Home > RWER > RWER issue 54: John Duffield

RWER issue 54: John Duffield

Real-World Economics Review, issue 54
You may read the whole paper here: http://www.paecon.net/PAEReview/issue54/Duffield54.pdf

Ricardian “comparative advantage” is illusory

John Duffield  

Part 1

             The doctrine of “comparative advantage” attributed to David Ricardo remains a staple in the apologetics of corporate globalization:  If international trade only benefits rich and poor countries alike, its opponents are summarily indicted for plain misanthropy.  In the nearly two centuries since its appearance, the Ricardian doctrine has been criticized by both Right and Left of the politico-economic spectrum, but critical consensus still awaits a definitive refutation of Comparative Advantage on its own terms.  This essay presents just such a refutation.1

            Ricardo’s paradigm case2 for Comparative Advantage is analytic, deriving from a model of the world economy consisting of two originally closed systems (autarkies), each producing the same two commodities (here, food and clothing) as the other.  Commodities of the same description are qualitatively identical, so that advances (or declines) in productivity are manifested only quantitatively.  A simple labor theory of value provides the numéraire for each closed system, such that the original intrasystemic exchange ratio of F (food) and C (clothing) for each system is set by the quantity of simple labor (labor undifferentiated by skill or productivity) required for the production of the respective commodities.  Equivalent-value exchange is the intrasystemic rule.  Intersystemic labor migration is prohibited, as is the intersystemic migration of know-how (skill); while intrasystemic labor mobility is assumed to be a completely free option, such that the producers of either commodity may switch to the production of the other commodity—thus, the “specialization” advocated by Ricardo—without encountering barriers to entry, such as training, transportation costs, monopoly, etc.  Although Comparative Advantage turns upon intersystemic differences in productivity, whereby the commodity-exchange ratio (“relative cost” of F and C) differs from one system to the other, technology per se is not quantified in the Ricardian problematic, which modern texts present as a single-factor model that turns upon labor cost per unit of output.  At this level of abstraction, the productivity differentials that animate Comparative Advantage inhere in the laborers themselves and thus are equivalent to skill differentials. 

            From the above problematic modern texts deduce that intersystemic trade is economically beneficial to all concerned (and harmful to none) if, and only if, the laborers of each system specialize in the production of that commodity for which the “relative cost,” given in the commodity-exchange ratios of the respective closed systems, is more favorable.  In the example detailed below, the laborers of system I should specialize in F because F costs half as much as C in system I, given the original (closed system) commodity-exchange ratio (1 F = 1/2 C); while F costs three-fourths as much as C in system II, given system II’s autarky exchange ratio (1/3 F  1/4 C, i.e., 1 F  3/4 C).  The “comparative advantage” of system I is thus in the production of F.  (The “” sign stands for “equals” or “exchanges for.”  Under Ricardo’s labor theory of value “exchanges for” signifies equality of labor content only if the exchange transpires within either closed system; intersystemic exchange in accordance with Comparative Advantage proceeds by the calculation of mutual benefit described below.)  By

the same token the laborers of system II should specialize in the production of C, since C in their (closed) system costs four-thirds as much as F (1 C = 4/3 F), while in system I C costs twice as much as F (1 C = 2 F).  System II’s comparative advantage is thus in C.  It will advantage the laborers of system I, specializing in F, if they can trade for C at a rate better than the 1 F = 1/2 C that would obtain in their system absent specialization.  Thus they will require a trading rate of 1 F > 1/2 C (“>” for “greater than” or “is worth more than”; “<” for “less than” or “is worth less than”).  It will advantage the laborers of system II, specializing in C, if they can trade for F at a rate better than their closed-system rate of 1 C = 4/3 F; i.e., their required trading rate will be 1 C > 4/3 F, which is equivalent to 1 F < 3/4 C.  Collating the requirements, intersystemic trade will benefit all concerned in the trading range given by (1/2 C < 1 F = nC < 3/4 C).  Within this range the rate coefficient n is undetermined; let it be arbitrarily set at 1 F = 2/3 C.  Then by trading for C the laborers of system I will enjoy the equivalent of a 33 1/3% increase in the C-productivity of their F over their closed-system rate

( [2/3 C – 1/2 C] ¸ 1/2 C = 1/3 = 33 1/3%).  To the extent that they trade for C instead of producing it, their real income increases by 33 1/3%.  (Their F-productivity per se is seen as unchanged.  The overall increase in real income will depend on system I’s current-income shares of F and C, a parameter that is independently assigned.)  By the same token, trading for F at 1 C = 3/2 F ( º 1 F = 2/3 C; the symbol “º” for “equivalent to”) brings to the laborers of system II the equivalent of a 12 1/2% increase in the F-productivity of their C over their closed-system rate ( [3/2 F – 4/3 F] ¸ 4/3 F = 1/8 = 12 1/2%).  Their real income increases by 12 1/2% to the extent that they trade for F instead of producing it.  (Their C-productivity per se is seen as unchanged; the overall increase in system II’s real income will depend on the current-income shares of F and C.)

