with wages advanced at the beginning of the production period. With w denoting the wage rate, r the general rate of profits, and $l_i$ the amount of labour needed per unit of commodity $i (i=1,2),$ domestic relative prices are then given by (1) $$\frac{p_1}{p_2}=\frac{w{l}_1(1+r)}{w{l}_2(1+r)}=\frac{l_1}{l_2}$$(1)

A second, separate issue is the question of the applicability of the labour theory of value with regard to international exchange ratios. In this case, the simplifying assumption adopted above is not sufficient, because of the possibility of different nominal wage rates (in a common world currency) and rates of profit in the two countries, so that we have (with the unstarred variables referring to the home country and the starred ones to the foreign country): (2) $$\frac{p_1}{p_2^{\mathrm{*}}}=\frac{w{l}_1(1+r)}{w^{\mathrm{*}}{l}_2^{\mathrm{*}}(1+r^{\mathrm{*}})}\ne \frac{l_1}{{\mathrm{l}}_2^{\mathrm{*}}}$$(2)

The international exchange ratio, although according to Ricardo (1951–73, I, 375) it is determined by the “natural prices” prevailing in the exporting country, is thus not governed by relative labour values even when the “technological” causes that lead to deviations of production prices from labour values are absent (and unless, of course, the nominal wage rates and the rates of profit are the same in both countries).

In a trade of barter, such as the exchange of English cloth for Portuguese wine, the value of the imports must be equal to the value of the exports, $p_c{Q}_c^E=p_w{Q}_w^P.$ Therefore, the price of cloth in terms of wine must be equal to the ratio of the quantities of wine and cloth actually traded, that is, $\frac{p_c}{p_w}=\frac{Q_w^P}{Q_c^E}.$ Because of $Q_j^i\equiv \frac{L_j^i}{l_j^i},$ this exchange ratio must correspond to $\frac{p_c}{p_w}=\frac{L_w^P}{L_c^E}·\frac{l_c^E}{l_w^P}.$ Accordingly, the international exchange ratio equals the amounts of embodied labour in a unit of cloth and of wine if, and only if, $\frac{L_w^P}{L_c^E}=1.$

After these preliminary observations we can now turn to the discussion of Grote’s numerical examples, which led him to conclude, wrongly, that with trade based on absolute advantages the labour embodied rule always applies. Grote started his deliberations by contemplating the special case of each country enjoying a symmetrically equal absolute advantage in one of the two commodities (Appendix A, p. 18), and by contending that in this case equal amounts of embodied labour are exchanged. The (very) special case considered in Grote’s example is deceptive and it led him to confuse cause and effect. Grote in his example starts out, like Ricardo, from a given trade situation: A certain quantity $Q_w$ of Portuguese wine is exchanged for a certain quantity $Q_c$ of English cloth, and what is assumed to be known are the total amounts of labour needed for producing those (unspecified) quantities in each of the two countries.²⁰ Therefore, the relative price of cloth in terms of wine is also given, independently of the quantities of labour embodied: It simply amounts to the reciprocal of the amounts traded, $Q_w/Q_c.$ If the total amount of labour embodied in a quantity $Q_w$ of Portuguese wine happens to be the same as the total amount of labour embodied in a quantity $Q_c$ of English cloth (as in Grote’s very special numerical example), then $L_w^P/L_c^E=1$ by assumption, and the relative price $Q_w/Q_c$ then also equals the ratio of the amounts of labour embodied in one unit of each commodity. However, the latter ratio cannot be said to determine the relative price.

In the next step, Grote suggested that the labour embodied rule would remain intact also when the absolute advantages enjoyed by each of the two countries become unequal from an improved production method becoming available in one of the countries for its import commodity, provided the direction of trade remains the same (Appendix A, p. 18). Although in this case symmetrically equal absolute advantages cease to prevail, still $L_w^P/L_c^E=1$ continues to hold, of course, by assumption. Grote’s argument, according to which the labour embodied rule of relative exchange values always obtains in foreign trade based on absolute advantages is clearly wrong. However, one of the advantages of Grote’s erroneous argument is that it puts into sharp relief his correct insight that foreign trade based on absolute advantages is not necessarily incompatible with the labour embodied rule of relative exchange value.

Grote then pointed out, quite rightly, that relative values cannot possibly be proportional to embodied labour quantities when ‘one of the countries, though possessing an advantage over the other in both the articles in which they trade, yet has a more decided advantage in the production of one than in that of the other’ (ibid.). Since the numbers in his examples refer, similar to those of Ricardo, to the amounts of labour that are needed in each country for the production of the unspecified quantities of the two commodities traded, Grote also expressed the range for mutually advantageous trade by referring, not to each country’s autarky price ratios or unit labour requirements, but to the equivalent calculation in terms of the total amounts of labour needed in each country for producing the quantities exchanged (Appendix A, p. 19):²¹ (3) $$\frac{800}{600}=\frac{L_c^E}{L_c^P}<\frac{L_c^E}{L_w^P}=\frac{800}{500}<\frac{L_w^E}{L_w^P}=\frac{1000}{500}$$(3)

Clearly, with $L_c^E>L_c^P\mathrm{ }$ and $L_w^E>L_w^P,$ the double condition $\frac{L_c^E}{L_c^P}<\frac{L_c^E}{L_w^P}<\frac{L_w^E}{L_w^P}$ implies that $\mathrm{ }{L}_w^P/L_c^E\ne 1,$ so that cloth and wine cannot possibly be exchanged at labour values. This implication was not spelled out explicitly by Grote, but he seems to have been perfectly aware of it. It should also be noted here (although Grote may not have recognised this) that this double condition for mutually advantageous trade had been obtained by Ricardo without making any assumptions about returns. Ricardo’s (and thus also Grote’s) classical formulation of the comparative advantage principle, unlike that of James and John Stuart Mill, is therefore compatible with non-constant returns and can accommodate incomplete specialisation due to diminishing returns as well as (firm-external) increasing returns.²²

Although Grote adhered to Ricardo’s specification of the trade situation in terms of the total amounts of labour needed for the (unspecified) quantities traded, $L_j^i,$ he nevertheless tried to determine the international exchange ratio in the post-improvement situation and he suggested, quite arbitrarily, that in this case ‘English labour will be exchanged against Portuguese at a disadvantage of 800 to 560’ (Appendix A, p. 19). At the same time, however, he continued to adopt an ex post perspective also in his discussion of the effects of improvements: The new, post-improvement trade situation is evaluated by calculating the gains or losses associated with a ‘continuence’ or ‘cessation’ of the existing trade at the given terms of trade (Appendix A, p. 18). Finally, it should be noted that Grote also stressed, like Ricardo, that the interest of a country in trading with another one depends solely on whether the commodity under consideration could be purchased with less labour abroad than is needed in producing it at home (Appendix A, p. 19).²³

Grote believed, no doubt erroneously, that from his numerical examples the following proposition could be inferred:

Where foreign trade therefore is carried on under the former circumstances [i.e., each country possesses an advantage peculiar to itself], it is subject to the same laws as the home trade; where it is carried on under the latter circumstances [i.e., the advantage in the production of each commodity is on the side of one of the countries], it is regulated by different laws.

