II. THE TRUE ORIGIN OF ABSOLUTE VERSUS COMPARATIVE COST ADVANTAGE

Ricardo’s famous numerical example is credited in economics textbooks for refuting the theory of absolute advantage in the cost of production. Therefore, it seems to be the logical starting point for searching for the earliest contraposition of absolute and comparative advantage. A careful reading of the relevant paragraphs in the Principles reveals that what Ricardo originally intended to illustrate with the four numbers was the proposition that “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” (vol. 1, p. 133).Footnote ² The rule he had in mind here was that of the respective quantities of labor embodied in these commodities (vol. 1, pp. 134–135). Ricardo then mentioned a possible implication of this proposition, namely that Portugal might import cloth from England although it “could be produced there with less labour than in England” (vol. 1, p. 135). As I have already shown in previous writings (Morales Meoqui Reference Morales Meoqui2011, pp. 754–755; and Reference Morales Meoqui2017, pp. 39–40), it would have been impossible for him to prove this implication in a mutually beneficial exchange without contradicting one of the core tenets of his theory of value.

The term “comparative advantage,” however, is nowhere to be found in the chapter “On Foreign Trade.” It appears for the first and only time in the Principles in chapter 19, “On Sudden Changes in the Channels of Trade,” where Ricardo stated the following: “A new tax too may destroy the comparative advantage which a country before possessed in the manufacture of a particular commodity; or the effects of war may so raise the freight and insurance on its conveyance, that it can no longer enter into competition with the home manufacture of the country to which it was before exported” (vol. 1, p. 263). Hence, Ricardo’s use of the term is not directly related to the two propositions he announced and fully proved in the famous numerical example.

The term “absolute advantage”—or “absolute cost advantage”—on the other hand is entirely absent from the Principles. More importantly, there is not even a trace of a critique towards Smith with regard to the proper rule for international specialization. This is indeed surprising, considering that Ricardo had announced in the preface of the book that he would “advert more particularly to those passages in the writings of Adam Smith from which he sees reason to differ” (vol. 1, p. 6). As he was quite explicit with his criticism of Smith on many other issues, I wonder why he was so subtle in the case of Smith’s rule for international specialization.

The search for the term “absolute advantage” in the Wealth of Nations yields an equally unexpected result. Smith mentioned it only twice there, in relation to the advantages of the colonial trade for the metropolis.Footnote ³ The term does not appear, though, in the two adjacent paragraphs, which most scholars quote as textual evidence for his alleged adherence to the theory of absolute advantage in the cost of production.Footnote ⁴

So, while the terms “absolute advantage” and “comparative advantage” are to be found in the Wealth of Nations and the Principles, respectively, there is no contraposition of them in these books. Instead, the terms are mentioned on the fly, and neither Smith nor Ricardo bothered to offer a precise definition of them.

The real genesis of the popular contraposition of absolute and comparative advantage is John Stuart Mill’s Essays on Some Unsettled Questions of Political Economy (Reference Mill1844). According to his account in the preface of the book, the five essays were written in 1829 and 1830. With the exemption of the fifth, though, the essays remained unpublished until 1844 due to a perceived lack of public interest in the respective topics. In the first essay, titled “Of the Laws of Interchange Between Nations; and the Distribution of the Gains of Commerce among the Countries of the Commercial World,” Mill stated the following:

The above passage is indeed the very first occasion in which the later inseparable tandem of absolute and comparative advantage appeared together. Mill defines “absolute advantage in the cost of production” as the ability to produce an article with less labor and capital than a foreign manufacturer. He further claims that it is not necessary to have an absolute advantage in the cost of production to export to another country. One won’t find a clear-cut definition of comparative costs or comparative advantage in the essay, though. Instead, Mill offers there several numerical examples that allegedly illustrate that international trade is based on a comparative rather than an absolute advantage in the cost of production. Since he conveniently branded his propositions in the essay as further developments of Ricardo’s principles of foreign trade (Mill Reference Mill1844, pp. 5 and 14), Mill laid the foundation for the currently predominant view that Ricardo had discovered a new law for international specialization. This explains why nobody referred to a law or principle of comparative advantage prior to the publication of the essay. There is no reference to this law, for example, in James Mill’s rendering of Ricardo’s insight in the Elements (Mill [1821] Reference Mill1826) and in the article “Colony” for the Encyclopædia Britannica (Mill [1818] Reference Mill1825). John Stuart’s father was a close friend and collaborator of Ricardo. He passed away in 1836, a few years before the unpublished essays of his son were finally sent to the printing press.

