Comparative Advantage – the Unicorn of Free Trade
Saturday, July 18, 2009 at 01:00PM
Skeptic in Economics, Favorites, Free trade

(I finished the following essay in April 2006, before I had a blog and before I found additional relevant materials.  I have not updated the essay, and I apologize that the link to the Paul Davidson quotation redirects from IHT to NYT and then disappears.)

Advocates of liberalized trade often argue that the doctrine of "comparative advantage" articulated by economist David Ricardo in 1817 assures that all free-trading nations will be better off than if trade barriers remain. For example, in chapter five of the updated and expanded edition of Tom Friedman's The World is Flat—a Brief History of the Twenty-First Century, the author says, "My mind just kept telling me, 'Ricardo is right, Ricardo is right, Ricardo is right.'" At 261. He makes clear that he would otherwise have grave concerns about the effects of a flattening world on his family and America. For many others too, the decision to give intellectual and political support to trade liberalization depends significantly, sometimes solely, on faith that comparative advantage will inevitably make us glad one year, five years, and ten years later that we dismantled trade restrictions.

If the comparative advantage idea is incorrect (which I do not contend) or simply inapplicable to our real world (which is exactly what I contend), that does not mean trade liberalization is the wrong policy, but it does mean that trade liberalization has to be justified by other facts and less convincing arguments. The persuasive power of the comparative advantage idea is that it can be portrayed as a kind of "force of nature" like gravity or at least like the pervasive, familiar, and powerful economic force of "supply and demand." Such a force, it is argued, should not be resisted because (i) it is irresistible and if one attempts to resist it he will be catastrophically overwhelmed, and (ii) it is foolish and self-destructive to delay or prevent receipt of the enormous win-win benefits that will surely come to every nation that liberalizes trade. The myth that win-win results are inevitable is especially powerful politically because even people at risk are reluctant to frustrate "the greater good," especially when they seem to be foolishly tilting at windmills.

On the other hand, if comparative advantage and arguments (i) and (ii) above are excluded, we are left with decisions that look much more like difficult, closely-balanced business case analyses: How much new export business will be generated for American businesses? How much will business profits and asset values go up in the favored domestic businesses, and how much will profits and asset values decline in the disfavored domestic businesses? How much will domestic personal income go down as a result of lost jobs? How much will the aggregate cost of imported goods and services go down? Are these projections consistent with what we have actually measured following trade liberalizations of recent decades? Who wins and who loses and what adjustments, if any, will be made? In which time frames will each of these effects occur? Are the gains less certain than the losses? If favored domestic businesses invest more and create more jobs, will the investment and job creation be in America or elsewhere? On balance, deal or no deal?

When we do that kind of hardheaded analysis, we notice that a typical trade liberalization deal confers immediate, tangible, dollar benefits on certain domestic businesses while those domestic businesses and workers who suffer immediate ill effects receive only a promise that comparative advantage, operating like a deus ex machina to rescue the players from a predicament, will bring them bounty later. Once it is widely understood that the future benefits are speculative and will only occur if and when Americans win the competition for them, those who are about to be the immediate losers might well insist on making a better deal for themselves now. The political process might play out differently if it were more obvious that our trade negotiators are picking winners and losers among Americans and that in playing this game of political economy maintaining or raising a trade barrier is as legitimate as lowering one. Nothing personal; its just business.

Comparative advantage is not an idea with any predictive value in the real world but is only a minor hypothetical qualification to Adam Smith's demonstration that business tends to flow to a nation that has the absolute advantage in production costs.

Ricardo's comparative advantage idea is nothing more than a minor exception to Adam Smith's general rule that absolute advantage in cost explains how free international trade will play out. Smith asserted that liberalizing trade would cause each type of good to be made in the nation with an absolute cost advantage for that particular good. Ricardo said that is not necessarily always true and described a case where it is possible that balanced two-way international trade can occur between nations even where one has an absolute advantage in the production of every good.

Unfortunately, it is common for contemporary free trade advocates to ignore the limiting conditions on Ricardo's idea and to argue that comparative advantage insures there will always be balanced two-way trade and that both nations will inevitably benefit even when one nation has an absolute advantage with respect to every good. This is a blunder because in fact the conditions assumed in the idea of comparative advantage rarely if ever occur in the real world. Nobody has ever identified important real-world situations in which a nation having a substantial absolute cost disadvantage has actually increased its exports to an advantaged nation through the effects of comparative advantage. Nevertheless, comparative advantage advocates ignore or deny Adam Smith's rule of absolute advantage, which we can see operating everyday with our very own eyes and in the econometrics and trade statistics, and they distort Ricardo's hypothetical exception into a "theory" that purportedly predicts what will happen in every case, even though so far it has happened in none. (I do not call comparative advantage a "theory" because that is a word scientists use for a statement that attempts to explain a set of reproducible observations of real world phenomena and which can then be tested by other real world observations.)