And the reason of this distinction is, that in the first case, the trade is one which would naturally exist and be carried on in precisely the same manner, if the two countries were districts in one province; whereas in the latter case, no trade would ever take place between the two countries if they formed merely parts of the same island or province. (Appendix A, p. 19; emphasis added)

Supposing there was a ‘discourse’ among Ricardo, Mill, and Grote on these issues during their walk in Kensington Gardens on 28 March 1819, what would have been Ricardo’s reaction to Grote’s argument? It seems safe to assume that Ricardo would have firmly rejected Grote’s distinction between the two cases of absolute and comparative advantages with regard to the validity/invalidity of the labour embodied rule in foreign trade, because he had stated unambiguously and indiscriminately:

The same rule which regulates the relative value of commodities in one country, does not regulate the relative value of the commodities exchanged between two or more countries. (1951–73, I, 133)

The existence of different laws of value in home and foreign trade Ricardo had explained by the domestic mobility, and the international immobility, of ‘capital and population’.²⁴ Grote followed him in this regard, but pointed out that an important difference exists between the two “species” of absolute advantage- and comparative advantage-based trade:

If Cornwall have an advantage in the production of wine, and Devonshire in that of cloth, the natural effect of the division of labour will lead to an exclusive production of wine by the former, and of cloth by the latter. But this is the precise condition in which the trade between England and Portugal was placed, by the first hypothesis. Since therefore this trade is in its natural state – in that state into which it would really be cast were both countries incorporated into one – there is no disturbing cause, in this species of foreign trade, which can occasion any deviation from those established laws which regulate exchange in the home trade. (Appendix A, p. 19)

With this statement Ricardo would certainly have disagreed, because the international immobility of capital and labour constitutes a ‘disturbing cause’ that must affect also absolute advantage-based trade, since it leads to country-specific (nominal) wage rates and rates of profits. Grote continued:

But the case is widely different, where the disadvantage in both sorts of production is on the side of the same country. Here the distinction of government and laws introduces an unnatural state of things, which never could have been permitted to exist, had individual interest operated exclusively on the human mind, without the control of any other principles. If Cornwall has an advantage over Devonshire in the production of both the wine and of the cloth, there will be no trade at all between the two. Individual interest will direct the removal of the capital of Devonshire into Cornwall. (Appendix A, p. 19)

On this point there is no disagreement between Ricardo and Grote, although Ricardo would probably not have concurred with calling comparative advantage-based foreign trade ‘an unnatural state of things’. However, with his insistence on the distinction between the two types of trade, as ‘the natural effect of the division of labour’, on the one hand, and as ‘an unnatural state of things’, on the other, Grote pointed to an apparent incongruity in Ricardo’s argument in his chapter “On foreign trade”, which seems to have been little recognised.

4. Comparative advantage-based trade and the labour embodied rule: the “shoemaker-hatter” example

From Grote’s pronouncements the reader is induced to ask the following question: Can there be trade based on comparative advantage within a single country, or is such trade possible only between two or more countries? Ricardo’s statements on this question are ambiguous, if not outright inconsistent. On the one hand, he maintained (referring to his famous cloth-and-wine example):

Thus England would give the produce of the labour of 100 men, for the produce of the labour of 80. Such an exchange could not take place between the individuals of the same country. The labour of 100 Englishmen cannot be given for that of 80 Englishmen, but the produce of the labour of 100 Englishmen may be given for the produce of the labour of 80 Portuguese, 60 Russians, or 120 East Indians. (1951–73, I, 135–136; emphasis added)

On the other hand, however, Ricardo attached a footnote to the end of this paragraph, in which he suggested that similar to the case of two individuals it is better also for two countries not to attempt to produce everything domestically, but to specialise in certain productions and to exchange. Adapting an example from the WN, in which Smith explains the advantages of the division of labour with reference to a tailor and a shoemaker ([1776] 1976, 456–457), Ricardo reports the case of two men making shoes and hats, one of them having a productivity advantage in the making of both goods, but a more decided one in the production of shoes (1951–73, I, 136n). In this case, Ricardo contends, it is in the interest of the first man to specialise in shoes and to buy hats from the second one, and conversely for the other man:

Two men can both make shoes and hats, and one is superior to the other in both employments; but in making hats, he can only exceed his competitor by one-fifth or 20 per cent., and in making shoes he can excel him by one-third or 33 per cent.;$-$ will it not be for the interest of both, that the superior man should employ himself exclusively in making shoes, and the inferior man in making hats? (1951–73, I, 136n)

Ricardo’s reasoning seems compelling,²⁵ but is it also compatible with his insistence on the different rules regulating relative values in domestic and foreign trade? Contrary to Ricardo’s assertion, the superior man in his example can have no interest in exchanging his day’s work of shoes for the daily quantity of hats made by the less productive man, because in one day he could produce a greater number of hats himself. The “shoemaker-hatter example” shows indeed that it is not the immobility of capital and labour alone which prevents exchange ratios from being in proportion to the amounts of embodied labour. In order to reconcile Ricardo’s footnote with his insistence on the different rules governing home and foreign trade it is necessary to take into account also the existence of labour heterogeneity.²⁶

Let us take a closer look at Ricardo’s “shoemaker-hatter example” to illustrate the point. In order to simplify the exposition, suppose that the two commodities are produced by (direct) labour alone, and that wages have to be advanced at the beginning of the (uniform) production period. Denoting with $l_j^i$ ($i=S,H;j=s,h)$ the amounts of labour needed by the shoemaker (S) and the hatter (H), respectively, in the production of a unit of shoes (s) and hats (h), and taking into account Ricardo’s specific numerical assumptions about the unequal productivity advantages of the shoemaker with regard to the two commodities, we have: $l_s^S=0,75l_s^H,$ and $l_h^S=0,8\overline{33} l_h^H.$ Clearly, with homogeneous labour the two individuals would have no incentive to exchange their products. Because under free competition a uniform wage rate (w) and a uniform rate of profits (r) must obtain, the shoemaker can produce both commodities at lower costs ($c_j^i$): (4) $$c_s^S=w{0,75l}_s^H\left(1+r\right)<c_s^H=w{l}_s^H\left(1+r\right)$$(4) (5) $$c_h^S=w{0,8\overline{33}l}_h^H\left(1+r\right)<c_h^H=w{l}_h^H\left(1+r\right)$$(5)