In the next two sections, I analyze more in detail those passages of the Wealth of Nations and the Principles that are mostly associated with Smith’s alleged theory of absolute cost advantage and the so-called Ricardian theory of comparative advantage, respectively. The aim is to find out whether the absolute-versus-comparative-advantage framework accurately conveys what they wrote about the basis of international trade and the rule for specialization.

III. SMITH’S GENERAL RULE FOR SPECIALIZATION

As cost-conscious consumers, we usually look to buy the necessities of life as cheaply as possible. While searching for bargains, we do not bother much whether the desired commodities were produced within the country or beyond its borders. Moreover, whenever we are confronted with the rare choice of making a product by ourselves or buying it from someone else, our decision is often guided by the maxim that one should never attempt to make something that costs less to buy.

In this respect, we are probably quite similar to Smith and his contemporaries. As a keen observer of the economic activities of his time, Smith realized that these common practices were beneficial not only for the individual consumer but for the country as a whole. Consequently, he recommended that governments abstain from protecting national industries that could not produce commodities cheaper than their foreign competitors could, stating in the Wealth of Nations:

He further explained his point of view in the following paragraph:

Smith believed that individual consumers, as well as the country, benefit from acquiring commodities at cheaper prices. In line with this belief, it was of no importance to him whether the origin of the advantages that one country has over another for producing certain commodities more cheaply was natural or acquired (WN IV.ii.15, pp. 458–459). The benefits of buying commodities from the cheapest seller were so obvious to him that he thought they did not require much explanation:

It takes indeed little effort to convince consumers that they should buy a cheaper commodity instead of a dearer one of similar quality, irrespective of its place of origin. They realize perfectly well that with the money saved, they can afford to buy additional commodities and services of local and foreign origin, and thus increase their level of consumption. By contrast, it is always a daunting task for interested merchants and manufacturers to lure their compatriots into buying dearer commodities from them. Promotional campaigns in favor of buying locally produced goods often fail to convince enough people to have a significant economic impact. This is why protectionists usually prefer to call for tariffs and quotas, whose sole purpose is to make foreign commodities artificially dearer for consumers. If protectionists had any real faith in their arguments about the alleged benefits of buying locally produced commodities, they would present their case to the public and then let consumers make their purchase decision based on undistorted market prices.

Smith was well aware of the fact that the cheapest producer of a commodity might not necessarily be the most productive one, as can be seen in the following quote:

Food producers in richer countries usually invest more than their competitors in poorer countries in a comparable acreage. This higher expenditure often results in a higher level of production but not necessarily in proportion to the additional spending. Consequently, the producers in richer countries are not always able to match the prices of their competitors from poorer countries, despite having a productivity advantage over the latter. Smith attributed this to the fact that the agricultural sector does not admit the same degree of division of labor as manufactures. He then added: “But though the poor country, notwithstanding the inferiority of its cultivation, can, in some measure, rival the rich in the cheapness and goodness of its corn, it can pretend to no such competition in its manufactures; at least if those manufactures suit the soil, climate, and situation of the rich country” (WN I.i.4, p. 17). His dictum might have been valid for the economic realities of the late eighteenth century. Nowadays, however, developing countries export a wide range of manufactures to Europe and the US as well.

In summary, it seems uncontroversial to affirm that Smith considered the possibility of acquiring commodities at lower prices as the main motivation for trading with other countries. Thus, it seems only logical to conclude that he viewed the relative cheapness of foreign commodities as the basis of international trade. Moreover, Smith’s corresponding rule for specialization was that one should not attempt to make a commodity that costs less to buy. A positive formulation of the rule would be that one should acquire commodities abroad whenever they are offered more cheaply than what their internal production would cost. The next section is dedicated to proving Ricardo’s agreement with, and recurrent use of, what I like to call the “classical rule for specialization.”