1               “In the theory of international trade Ricardo stated explicitly for the first time the law of comparative advantage . . . . The law of comparative advantage survives as an important part of the theory of international trade today.”  Graham Bannock, R.E. Baxter and Evan Davis, The Penguin Dictionary of Economics, 7th ed. (London:  Penguin, 2003), “Ricardo, David,” pp. 336-7.  “Ricardo may not have been the first economist to advance the concept of comparative advantage.  Another Englishman, Colonel Robert Torrens, included a brief, very rough formulation of the law of comparative advantage in one paragraph in ‘An Essay on the External Corn Trade’ (1815), but Ricardo’s treatment of the topic is more explicit and influential . . . . historians of economic thought consider it more likely that Ricardo arrived at his conclusions independently.”  Morgan Rose, “ A Brief History of the Concept of Comparative Advantage,” http://www.econlib.org/library/Columns/Teachers/comparative.html.  “There is no body of economic theory that has achieved greater professional acceptance than David Ricardo’s theory of comparative advantage and the modern emendations of Ricardo’s ‘law.’  Criticisms of comparative advantage and its extensions from a theoretical perspective have not resulted in any substantial weakening of the overall strength of this body of theory nor of its corollary, a free trade regime.”  James M. Cypher and James L. Dietz, “Static and Dynamic Comparative Advantage,” http://www.highbeam.com/doc/1G1-20970221html.  (Article orig. in Journal of Economic Issues, June 1, 1998.)  “At the deepest level, opposition to comparative advantage—like opposition to the theory of evolution—reflects the aversion of many intellectuals to an essentially mathematical way of understanding the world.  Both comparative advantage and natural selection are ideas grounded, at base, in mathematical models . . . . ’’ Paul Krugman, “Ricardo’s Difficult Idea,” http://web.mit.edu/krugman/www/ridardo.htm.  “Nobel laureate Paul Samuelson  (1969) was once challenged by the mathematician Stanislaw Ulam to ‘name me one proposition in all of the social sciences which is both true and non-trivial.’  It was several years later than [sic] he thought of the correct response:  comparative advantage.  ‘That it is logically true need not be argued before a mathematician; that it is not trivial is attested by the thousands of important and intelligent men who have never been able to grasp the doctrine for themselves or to believe it after it was explained to them.’ ”   Paul A. Samuelson, http://www.wto.org/english/res_e/reser_e/cadv_e.htm   (citing P.A. Samuelson ,  “The Way of an Economist,” in P.A. Samuelson, ed., International Economic Relations:  Proceedings of the Third Congress of the International Economic Association,  London: Macmillan, 1969, pp. 1-11). 2               Ricardo’s doctrine is sketched in ch. 7 of his The Principles of Political Economy and Taxation, 1817; rev. 3d ed., 1821.  He died in 1823.  Ricardo’s doctrine is specifically addressed in section II of this essay.  The principal problematic utilized in this essay is the modern textbook case for Comparative Advantage as spelled out by Paul A. Samuelson and William D. Nordhaus, Economics, 14th ed. (New York:  McGraw-Hill, 1992), pp. 663-5.  The Penguin Dictionary of Economics  (Bannock et al., op. cit., pp. 336-7) offers a modern explication that utilizes Ricardo’s original commodities (wine and clothing) and productivity figures, although the latter are couched in terms of man-hours instead of Ricardo’s man-years.You may read the whole paper here: here:http://www.paecon.net/PAEReview/issue54/Duffield54.pdf

  1. September 29, 2010 at 6:46 am

    I believe that the Reicardian comparative advantage was refuted “in its own terms” by Samir Amin in a book from the 70s, I think it was called The global accummulation of capital, or something like that. I also have a faint memory that his argument was attributed to the Greek economist Arghiri Emmanuel.