Accordingly, ‘there will be no trade at all between the two’, similar to the case of regional productivity differences between Devonshire and Cornwall, where in competitive conditions the production of both commodities will ‘naturally’ be concentrated in Cornwall alone. But it is possible, of course, to follow Ricardo (as well as Smith and other classical economists),²⁷ and to postulate the existence of persistent wage differentials, which have already become thoroughly established, and to use those differentials for reducing the shoemaker’s and the hatter’s differently skilled labours to equal amounts of common standard labour.²⁸ By setting $w^H=1,$ for instance, a unit of “shoemaker-labour” can be reduced to a corresponding amount of standard “hatter-labour” by using as a weight the shoemaker’s money wage rate $w^S.$ Then, for trade to exist between the two individuals, the following inequalities must hold: (6) $$c_s^S=w^S{0,75l}_s^H\left(1+r\right)<c_s^H=w^H{l}_s^H\left(1+r\right)=l_s^H\left(1+r\right)$$(6) (7) $$c_h^S=w^S{0,8\overline{33}l}_h^H\left(1+r\right)>c_h^H=w^H{l}_h^H\left(1+r\right)=l_h^H\left(1+r\right).$$(7)

This implies: (8) $$\frac{l_h^H}{l_h^S}=\frac{l_h^H}{{0,8\overline{33}l}_h^H}=1,2<w^S<1,\overline{33}=\frac{l_s^H}{0,75l_s^H}=\frac{l_s^H}{l_s^S}$$(8) Accordingly, with a wage differential between 20 and 33 percent both men would have an incentive to specialise and to exchange based on their comparative advantages.²⁹ In this case the quantities of the two commodities exchanged, $Q_H$ and $Q_S,$ would contain equal amounts of embodied standard-labour, because the exchange ratio is determined, given a $w^S$ in the appropriate range, by $\frac{p_s}{p_h}=\frac{{w^{S}l}_s^S}{l_h^H},$ so that (9) $${L_h^H\equiv l}_h^H{Q}_h=w^S{l}_s^S{Q}_s\equiv L_s^{\mathit{\text{HS}}}$$(9)

where $L_s^{\mathit{\text{HS}}}$ is the amount of standard “hatter-labour” contained in the quantity of shoes that are exchanged for hats. Ricardo’s footnote reference to the possibility of mutually advantageous trade between the two individuals therefore is not inconsistent with his insistence on the different rules which regulate ‘the relative value of commodities in one country, … [and] the relative value of the commodities exchanged between two or more countries’ (1951–73, I, 133). And contrary to Grote’s assertion, comparative advantage-based trade can ‘naturally’ occur also in a single country, provided wage differentials already exist and are in the appropriate range. Ricardo implicitly appears to have taken the existence of such wage differentials for granted in his argument, without entering into a discussion of the problem of their formation and persistent establishment.

5. Notes on further issues and concluding remarks

Grote noted an important difference between trade based on absolute and on comparative advantage: While the former is compatible with the labour embodied rule governing exchange ratios, the latter is not. From this correct observation Grote wrongly inferred, by basing himself additionally on the special case of symmetrically equal absolute advantages, that the embodied labour rule generally obtains in all foreign trade based on absolute advantages. The ramifications of this erroneous inference permeate his entire essay, because he carried over his distinction between absolute and comparative advantages as a cause for different exchange rules also to his analyses of the effects of taxation, of improvements, and of the value of money in the countries trading together. However, some original ideas are nevertheless to be found also in the remainder of his essay.

In his discussion of the effects of taxation on foreign trade Grote rightly noted that it can be analysed along the same lines as those of improvements, ‘for these two operate in opposite directions’ (Appendix A, p. 21). In conducting the analysis from the point of view of a profit-maximising merchant Grote followed Ricardo, but his statement that the exporter of cloth ‘can only reimburse himself by selling the wine which he imports at a higher price’ (Appendix A, p. 20) conflicts with Ricardo’s observation that ‘every transaction in commerce is an independent transaction’ (1951–73, I, 138).³⁰ A comparison of Grote’s and Ricardo’s ideas on the effects of taxation on foreign trade is made difficult by the fact that the latter are not easy to pin down. Ricardo clearly considered taxes to be distortionary, because they alter relative prices and thereby change the allocation of resources.³¹ In chapter IX, “Taxes on raw produce”, for instance, Ricardo argued that taxation could give ‘rather a different direction … to foreign trade’ (1951–73, I, 171–172). Because the proportions in which raw produce enters into the production of commodities vary, the imposition of a tax on raw produce will alter relative prices. It is for this reason also that he maintained in “Funding System” that ‘an injudicious tax may induce us to import what we should otherwise have produced at home, or to export what we should otherwise have received from abroad’ (1951–73, IV, 189). Moreover, Ricardo twice changed his mind on the question whether a general tax on commodities could raise prices without an increase in the quantity of money. In 1811, in his Reply to Bosanquet (1951–73, III, 242–243), he held that this was possible, but in 1817, in edition 1 of the Principles (1951–73, I, 213), he denied the possibility. In edition 3 (1821), however, he reverted to his earlier view and inserted two footnotes to this effect (1951–73, I, 213 and 196).³² It seems unlikely, however, that Grote’s later insertion in his discussion of tax-induced price increases (Appendix A, p. 20) was related to Ricardo’s initial change of mind on this issue.

Grote’s analysis of the effect of a commodity tax on foreign trade can also be compared with that of James Mill, who addressed this problem in the first edition of his Elements (1821, 231–235). Mill’s treatment differed markedly from both Grote’s and Ricardo’s. As Faccarello has pointed out, also in the discussion of this problem Mill ‘reasons in terms of units of commodities, and the “four magic numbers” are defined, in each country, as the per-unit domestic prices of these commodities and no longer as the quantities of labour respectively necessary to produce, in each country, the amounts exchanged internationally’ (2022, 70).

Grote closely followed Ricardo’s analysis also with regard to the determination of the value of money or gold.³³ Ricardo had sketched two main factors that influence the value of gold: the amount of labour needed in the production of gold in the country in which it is mined (plus the labour needed for its transportation), and the distribution of the precious metals among the countries trading together, which is bound to change whenever gold rather than other commodities are used in exchange for imports. By considering a trade situation between England, France and Mexico in cloth, wine and gold, Grote provided an original and interesting account of the combined effect of the different factors influencing the value of money in the three countries. However, it must be acknowledged that Grote’s discussion of foreign trade in cloth and gold exchanged between England and Mexico also suffers from his erroneous claim that commodities will be exchanged in proportion to quantities of embodied labour when the two countries each possess a superiority in the article they export. This error is carried over also to his analysis of the trade in cloth, wine, and gold between England, France and Mexico.