IV. RICARDO’S AGREEMENT WITH SMITH ON THE CLASSICAL RULE FOR SPECIALIZATION

It is quite easy to find passages in the Principles where Ricardo explicitly stated that he also considered the relative cheapness of foreign commodities as the basis of most international exchanges. Perhaps the clearest statement in this respect is the one already quoted in the introduction, where he affirms that “the motive which determines us to import a commodity, is the discovery of its relative cheapness abroad: it is the comparison of its price abroad with its price at home” (vol. 1, p. 170). He further emphasized this rule by pointing out that if a farmer decides to sell his corn below the price at which it is currently imported, this importation will stop under the supposition that the importer cannot reduce the price of the foreign corn even further (vol. 1, p. 269). In the following passage, Ricardo even expressed his explicit agreement with Smith on the benefits of acquiring commodities in the markets with the cheapest prices:

The above quotes are at odds with the current depiction of Ricardo’s numerical example in economic textbooks. If the textbook interpretation were accurate, then the “master logician of political economy” (Maneschi Reference Maneschi2004, p. 435) would have contradicted himself with the four numbers. In reality, though, the numerical example is perfectly aligned with these quotes, as will be shown below.

In the featured barter trade in chapter 7 of the Principles, an unspecified amount of cloth produced by 100 Englishmen working for a year is exported to Portugal in exchange for an unspecified amount of wine, which required the labor of 80 Portuguese men also working for a year. This exchange of unequal quantities of labor could not take place between individuals of the same country, as Ricardo indicated in the following paragraph:

The featured barter trade could not take place between individuals of the same country, because according to Ricardo’s theory of value, the produce of 100 Englishmen would have to have a higher cost of production and exchange value than the produce of 80 Englishmen. It could take place between English and Portuguese traders, though, since the relative value of two lots of commodities produced in different countries is not determined by the respective quantities of labor devoted to the production of each lot. As Ricardo explicitly indicated, this is due to a difference in the rate of profit between countries (vol. 1, p. 136). The difference persists as long as capital is relatively less mobile internationally than nationally.

A possible implication of the above proposition is that it might be beneficial for Portugal to import cloth from England, although she could produce it internally with less quantity of labor. What makes this logical implication so counterintuitive is the inference that the English cloth made by 100 men would have to have a higher cost of production than the Portuguese cloth made by only 90 men. This inference will be proven wrong by showing that the English cloth had to have a lower cost of production than the Portuguese cloth. Consequently, the English cloth could be offered more cheaply in the Portuguese market than the locally produced cloth. In correspondence with the current zeitgeist in economics, the proof will be presented in mathematical language.

Before diving into the proof, though, some terminological clarifications are required. In chapter 7 of the Wealth of Nations, Smith distinguished between the market price and the natural price of a commodity. The latter was the price that is “sufficient to pay the rent of the land, the wages of the labour, and the profits of the stock employed in raising, preparing, and bringing it to market, according to their natural rates” (WN I.vii.4, p. 72). At the natural price, Smith added, the commodity is sold for what it really costs the person who brings it to market. This is important to remember, as the natural price also played a central role in Ricardo’s theory of value. After praising Smith’s chapter (vol. 1, p. 91), Ricardo stated: “In speaking then of the exchangeable value of commodities, or the power of purchasing possessed by any one commodity, I mean always that power which it would possess, if not disturbed by any temporary or accidental cause, and which is its natural price” (vol. 1, p. 92). He also stated later in the Principles that “the real and ultimate regulator of the relative value of any two commodities, is the cost of their production” (vol. 1, p. 344). Both affirmations are equivalent, as the natural price of a commodity was for Ricardo just another name for its cost of production (vol. 2, p. 46). He used indeed both terms interchangeably in his writings.Footnote ⁵

Furthermore, it is inconsequential for the purpose of this proof whether the two countries have a common currency or not. In the latter case, one can simply presume that there is a given exchange rate between the two currencies at the moment of the exchange.