    His argument was also based in differentials in productivity, and showed that a less productive economy lost by taking part in exchange.

    • John Duffield
      September 30, 2010 at 8:40 pm

      Please sharpen your references. My recollection is that A. Emmanuel, Samir Amin, R. Prebisch et. al. criticized the doctrine from an external and empirical standpoint, i.e., essayed refutation of “comparative advantage” by reference to historical “unequal exchange” that issues in international polarization between the have’s and the have-not’s. This is certainly a telling line of criticism, but it is not refutation of the doctrine on its own terms: Duffield’s paper takes the neoclassical version on its own terms–i.e., with no empirical reference added–and demonstrates logical self-contradictions that reduce the doctrine to absurdity. The paper then shows that the Ricardian original of the doctrine was explicitly class-biased, thus belying the halcyon claims of mutual benefit alleged for the doctrine by modern apologists for corporate globalization. –In any event, the references cited in footnote 1 of the paper demonstrate that nothing offered by Amin et al. is recognized by mainstream academe as a refutation of the doctrine of comparative advantage on its own terms. They think they are still trading in ideological gold. The paper in point demonstrates that this is naught but fool’s gold.

  2. John Duffield
    September 30, 2010 at 7:54 pm

    EMENDATIONS in pdf print of the article: p. 66, line 17 of proof, RIGHT (justifying) col., should read “1, 16, m. ponens, simp.” — the pdf left out the “1”.
    p. 71, 2d line of text from top should have “(p. 64)” — the pdf reproduced the old “(p. 3)” of the original manuscript.
    p. 72, last two lines of text should have “(pp. 64, 67)” — the pdf reproduce the old “(pp. 3, 6)” of the original manuscript.

  3. John Duffield
    October 1, 2010 at 4:47 pm

    EMENDATIONS, cont. p. 73, 9 lines from end of 1st para should have “(above, p. 66)”–the pdf reproduced the old “(above, p. 5)” of the original manuscript.

  4. Peter T
    October 5, 2010 at 4:46 am


    I would be interested to know if you submitted a version of this paper to “orthodox” economic journals, or have had any reaction from mainstream economists to your argument.

    • John Duffield
      October 5, 2010 at 9:04 pm

      Peter T.: No submissions to “orthodox” journals, but I queried a representative sample of mainstream economists–three Ph.D.’s who would speak to a nonacademic such as myself. Two confessed incomprehension of my argument; all three declared their incredulity. All three conceded the accuracy of my account of the received doctrine of Comparative Advantage in the opening pages of my essay. The tenor of the discourse was “theological” in Joan Robinson’s pejorative sense of the term: I, a nonprofessional, was guilty of questioning the unquestionable, i.e., part of the core ideology of mainstream economics. (I am reminded of the incredulity that greeted the disproofs of what turned out to be scientific dogmas, such as determinism and, on a more modest level, the Weltanschauung, both Aristotelian and religious, overturned when Galileo raised his lens to the sky and related his discoveries his “Starry Messenger.” Of course, my reductio ad absurdum of Comparative Advantage is minor-league stuff; exactly proportionate to the scientific stature of neoclassical economics.) I predict that the response of mainstream economics to my essay will be (1) to ignore it; (2) if/when that becomes impossible the argument will be dismissed by (a) misconstruing it and (b) adducing ancillary irrelevancies that do not touch the substance of the argument; and (3) to the extent that, after all, the argument cannot be completely dismissed, its import will be minimized by various ad hoc strategems. Throughout, the ideological role of neoclassical economics in perfecting the necessary illusions of corporate capitalism will be on display. Other precedents relevant to possible ideological responses include the Cambridge Capital Controversy, which did nothing to diminish the hegemony of the neoclassicals; Mirowski’s scathing criticisms of neoclassicism according to Samuelson (in Mirowski’s “More Heat Than Light”), which were similarly shrugged off; and Marx’s argument to the effect that capitalism required both crises and unemployment–initially dismissed but, ever since Keynes, co-opted as indicating only minor blemishes on a fundamentally sound system, corrigible by the interventions of professional economists (of course), ever on the lookout for pelf and position.

  5. Peter T
    October 6, 2010 at 8:54 am


    thanks for the reply. I agrree with your view of the likely reception, but I am not without hope – I anticipate that as environmental limitations bite we will perforce return to smaller economic units as more stable and more amenable to the stringent local control demanded by environmental sensitivities. Economic theory will catch up after 50 years or so, find your argument, and deploy to support the then status quo.