There is, nevertheless, some sound reasoning and much food for thought to be found in Grote’s two essays on foreign trade. Some of his statements in the first essay (Appendix B, pp. 22–23) show a remarkably clear understanding of the comparative advantage principle and its scope of application; his reading of Ricardo’s four numbers in the cloth-and-wine example agrees with the “Sraffa-Yukizawa-Parrinello interpretation”; he recognised that Ricardo’s approach to the analysis of the effects of improvements could be extended to those of taxation; and he combined in an original way elements of Ricardo’s approach to the determination of the value of money and the distribution of the precious metals. Finally, it seems noteworthy that Grote also recognised that with unrestricted international capital mobility absolute cost advantages prevail over comparative ones:³⁴

If Cornwall has an advantage over Devonshire in the production of both the wine and of the cloth, there will be no trade at all between the two. Individual interest will direct the removal of the capital of Devonshire into Cornwall. It would indeed direct the same removal in the case of England and Portugal, but here its suggestions are stifled and overpowered by more powerful motives. (Appendix A, p. 19)

However, Grote failed to spell out the implication of establishing a regime of unrestricted international capital mobility, namely that a country may then be unable to compete internationally in any commodity, and thus to participate in the international division of labour.³⁵

Acknowledgements

I would like to thank two anonymous referees of this journal for their most helpful comments and suggestions and the Syndics of Cambridge University Library for granting me permission to publish a transcription of the Grote manuscript on “Foreign trade” (Add. Mss 7510 IV.C.1). I also wish to thank Richard van den Berg for providing me with a copy of Grote’s manuscript note from the British Library and with some information on Grote’s books in the Senate House Library, London. An earlier version of this paper was presented at the 26^th Annual Conference of the European Society for the History of Economic Thought in Liège, Belgium, 1–3 June 2023, and at the September Conference on Economic Theory and Policy at Meiji University, Tokyo, 14–15 September 2023. I am grateful to the session participants, and in particular to my discussants, Taichi Tabuchi and Jou Ishi, for very helpful comments and suggestions. Comments on an earlier version of this paper were received also from Tony Aspromourgos, Saverio M. Fratini, Harald Hagemann, Heinz D. Kurz, Pedro S. Machado, Patrick Mellacher, Sergio Parrinello, Taichi Tabuchi, and Alex M. Thomas, and are gratefully acknowledged. All remaining errors are, of course, my responsibility.

Disclosure statement

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

Notes

1 The manuscript is noted in the introductory notes to Ricardo (1951–73, VI, xxxiii–xxxiv). It consists of three double quarto sheets and was transferred with the Ricardo Papers to the Cambridge University Library (Add. Mss 7510 IV.C.1). It exhibits the same handwriting and general appearance as Grote’s other papers on political economy in the British Library (Add. Mss 29,530), which comprise another, and probably earlier, manuscript note with the same title. A transcript of the latter is provided in Appendix B, while the manuscript essay from the Ricardo Papers is reproduced in Appendix A.

2 See Grote (1873, 32–37).

3 Faccarello (2022) provides a detailed account of the subtle changes James Mill introduced in the three editions of his Elements of Political Economy (1821, 1824, 1826) to Ricardo’s exposition. On John Stuart Mill’s contribution to the transformation of Ricardo’s trade theory into the so-called “Ricardian trade model” see, for instance, Faccarello (2015, 2017, section 1.1), Tabuchi (2017) and Gehrke (2017).

4 Grote’s diary entry for 23 March 1819 reads: “I went up to Brook Street to breakfast with Mr. Ricardo; was very politely received by him; walked with him and Mr. Mill in St. James’s Park until near 12, …; wrote out again in the evening some of my remarks upon Foreign trade, and arranged them in a different manner.” (Grote 1873, 36) Grote therefore may have discussed his ideas on foreign trade with Ricardo and Mill not only on 28 March, but also earlier.

5 Thweatt’s attempt to attribute priority for the first correct formulation of the comparative advantage principle to James Mill and the “Millian group”, rather than to Ricardo, was fully endorsed by Rothbard, who similarly referred to Grote’s manuscript essay in the Ricardo Papers (apparently without having seen it) as an “important, unpublished essay setting forth the Millian view on comparative advantage” ([1995] 2006, 98).

6 This reading of Ricardo’s four numbers was first suggested by Sraffa (1930). Subsequently, the same reading was then also proposed, independently of Sraffa (1930) and of each other, by Yukizawa (1974, 1978) and Parrinello (1973, 1988). More recently, Ruffin (2002) also suggested this interpretation, and Maneschi (2004) then introduced the notion of “Sraffa-Ruffin interpretation”. However, in view of the earlier contributions by Yukizawa and Parrinello, which were apparently overlooked by him, it seems more appropriate to re-name it as the “Sraffa-Yukizawa-Parrinello interpretation”. For a summary account of Yukizawa’s contributions, of which no English translation exists, see Tabuchi (2017). For a concise exposition of this interpretation and its implications, see Faccarello (2023). For an assessment of Ruffin’s contribution and a critique of his reconstruction of Ricardo’s discovery of comparative advantage, see Gehrke (2015).

7 Incidentally, it may be noted that Ricardo’s approach to the explication of the comparative advantage theory, unlike John Stuart Mill’s, eschews counterfactual reasoning.

8 See Grote (1873, 12 and 28). Grote and James Mill first met in 1818 or 1819, “the introduction being effected through Ricardo” (Bain [1882] 1967, 180). By March 1819 their acquaintance must still have been fairly recent, because Grote only related his first impressions of Mill to his future wife, Harriet, in a letter of May 1819 (ibid., 181).

9 These manuscript notes, consisting altogether of some 230 pages, are kept in the British Library under the title: “G. Grote, Tracts on Political Economy, 1818–1823”, with the reference number “Additional manuscripts 29,530”. On the spine of the volume it is noted: “Presented by Mrs Grote”. The pages on foreign trade, reproduced in Appendix B, are on sheets 117 to 120 (only recto pages are numbered).

10 Apart from Ricardo’s Principles, Grote in his diary also refers to his reading of Smith’s Wealth, Say’s Economie politique, Turgot’s “Valeur et monnaies”, and Melon’s Essai politique sur le commerce; see Grote (1873, 29–32).