Let’s define CC_ENG and CW_POR as the respective costs of production of the batches of English cloth and Portuguese wine traded. CW_ENG is the estimated cost of production if the lot of wine would have been produced in England, while CC_POR is the estimated cost of production of the lot of cloth in Portugal. Moreover, I will make the following three assumptions:

1. (1) Since Ricardo’s numerical example features a barter trade, the lot of English cloth must have the same cost of production (natural price) as the lot of Portuguese wine.Footnote ⁶

2. (2) The embodied quantities of labor determine the relative value of commodities made in the same country but not of commodities produced in different countries.Footnote ⁷

3. (3) When the production of an unspecified amount of cloth requires more (less) quantity of labor than the making of an unspecified amount of wine in the same country, then one can also presume that the cost of production of the cloth is higher (lower) than the cost of production of the wine.Footnote ⁸

These three assumptions are supported by explicit statements in the Principles and Ricardo’s correspondence, as indicated in the respective footnotes. Moreover, none of them is at odds with reality. This contrasts sharply with the many unrealistic assumptions of the textbook trade model of comparative advantage, which lack any foundation in Ricardo’s writings and contradict quite openly core tenets of his economic theory.

As stated in assumption (1), the English cloth and the Portuguese wine traded have to have the same cost of production, or CC_ENG = CW_POR. Because of the assumptions (2) and (3), and the fact that the making of the wine in England would have required twenty additional men working for a year, compared with the production of the cloth, it must be that CW_ENG > CC_ENG. The exchange is therefore beneficial for England, since it would have cost her more to produce the wine locally than to buy it from Portugal.

Since CC_ENG = CW_POR and CW_ENG > CC_ENG ⇒ CW_ENG > CW_POR, by applying transitivity, thus, it is proven that the cost of production of wine in England has to be higher than in Portugal. There is nothing counterintuitive about this conclusion. On the contrary, it seems perfectly logical that the cost of production of the produce of 120 men (CW_ENG) is higher than that of 80 men (CW_POR). The counterintuitive insight appears in the analysis of Portugal’s numbers.

Because of the assumptions (2) and (3), and the fact that the imported cloth would have required the labor of 90 men in Portugal, while the exported wine required only the labor of 80 men, we can presume that the local production of the cloth would have cost more than the making of the wine, or CC_POR > CW_POR. The exchange is therefore beneficial for Portugal, since it would cost her more to make the cloth than to buy it from England with the exportation of wine.

Since CW_POR = CC_ENG and CC_POR > CW_POR ⇒ CC_POR > CC_ENG, by simply applying transitivity, it is proven that the cost of production of cloth in Portugal has to be higher than in England. Thus, Portugal had no production cost advantage over England in cloth making. This result is not contradicted by the fact that the making of the English cloth requires the labor of ten additional men working for a year compared with that of the Portuguese cloth, as Ricardo explained that the embodied quantities of labor do not determine the relative value of commodities produced in different countries. Thus, the cloth may well be produced more cheaply in England than in Portugal, although its production would have required less quantity of labor in the latter.

Furthermore, this interpretation is perfectly aligned with the content of the subsequent pages of chapter 7. Ricardo stated there that “cloth cannot be imported into Portugal, unless it sell there for more gold than it cost in the country from which it was imported; and wine cannot be imported into England, unless it will sell for more there than it cost in Portugal” (vol. 1, p. 137). The cloth and wine can be imported, though, only as long as local producers are unable to undercut or at least match the prices at which these commodities are currently imported. After already mentioning on page 137 that the price of cloth in Portugal is indeed higher than in England, Ricardo indicated in the very next page that the price of wine in England is 50l. per pipe, and the price of a certain quantity of cloth is 45l., while in Portugal the price of the same quantity of wine is 45l., and that of the same quantity of cloth 50l.

By taking the famous four numbers as a basis, and assuming that 45l. is the natural price (cost of production) of cloth in England and wine in Portugal, it is possible to make a sound estimate of the respective costs of production if the lot of cloth had been produced in Portugal and the pipe of wine in England. In the case of the latter, one has to multiply the 45l. by 1.2 (120/100), resulting in an estimated cost of production of 54l. for a pipe of English wine. After multiplying the 45l. with 1.125 (90/80), one obtains an estimated cost of production of 50.625l. for the lot of Portuguese cloth. In both cases, thus, the estimated costs of production for the English wine and the Portuguese cloth are above 45l., the natural price of cloth in England and wine in Portugal. This further confirms that the cost of production of the lot of cloth is higher in Portugal (50.625l.) than in England (45l.). Moreover, the estimated costs of production of the Portuguese cloth and the English wine are both above 50l., the price of wine in England and cloth in Portugal indicated by Ricardo. Under these conditions, the cloth and the wine would not be supplied by local producers. Evidently, the master logician of political economy did not leave any loose ends.