  6. merijnknibbe
    October 7, 2010 at 8:40 am

    Dear John,

    I always try to make things ‘real’. Now. let’s go back to the days of Ricardo (and before). Some of the succes stories of pre-1800 global markets are products like sugar, tobacco, tea, coffee, cotton, spices as well as one of the favorite examples of Adam Smith, the ‘Mother of all trades’ (moedernegotie), the wheat and barley trade from the Baltic and the Northern part of Germany to Amsterdam and from Amsterdam to other Dutch cities, Portugal, Italy etcetera (modern research indicates that a large part of this international grain trade also involved lots of wheat from nearby Zeeland and Friesland).

    All of these markets were, on the supply side, dependend on at least some amount of some kind of slave labor.

    With regard to the grain trade: close to the center of the international grain trade, Amsterdam, markets for land, labor and capital were ‘free’. Seventeenth and even sixteenth century land and labor markets in Friesland, Holland and Zeeland were, even compared with nineteenth (!) century England, quite modern. The same holds for the capital market. Somewhat gradually, this changed when one travelled to the east and in Poland, bound labor and restricted land markets were the rule.

    With regard to the other products: the very increase in trade in these products played the leading role in the extension of the slavery system in for instance the America’s but also in present day Indonesia.

    All these slavery systems were characterized by commercial, monetized, profit and trade oriented enterprises producing the traded goods, in surroundings which were much less monetized and much less market oriented. They were also dependent on trade to obtain money, which was sometimes hoarded but in any case used to buy amazing amounts of luxuries and to build even more amazing houses (to be sure: the Amsterdam capital market also provided capital to for instance plantations in Suriname – these were modern enterprises!). Once in existence, any kind of ‘comparative advantage’ was more or less frozen by systems of bound labor and land.

    My question to you (and others): was Ricardo in fact describing such a system with slave labor? Remember: in the days of Ricardo, grain, cotton, spices, tea, coffee and the like were relatively important in international trade. And do you describe what might happen to comparative advantage when labor is not bound anymore?

    My description of the slavery system is of course extremely short, crude and simplified, I admit. Some literature, however:

    Tielhof, Milja van, ‘The ‘mother of all trades’: the Baltic grain trade in Amsterdam from the late 16th to the early 19th century’

    Peter Priester, ‘Wheat yields in Zeeland from c. 1585 – 1995,’ in Bas J. P. van Bavel and Erik Thoen, Land productivity and agro-systems in the North Sea area. Middle Ages – 20th century. Elements for comparison. CORN Publication series. Comparative Rural History of the North Sea Area 2. Turnhout (Belgium) Brepols (1999)

    Otto Knottnerus, Yeomen and Farmers in the Wadden Sea Coastal Marshes, c. 1500-c. 1900. In: P. Hoppenbrouwers en B. van Bavel (red.), Access to Land in the North Sea Area from the Middle Ages to the 19th century, Turnhout 2004, pp. 149-186.


  7. John Duffield
    October 9, 2010 at 2:21 am

    merijnknibbe: Ricardo’s own problematic of Comparative Advantage was explicitly one of free labor vis-a-vis the capitalist production of wine (Portugal) and cloth (England), per footnotes 4, 6 and 7 of the second part of my essay. Ricardo’s definition of “capital,” given at the end of my footnote 5, is also relevant. His presentation was in the nature of an analytic model, not an historical synthesis such as offered by in your comment. Ergo, my essay–like Ricardo’s own presentation–already describes “what might happen to comparative advantage when labor is not bound anymore” (quoting your question, above).

  8. May 22, 2011 at 5:27 am

    John Duffield, please write to me at ianfletcher@prosperousamerica.org

    I am the author of Free Trade Doesn’t Work.

  9. Jorge Morales Meoqui
    November 27, 2012 at 12:20 pm

    Dear John Duffield,

    I could not find in your paper a refutation of the two propositions that David Ricardo actually stated and fully proved in his famous numerical example in the Principles: first, that his labor theory of value does not regulate the exchange value of commodities in international trade; and second, that a country might import a certain amount of a commodity despite having a real cost advantage in producing the same amount of the commodity with respect to the exporting country. The second proposition is a corollary of the first.

    Here is a link to my paper in case you are interested in reading my interpretation of Ricardo’s numerical example.


    I would be very interested in reading your comments.

    Best regards,


  1. No trackbacks yet.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.