11 In 1820, Grote completed an essay on “Magic”, which was a thinly-disguised attack on religion inspired by Jeremy Bentham’s and James Mill’s philosophical radicalism. In the following year, he was asked by James Mill to work on an extensive set of manuscripts from Bentham, out of which he composed An Analysis of the Influence of Natural Religion on the Temporal Happiness of Mankind, published in November 1822 (see Fuller 2008). In 1821, he also published a short pamphlet on parliamentary reform (Grote 1821).

12 Harriet Grote also conveyed how deeply affected her husband was by Ricardo’s untimely death: “I never saw George so oppressed by any event before.” (Bain [1882] 1967, 211)

13 John Stuart Mill’s account of the study group’s discussions on political economy, which were based on a reading of his father’s Elements, Ricardo’s Principles and Bailey’s Critical Dissertation, closes with the remark: “When we had enough of political economy, we took up the syllogistic logic in the same manner, Grote now joining us.” ([1873] 1981, 125; emphasis added).

14 Approximately one-third of the books that originally belonged to Grote’s library are still held by the Senate House Library; the rest was sold or has been dispersed on the library’s book-shelfs and cannot be identified anymore. (Information provided by Mrs Lucy Evans, Special Collections Manager, Senate House Library, London.).

15 The copy of Mill’s Elements was presented and inscribed by James Mill to Grote’s wife, Harriet, but it ended up in her husband’s library.

16 See Aldrich (2004, 289).

17 See Thweatt (1976, 231).

18 See Ricardo (1951–73, I, 137–139). See also Faccarello (2015 and 2017, section 2.1) and Kurz (2015, section 8; 2017).

19 If “the sole purpose of the four numbers [in Ricardo’s numerical example] was to illustrate the proposition that the relative value of commodities produced in different countries is not determined by the respective quantities of labour devoted to the production of each,” as has recently been maintained by Morales Meoqui (2023, 3), then Grote’s essay could be seen as an attempt to contribute to a better understanding and explanation of this fact.

20 As Bhering and Serrano (2022) have rightly pointed out, these are not arbitrary quantities, but the “reciprocal net effectual demands” (which correspond to the total effectual demands net of the domestic production on superior quality lands); in this case for wine in England and cloth in Portugal.

21 The amounts of labour needed in country i $(i=E,P)$ for the (unspecified) quantities actually traded of commodity $j,Q_j \left(j=c,w\right),$ are given by $L_j^i=l_j^i{Q}_j.$ Since the quantities exchanged are the same for both countries, the comparison of the total amounts of labour is of course equivalent to that of the (unknown) unit labour coefficients, $\frac{L_c^P}{L_c^E}>\frac{L_w^P}{L_w^E}\iff \frac{l_c^P}{l_c^E}>\frac{l_w^P}{l_w^E}$ (see also Machado and Trigg 2021, 380n).

22 See Parrinello (1988, 587). For classical trade models with non-constant returns in the Ricardian tradition see, e.g., Parrinello (2022) and Bhering and Serrano (2022).

23 As Ricardo stressed in one of his notes on Malthus’s Principles, the comparison of “real costs” (i.e. labour costs) between countries is irrelevant for the particular interest of a country in trading with another one: “It can be of no consequence to America, whether the commodities she obtains in return for her own, cost Europeans much, or little labour; all she is interested in, is that they shall cost her less labour by purchasing them than by manufacturing them herself.” (1951–73, II, 383).

24 See Ricardo (1951–73, I, 135–137).

25 The main reason why Ricardo’s reasoning seems so compelling is that the overall quantities of shoes and hats are maximized when the two individuals specialise according to their comparative advantages.

26 The superior man can be naturally more talented, better skilled, or equipped with a better technology. For simplicity he will be designated in the following simply as “better skilled”.

27 See Ricardo (1951–73, I, 20–22). See also Kurz and Salvadori (1995, 322–327).

28 From an analytical point of view, assuming labour heterogeneity in combination with wage differentials is similar to assuming within-country immobility of capital and labour: To suppose that the shoemaker receives a persistent wage premium because of his better skills is similar to supposing that workers in Cornwall are more productive and receive higher wages than in Devonshire (or that the Cornish technology is superior to the one in Devonshire) and that capital, labour, and technologies are immobile between the two regions.

29 More generally, the wage differential must be higher than the shoemaker’s lowest productivity advantage, but fall short of his highest productivity advantage. Tabuchi (2023, 3–5) has provided a numerical example to show that the “shoemaker-hatter” example can be interpreted on the basis of Ricardo’s ex post approach. I am grateful to Taichi Tabuchi for making a draft version of this paper available to me.

30 See Faccarello (2023, 38).

31 Ricardo therefore advocated import duties and bounties on exportation as counter measures to rectify the tax-induced distortions of relative prices.

32 See Sraffa’s note on the manuscript “A note on prices and taxation” (1951–73, IV, 320).

33 James Mill, in “Colony” (1818), had omitted to provide a monetary analysis, but had simply referred the reader to Ricardo’s discussion. On the value of gold and silver currencies, among nations trading together, see also Ricardo (1951–73, VIII, 2).

34 Ricardo had already alluded to this possibility when he observed: “It would undoubtedly be advantageous to the capitalists of England, …, that under such circumstances, the wine and the cloth should both be made in Portugal, and therefore that the capital and labour of England employed in making cloth, should be removed to Portugal for that purpose.” (1951–73, I, 136)

35 For classical trade models with international capital mobility, which show that a country may be unable to compete internationally with regard to any commodity, see Parrinello (2010) and Bellino and Fratini (2022).

Appendix A:

Transcript of the manuscript “Foreign Trade” in the Ricardo Papers (Add. MSS 7510 IV.C.1, Cambridge University Library) in the hand of George Grote

Foreign trade

As it is acknowledged that the laws which regulate that species of exchange which takes place between two countries are not universally the same as those which regulate other species of exchange, it forms a proper subject of inquiry, to ascertain the grounds of this difference, to determine whether it is found in all cases, and to measure the extent to which it may ever be carried.

The motive which induces two countries to trade together is the same as that which influences two individuals – viz: that mutual commerce enables each country to enjoy more of both articles of trade, than it could have done if there had existed no trade at all. This object may be attained under two different circumstances – either when each country possesses a peculiar advantage in the production of its own article, over the other – or when one of the two countries, though possessing an advantage over the other in the production of both the articles in which they trade, yet has a more decided advantage in the production of one than in that of the other. In the former of these two cases, the laws which regulate foreign trade will be precisely the same as those which regulate the home trade, and equal quantities of labour will be exchanged against each other exactly in the same manner as if both articles were produced in the same country.

But in the latter of the two cases, the laws which regulate foreign trade will differ from those which regulate the home trade, and equal quantities of labour will not be exchanged against each other.