This novel interpretation of the numerical example establishes, arguably for the first time, a logical connection between the quantities of labor and the prices indicated by Ricardo in chapter 7 of the Principles. Moreover, it gives the simplest possible answer to the question of why the merchant is able to sell the English cloth in Portugal: because it can be offered more cheaply than the locally produced cloth. Thus, the new interpretation of the numerical example eliminates the two alleged difficulties in Ricardo’s text mentioned by Faccarello (Reference Faccarello2015, pp. 759–764).

All of the above indicate that Ricardo had the classical rule for specialization in mind when he carefully selected the four numbers. They are indeed perfectly in line with Smith’s recommendation that one should always buy from the cheapest seller, irrespective of whether this seller is located in the same country or abroad. This further confirms a previous assertion (Morales Meoqui Reference Morales Meoqui2014) that the level of compatibility between the theories of international trade of Smith and Ricardo is significantly higher than currently reflected in the economic literature.

On the other hand, the famous four numbers offer no proof whatsoever for the claim that a foreign producer does not need to have a lower cost of production to sell a commodity of similar quality in another country, which is the way they have been interpreted for over 170 years. This also explains why Ricardo never claimed to have discovered a new law, principle, or rule for international specialization, and why the contraposition of absolute and comparative advantages in the cost of production is nowhere to be found in the Principles. These notions originated from an erroneous interpretation of Ricardo’s numerical example popularized by J. S. Mill.

V. J. S. MILL’S MISINTERPRETATION OF RICARDO’S NUMERICAL EXAMPLE

Although the textbook trade model of comparative advantage is commonly referred to as the Ricardian model, many of its prominent features and unrealistic assumptions can be traced back to J. S. Mill. He was indeed the first, for example, in proposing to leave capital and the cost of carriage out of consideration (Mill Reference Mill1844, pp. 5–7). Moreover, he also came up with the method of determining comparative advantage by comparing cost ratios, with the implicit assumption that the costs remain constant (Mill Reference Mill1844, p. 12). These features and assumptions of the textbook trade model of comparative advantage, for which Ricardo has been unfairly criticized in the past, are significant departures from the original numerical example in the Principles. The comparison of cost ratios may even recommend an exchange, which Ricardo would have deemed detrimental for the country (Morales Meoqui Reference Morales Meoqui2017, pp. 42–43).

In addition to this, the previous section has revealed that J. S. Mill misunderstood the four numbers in a more fundamental way than previously thought. From Ricardo’s indication that the quantities of labor embodied in the cloth and wine did not determine how much wine should be given for the cloth, J. S. Mill derived the following: “The principle, that value is proportional to cost of production, being consequently inapplicable, we must revert to a principle anterior to that of cost of production, and from which this last flows as a consequence,—namely, the principle of demand and supply” (Mill Reference Mill1844, p. 8).

This contradicts what Ricardo repeatedly stated in the Principles: that “it is the cost of production which must ultimately regulate the price of commodities, and not, as has been often said, the proportion between the supply and demand” (vol 1, p. 382). The reason for this contradiction between Ricardo and J. S. Mill pops up immediately when contrasting their respective definitions of the “cost of production.” For Ricardo, the natural/necessary price was synonymous with the cost of production, and the latter always included the ordinary profits (vol. 1, p. 47; vol. 2, p. 369; vol. 4, p. 25). By contrast, J. S. Mill’s notion of the cost of production was different from that of the necessary price, as it did not include the ordinary rate of profit.Footnote ⁹ It rather corresponded to what Smith called the “prime cost” of a commodity (WN I.vii.5, pp. 72–73). Nowadays it is mostly referred to as the “cost price” of commodities.

Two bundles of commodities with the same cost price but different profit rates do not have the same value. Therefore, J. S. Mill mistakenly inferred from the divergence of profit rates between countries that the cost of production could not be the ultimate regulator of the relative value of any two commodities in international exchanges. This inference is obviously incorrect in the case of Ricardo’s notion of cost of production, which includes the ordinary profits in the location where the commodities were produced. Thus, the determination of the commodity terms of trade in international exchanges was not a problem that Ricardo had left for subsequent generations of economists to solve.

J. S. Mill first published the solution for this alleged problem, the theory of reciprocal demand, in the Essays (Reference Mill1844), and later reproduced it with some extensions and omissions in chapter XVIII of his Principles of Political Economy with some of their Applications to Social Philosophy ([1848] Reference Mill and Robson1965). It was positively received by John Elliott Cairnes (Reference Cairnes1874) and Charles Francis Bastable ([1887] Reference Bastable1897). The initial statement in the form of arithmetical examples was later translated into geometrical form by Alfred Marshall (Reference Marshall1879, Reference Marshall1923) and Francis Ysidro Edgeworth (Reference Edgeworth1894). From this gradually emerged a new branch of international trade theory, which, following Marshall, was called the “pure theory of international trade.” J. S. Mill’s analysis of the relationship between reciprocal demand and the commodity terms of trade has been considered his chief claim to originality in the field of economics (Viner Reference Viner1937, p. 535), so much so that Frank D. Graham’s critique of Mill and his followers did not mention Ricardo at all (Graham Reference Graham1923). Thus, it was indeed J. S. Mill “who gave the analysis of comparative advantage the form that became an engine of analysis for generations to come and the starting point for all further developments in trade theory” (Ruffin Reference Ruffin2002, p. 742). Consequently, he also bears the bulk of the responsibility for almost two centuries of extreme confusion about Ricardo’s numerical example.

Notwithstanding, J. S. Mill cannot be blamed for all the flaws in the current textbook narrative on the law of comparative advantage. So far, I have not found in his writings any explicit reference to the existence of such a law, and neither did he attribute the theory of absolute cost advantage to Smith. Moreover, he even appears to have realized that the contraposition of absolute and comparative cost advantage was prone to originate confusion about the proper rule for specialization, as he stated in his monograph:

As he further clarified in the next paragraph, what he meant to explain is why a “thing may sometimes be sold cheapest, by being produced in some other place than that at which it can be produced with the smallest amount of labour and abstinence” (Mill [1848] Reference Mill and Robson1965, vol. III, p. 587). Thus, J. S. Mill also considered the relative cheapness of foreign commodities as the basis for most international exchanges and used the same rule for specialization as Smith and Ricardo. His attempts of clarification ultimately failed, though, because he reiterated his previous claims that the cost of production did not regulate the relative value of commodities produced in different countries, and that Ricardo had allegedly shown that “it is not a difference in the absolute cost of production, which determines the interchange, but a difference in the comparative cost” (Mill [1848] Reference Mill and Robson1965, vol. III, p. 589).

Since the mid-nineteenth century, successive generations of economists have been trained to thinking about the proper rule for specialization mostly in terms of a dichotomy between absolute and comparative cost advantage. Hence, some scholars may conclude that I am claiming here that both Smith and Ricardo were advocates of the theory of absolute cost advantage, as John King (Reference King2013, p. 466) suggested with respect to one of my previous papers. This is why I would like to emphasize that it is definitely not the case.

If the theory of absolute cost advantage is understood in the sense that commodities are always bought from those places where its production is cheapest, then I believe that neither Smith nor Ricardo would have subscribed to it. They knew too well that governments might grant subsidies and preferential tariffs to particular manufacturers and countries, which may offset any production cost disadvantage. But even in the hypothetical case that all governments would adhere to the principle of free trade and refrain from these policy tools, commodities might not always be acquired from the cheapest producer. For one thing, the cheapest producer might not have the production capacity to satisfy the world demand of the commodity. In addition to this, one should remember that the cost of production includes not only the costs of the producer of the commodity but also all the other expenses necessary to bring it to market—for example, the cost of carriage, insurance, and advertising, among others—as well as the ordinary profits for the producer and merchants. The cost of production of a commodity keeps adding up at every stage of the process of production and distribution. Thus, the natural price of a commodity at the factory gate is only a fraction of its natural price at the retail counter. It is therefore quite common to discover that the cheapest provider for a specific market is someone other than the producer who can afford to offer the lowest ex-factory price for the commodity.

For all the reasons mentioned above, it seems to me that neither absolute nor comparative advantage in the cost of production are proper notions to explain the basis of international trade and the proper rule for specialization. Given the problematic origin of these terms and the ongoing confusion generated by their contraposition, I would not recommend using them to popularize the genuine and still valid insights of Smith and Ricardo on international trade.