Suppose England producing cloth by the labour of 500 men, to trade with Portugal producing wine by the labour of the same number while the same quantity of cloth requires in Portugal 1000 men to produce it, and the same quantity of wine requires in England 1000. Each country here enjoys an equal advantage in the production of its own article. In this, which is the simplest case imaginable in all foreign trade, it seems clear that the laws which regulate prices will be exactly the same as in the home trade, and that English and Portuguese labour will be exchanged upon an equal footing. For nothing can enable England to obtain more wine for the cloth which she sends to Portugal, which might not as well be supposed to enable Portugal to obtain more cloth for the wine which she sends to England.

Let us now suppose that the advantage which each country possesses in the production of its own peculiar article is not equal; that for instance, the wine and the cloth required in Portugal, respectively, 500 and 600 labourers – and in England, respectively, 1000 and 500. Here each country enjoys an advantage in the production of the article which it sends to the other, but the advantage which England has over Portugal in the production of the cloth is only as 500 to 600, while that which Portugal has over England in the production of the wine is as 500 to 1000. But though England gains more by the continuance of the trade, and would lose more by its cessation, than Portugal, yet the labour of the two countries will continue to be exchanged on an equal footing. For if it be held true, that they will be exchanged on an equal footing in the first case supposed, where the advantages are equal, there will not be the smallest reason for a disturbance of this principle, when, by an improvement introduced into either of the countries, the advantages become unequal, provided there still continues, on the side of each country, a superiority in one of the articles produced. Thus, if an improvement were discovered in Portugal by which the cost of producing cloth were reduced from 1000 to 600 labourers, this would not at all discompose the equal ratio according to which Portuguese and English labour were before bartered. For under those circumstances Portugal would pay only the produce of 500 of her labourers for that which cost as many in England, and therefore it would not make any difference to the former whether she could produce the cloth for 600 or for 1000 labourers herself, for she purchases it of England for 500.

But the trade between the two countries may exist where one of them has a decided advantage in the production of both the articles, though not so great in the one as in the other – where, for instance, Portugal can produce the wine by 500, and the cloth by 600 labourers, while England cannot produce the wine under 1000 and the cloth under 800.

And in this case the labour of the two countries will not be exchanged at an equal ratio. England cannot obtain from Portugal the produce of 800 Portuguese labourers for the cloth which took 800 English labourers to make it – for by the supposition, Portugal can herself produce it for 600. So that in order to make it the interest of the latter country to continue the trade, England must offer to her a quantity of English labour equal to 800, and receive in exchange a quantity of Portuguese labour equal only to 560 or 570. English labour will be exchanged against Portuguese at a disadvantage of 800 to 560.

It appears therefore that in order to ascertain the proportion according to which the labour of the two countries exchanges one against the other, we should enquire whether each country possesses an advantage peculiar to itself in the production of one of the articles of trade, or whether the advantage in both of the articles belongs to one of the two. If each country possesses a peculiar advantage, then will the labour of the two countries be exchanged on an equal footing, although the superiority possessed by one in the production of her commodity should be far greater in degree than that possessed by the other in the production of hers. But if the advantage in the production of each commodity should be on the side of one of the countries, then the labour of the two countries will be exchanged, though on an unequal footing, yet in the ratio least unfavourable to that country on whose side the disadvantages are.

It appeared for instance in the case just quoted, that English labour was exchanged against Portuguese not in the ratio of its disadvantage in wine, where it was greatest, but in that of its disadvantage in cloth, where it was the least.

Where foreign trade therefore is carried on under the former circumstances, it is subject to the same laws as the home trade; where it is carried on under the latter circumstances, it is regulated by different laws. And the reason of this distinction is, that in the first case, the trade is one which would naturally exist and be carried on in precisely the same manner, if the two countries were districts in one province; whereas in the latter case, no trade would ever take place between the two countries if they formed merely parts of the same province.

If Cornwall have an advantage in the production of wine, and Devonshire in that of cloth, the natural effect of the division of labour will lead to an exclusive production of wine by the former, and of cloth by the latter. But this is the precise condition in which the trade between England and Portugal was placed, by the first hypothesis. Since therefore this trade is in its natural state – in that state into which it would really be cast were both countries incorporated into one – there is no disturbing cause, in this species of foreign trade, which can occasion any deviation from those established laws which regulate exchange in the home trade.

But the case is widely different, where the disadvantage in both sorts of production is on the side of the same country. Here the distinction of government and laws introduces an unnatural state of things, which never could have been permitted to exist, had individual interest operated exclusively on the human mind, without the control of any other principles. If Cornwall has an advantage over Devonshire in the production of both the wine and of the cloth, there will be no trade at all between the two. Individual interest will direct the removal of the capital of Devonshire into Cornwall. It would indeed direct the same removal in the case of England and Portugal, but here its suggestions are stifled and overpowered by more powerful motives. The existence therefore of this disturbing cause readily points out to us the reason, why foreign trade, when carried on under these circumstances, is governed by laws of its own – while in the other case it remains under the influence of those principles which regulate exchanges in general.

In considering the effects of Taxation upon Foreign Trade, we must distinguish between the two sorts of Foreign Trade before noticed. If England has an advantage in the production of cloth, and Portugal in that of wine, a tax laid upon cloth in England will raise the price of that article both in England and in Portugal. A tax laid on wine in Portugal will produce the like effects on the price of wine in the two countries. But if England is under a disadvantage, in comparison with Portugal, in the production of the cloth as well as in that of the wine, then any tax imposed upon cloth in England will merely increase the disadvantageous ratio according to which English labour is already exchanged against Portuguese. It will raise the price both of cloth and of wine in England.

For the English exporter of cloth, when the cost of producing it has been increased by the tax, cannot obtain any greater quantity of wine for it in Portugal, since it will be the interest of Portugal rather to produce cloth for herself than to pay the increased price. As therefore the cost of the cloth which he exports has been increased by the tax, and as he cannot obtain any greater quantity of wine in return for it, he can only reimburse himself by selling the wine which he imports at a higher price. If England produces the cloth by 800 labourers, and Portugal can produce it by 600, a tax which should raise the cost of production of cloth in England from 800 to 850, will only oblige England to give the labour of 850 Englishmen, instead of 800, in exchange for that of 560 Portuguese. ([Later insertion:] On re-perusing this, I think I was wrong to say, that the price of wine and cloth will both rise in England – for money would be exported – but England would obtain less wine from Portugal in exchange for the same quantity of cloth.)

To apply these general principles of foreign trade to the particular case of a trade with Mexico, it seems that if any country trades with Mexico upon equal terms, having an advantage in cloth or some other article of demand to counterbalance that which the latter state possesses in silver, the value of silver will in that country be regulated by the quantity of labour required to produce it, precisely in the same manner as if Mexico were one of its own provinces. Distance of transport is of course to be reckoned as a part of the labour required, and is an element in the cost of commodities which is found in the home as well as in the foreign trade. Suppose however England and France, each trading only with Mexico, and each precisely at the same distance from it, but the former possessing an advantage over Mexico in cloth, the article which she exports, while the latter is at a disadvantage with respect to Mexico as well in wine, the article which she exports, as in silver, the article which she imports. Under these circumstances, English and Mexican labour will be exchanged on an equal footing, and silver will in England command precisely the quantity of labour which it has cost to produce and to bring it there. But French and Mexican labour will be exchanged at a ratio disadvantageous to the former, and a given quantity of silver will in France command more than the same quantity of labour which it has cost to bring it thither. In other words, silver will be higher in France than in England, though both at the same distance from Mexico.

This state of things would continue in all cases unaltered, if France and England traded each with Mexico alone, and had no commerce with each other. But if they trade immediately with each other, the relative value of silver will be determined by the nature of the commerce carried on between them. If that commerce be one in which each of the countries possesses an advantage in the article which it exports, then the relative value of silver will continue in the two countries as before. But if it be one in which England possesses an advantage over France in the article which she imports as well as in that which she exports, then another cause will be added to that which already existed, why silver should be relatively higher in France than in England. But if the trade between France and England be one in which France had a superiority in both articles of trade, then the disadvantageous ratio, at which French labour [9] is exchanged against Mexican labour, will be counterbalanced by the advantageous ratio, according to which it will be exchanged against English labour. Silver will become of the same relative value, or may be even relatively lower, in France than in England.

The exportation of money from a country is a sign that it can no longer carry on a trade upon an equal footing – that its labour must be exchanged at a disadvantage against that of the country with which it trades. This point is not clearly elucidated in the particular case supposed in the Chapter on Foreign Trade. On the contrary, the importation of money is a sign that a country is trading, or preparing to trade, at an advantage, and that its labour will be exchanged at a premium.

All improvements in manufacture, and all taxation (for these two operate in opposite directions) do not tend necessarily to make the country in which they occur trade at an advantage or at a disadvantage, nor do they always occasion an altered distribution of the precious metals. It is only when they are of such a kind and extent as to change a trade, in which each country possesses a superiority in one of the articles exchanged, into one where the superiority in both of the articles will belong to one of the two. Thus if England trades with Portugal, having an advantage in the production of cloth and a disadvantage in the production of wine, no improvement which England might discover in that department where she already possessed a superiority, would occasion any exportation or importation of money. It would merely lower the price of cloth both in England and in Portugal. But if England should discover an improved mode of making wine, whereby she might obtain a superiority over Portugal in that article as well as in the cloth, then immediately the nature of the trade is changed, and this alteration in its nature is indicated by an exportation of money from Portugal.

It is not however wholly impossible that an alteration such as that just mentioned in the nature of the trade might take place without any exportation of money. Suppose, for instance, that England sends cloth to Mexico and brings home silver, having a superiority over Mexico in the production of cloth. Now if Mexico should under these circumstances discover an improved mode of making cloth, whereby she should obtain a superiority in that article over England, the trade might still continue, but English labour would immediately be exchanged at a disadvantage. English cloth would command less silver in Mexico, and silver would in England become of greater exchangeable value.

When it is stated, that, if England and Portugal trade together in cloth and wine, each possessing a superiority in the article which she exports, under such circumstances English and Portuguese labour would be exchanged upon an equal footing – it is supposed, that the consumption of wine in England and of cloth in Portugal is such, that the one is exactly equal to balance the other. For if the demand for cloth in Portugal should very greatly diminish, and there be no other article in demand there in which England has an advantage, then the equal ratio of exchange between the labour of the two countries will be destroyed. When England has exported all the cloth which Portugal demands, and has no other commodity, in which she possesses a superiority, to offer in exchange for the additional wine which she wants, this additional quantity must be purchased by the transmission of some commodity, in which she has a disadvantage in comparison with Portugal. At first however she will export money, in which she has no disadvantage; until money is so lowered in Portugal, and so raised in England, that the labour of 800 English is equal to no more money than the labour of 600 Portuguese. The disadvantageous article may then be exported. By this exportation of money however, and by the consequent change in the ratio of English to Portuguese labour, the superiority which England formerly possesses in the cloth will be entirely lost. For English labour in all articles, and Portuguese labour in all articles must be of equal value, and consequently as the former is to the latter in one commodity, so it is in all.

Hence it appears, that articles in foreign trade will be exchanged according to the laws which regulate value in general, and in proportion to the quantity of labour which they contain, when the two countries engaged in the traffic possess each a superiority in the article which it exports. But that when the superiority in the production both of the imported and exported article is on the side of the same country, its labour will be exchanged at an advantageous ratio against that of the other country.

The grounds of the infringement, in the second case, of the general laws which regulate exchanges, are that a departure there takes place from that mode of conduct which individual interest enjoins; since it is merely upon the supposition, that this mode of conduct is in all respects pursued, that our rule for ascertaining the value of articles in general is constructed.

The extent of the disadvantages, to which the labour of one of the countries trading is subject, measures the possible deviation which may take place from the general law of exchangeable value.

Appendix B:

Transcript of the manuscript note “Foreign Trade” in “G. Grote, Tracts on Political Economy, 1818–1823” (Add. MSS 29,530, British Library)

Foreign trade

As Nature has bestowed different and peculiar excellencies of soil and climate on different countries, so the same portion of labour and capital will in some obtain a much greater portion of various commodities than in others. Hence there results a division of labour among countries, analogous to that among individuals, each devoting itself exclusively to the production of those articles in which it possesses some natural or acquired advantages, distributing its surplus produce among the rest. By means of this division of labour and subsequent exchange, the whole produce of the world is far greater than it would be, were every country to supply itself with every thing from its own soil – not to mention that this is in many instances wholly impracticable.

Hence therefore the origin of Foreign Trade; the effect of which, when perfectly free, is to distribute the aggregate mass of capital in the most beneficial manner, so as to afford the greatest possible quantity of annual production. Foreign Trade however though it increases greatly the mass of enjoyments, yet does not, any more than any other exchange, increase the quantity of value in a nation. The value of that which a nation imports is measured by the value of the produce of its labour which it gives in exchange. Whatever be the sum of commodities which this produce will purchase, it is evident that this value will be always the same. But as foreign trade puts us in possession of the same enjoyments at a cheaper rate, it affords an incentive to saving, and accumulation may take place equally well from a diminution of expenditure while production continues the same, as from an increase of production.

Two countries will trade together, so long as the quantity of commodities obtained by each, is greater by means of the exchange, than it would have been had each produced for itself. Suppose the cost of producing cloth is in England 50, and in Portugal 100, and the cost of producing wine is in Portugal 50, and in England 100. The cloth and the wine are exchanged against each other, and England buys, with the labour of 50 Englishmen realised in cloth, a quantity of wine which would have cost the labour of 100 Englishmen, had they been employed about that commodity. Portugal obtains an equal advantage with respect to the cloth. In this case mutual benefit results from the trade. But the trade might still continue, though only one of the countries should derive benefit from it. Suppose the cost of wine were to fall in England from 100 to 30, and that England were in consequence to manufacture wine for herself. Though Portugal has now an inequality in the production of wine of 30 to 50, yet if there were no other market where she could buy cloth, she might still send wine to England, though she would obtain a smaller quantity of cloth in exchange, and the price of the latter would rise in her country, inasmuch as a greater portion of its annual produce must be delivered to procure the same quantity of the cloth. The reason why Portugal submits to this disadvantage is, that in the production of wine she has only a disadvantage of 30 to 50; in the production of cloth, she has a disadvantage of 50 to 100. If such a situation as this were to occur it would indeed (as Ricardo observes) answer to England still to confine her production to the cloth in which she had the greatest advantage, and to buy her wine of Portugal – in which case the trade would, as before, be reciprocally advantageous. But they might possibly, I think, in the case just cited, continue to trade though one only of the parties was benefitted by it.

On reflecting further upon this subject, I am inclined to think that the trade could not continue unless it were mutually advantageous. It will be beneficial to each country in either of the two cases;

1^st When each of the two countries has an advantage peculiar to itself in the manufacture of a particular commodity; as when Portugal has an advantage in the production of Wine, and England in that of Cloth.

2^dly When the advantages in the Production of both the articles belong to the same Country, but not in an equal degree; as when Portugal has an advantage of 50 per Cent over England in producing wine, but only of 30 per Cent in producing cloth. In this case it will be beneficial to each country that the former should confine herself to the production of wine, in which her peculiar properties may be best displayed; and that England should manufacture the Cloth for both countries, the species of production in which she has the smallest disadvantage.

Now the only state of things which does not fall under one of these two heads is

3^dly When the advantages in each sort of production belong to one of the two countries, and in an equal degree; as when Portugal has an advantage over England of 50 percent in producing both wine and cloth; In this case I do not think there can be any trade between the two countries. Each must manufacture all the wine and cloth for its own consumption. To obtain the same quantity of wine and cloth, England will be obliged to devote twice as much of her Productive Powers as Portugal, but from this disadvantage she has no remedy.

Effect of taxation on foreign trade

I do not see how Taxation could subject our foreign trade to disadvantage (more than the home trade). When two countries carry on a trade, the amounts of values imported and exported must be measured against each other.

It answers to an English merchant to send cloth to Portugal, and to bring home wine, when the wine which he brings home will command more goods in England than the cloth which he sends out. It answers to a Portuguese merchant to send wine to England, and to bring home cloth in return, when the cloth which he brings home will exchange for more commodities in Portugal than the wine which he sends out. Each merchant measures the value of what his exported goods will command in his own country against what his imported goods will command in his own country.

Now the reason which induces two countries to engage in this trade is, that a greater quantity of cloth can be manufactured in England and a greater quantity of wine produced in Portugal, by the same quantity of capital and labour. Let us suppose the cost of production of cloth in England to be 50 and in Portugal 100; and also that of wine to be in Portugal 50 and in England 100. I mean of quantities of cloth and wine which were exchanged one against the other. England then in producing cloth and exchanging it for wine reduced her cost of production of wine really to 50 instead of £100. Suppose now that by the operation of a tax imposed in England, the cost of cloth is increased to 60. In this case probably the same quantity of cloth at the increased cost will continue to be exported, and the same quantity of wine to be imported. But the merchant to reimburse himself will be obliged to charge a higher price for his wine. So that the tax on cloth which first raised its cost, will ultimately operate as a tax on wine and enhance its cost to the English consumer.

It is possible indeed that if no other country but England could manufacture cloth at any thing like the same cost, cloth might rise in price in Portugal – or a smaller quantity of it might be given in order to obtain the same quantity of wine. It is not easy to ascertain how much of this effect would take place, for it would evidently answer to Portugal to give the increased price rather than relinquish the exchange with England, since she would obtain even after the advance, cloth at a cost of 60 which she could not make for herself under a cost of 100.

Perhaps it is impossible to determine the effect of taxation on foreign trade without the possession of some data which would enable us to ascertain by what law the general prices of foreign commodities are determined. The cause which enables us to discover, with any degree of accuracy, the prices of commodities at home, is, that we there find the love of gain operating on the human mind, pure and unopposed by any conflicting principles, and our knowledge of the effects of this motive tells us, that equal quantities of the Productive Powers must be equally rewarded. The transfer of Capital and equalisation of Profit which inevitably takes place in the home trade, is in the foreign trade prevented by the love of country and by other powerful principles which render emigration one of the bitterest of all remedies.

We can, I think, only discover two extremes between which the prices of Foreign Commodities must always be found. Let us suppose two countries, Portugal and England, each consuming an equal quantity of wine and cloth, but with advantages so disposed, that England can produce the cloth which she consumes for the labour of 500 Englishmen, and the wine which she consumes for the labour of 1000, and that Portugal can produce her own wine by 500 Portuguese and her own cloth for 1000. Now as the consumption of each country is equal when they trade together, 1000 Portuguese producing wine and 1000 English producing cloth, can supply both countries. It is obvious however that Portugal can never raise the price of the wine which she sends to England above the labour of 1000 Englishmen, for if she did, it would answer to England rather to produce the wine for herself. It is equally evident that England can never raise the price of her cloth above the labour 1000 Portuguese, by which Portugal could supply herself with the article. And if Portugal possessed an advantage in both sorts of Production – if she produced Wine for 500 and Cloth for 600, and if England produced {both} for 1000 labourers, though the trade would still continue, yet England could never raise the price of the cloth which it cost her 1000 labourers to produce, above that of 600 Portuguese labourers, the number for which Portugal can supply herself. But below this limit, it seems impossible to determine either the price of wine in England, or that of cloth in Portugal, for it will evidently answer to each country to continue the trade, so long as the article is obtained by importation at a less price than it would be if produced on her own soil. And as we can discover no general law for ascertaining the price of the productions of one country in another, so it seems impossible to determine the precise effect which Taxation would produce upon Foreign Trade, since this effect must operate altogether upon prices.