Comparative advantage is only relevant where there is full employment within both trading nations and no international mobility of capital or labor.

Paul Davidson, editor of the Journal of Post Keynesian Economics, recently explained the conditions that limit the scope of comparative advantage and how they are wrongly ignored by those who rely on comparative advantage as a reason to liberalize trade:

This law of comparative advantage is the basis of the conventional wisdom that free trade globalization is a 'win-win' game – where IF TRADE IS FREE, then there will be enough ADDITIONAL goods and services produced that even if the winners compensated the losers (losers of jobs, of profits for import competing firms, etc.), there would be more goods and services for everybody.

This win-win conclusion of this LAW OF COMPARATIVE ADVANTAGE requires two basic assumptions: (1) there is full employment in all the trading nations before free trade and there is full employment in all nations after free trade is established, AND (2) there is no international mobility of either capital or labor across national boundaries. If either of these assumptions are [sic] not applicable to the existing situation – and in our world both assumptions are false – then logically it can NOT be demonstrated that free trade is a 'win-win' situation. If capital is internationally mobile and full employment is not universal in all trading nations, then trade patterns based on free trade globalization circumstances will be based on the theory of ABSOLUTE ADVANTAGE – where capital will place jobs in the lowest cost labor nations – with considerable real economic losses possible for those nations whose labor costs exceed those in the low cost nations. In other words – until and unless the real income of workers in industries of developed nations falls to the level of Chinese and Indian unskilled and semi-skilled workers, and social safety nets in the US, Europe, etc are reduced to levels of comparable social safety nets with those cheap labor nations, free trade will be a no win situation for workers in developed natinos [sic] whose jobs can be outsourced. Only semi-skilled workers whose jobs can not be outsourced, e.g., waiters, plumbers, electricans, [sic] personal trainers, etc. will see gains due to cheaper imports from free trade." posted January 27, 2006. (Emphasis in original.) [UPDATE 1/4/2001: This link to might work.]

In his online textbook on international economics, Steven Suranovic concludes a lengthy comparative advantage overview chapter this way:

"The usual way of stating the Ricardian model results is to say that countries will specialize in their comparative advantage good and trade them to the other country such that everyone in both countries benefit [sic]. Stated this way it is easy to imagine how it would not hold true in the complex real world.

"A better way to state the results is as follows. The Ricardian model shows that if we want to maximize total output in the world then,

"first, fully employ all resources worldwide;

"second, allocate those resources within countries to each country's comparative advantage industries;

"and third, allow the countries to trade freely thereafter.

"In this way we might raise the well being of all individuals despite differences in relative productivities. In this description, we do not predict that a result will carry over to the complex real world. Instead we carry the logic of comparative advantage to the real world and ask how things would have to look to achieve a certain result (maximum output and benefits). In the end we should not say that the model of comparative advantage tells us anything about what will happen when two countries begin to trade, instead we should say that the theory tells us some things that can happen." (Emphasis in original.)

Special attention should be paid to the relationship between comparative advantage and the level of employment. Proponents often suggest or imply that comparative advantage guarantees there will be full employment, or at least no decline in overall employment. To the contrary, unless there is already full employment (and the other conditions also exist) comparative advantage does not predict any economic effect at all (although the theory of absolute advantage certainly does). In America, where we have never had full employment, the unemployed may find work, as the unemployed mostly have in the past, but they can be sure that comparative advantage had nothing to do with it. And policy makers concerned with creating domestic jobs for today's unemployed and tomorrow's new workers need to look elsewhere for solutions.

Still not convinced?

If you still believe a theory or force of comparative advantage operates in the real world and will secure free trade benefits to America, here are some additional points to consider.

Why is it that comparative advantage advocates always use hypothetical examples to support their idea? Ricardo used a hypothetical example about wine and cloth, and a variety of hypothetical examples have been advanced by others. Since Ricardo's time, have there not been at least a few cases where comparative advantage actually worked and, if so, why does nobody base the argument on observed facts about the real world? If it hasn't happened yet or we can't measure it, why have faith in it? Isn't comparative advantage like a unicorn – a beneficent creature prominent in the literature of political economy that has never actually been captured or photographed by mere mortals?

If comparative advantage is working for third world countries, why do international non-profit organizations concerned with eliminating poverty and fostering economic development in undeveloped regions of the world, so uniformly decry the actual results of trade liberalization?

If comparative advantage is working for America, why are aggregate personal incomes and wealth accumulation expanding greatly in China and India and less so or not at all in America? Is that not a relative decline in America's economic position in the world?

American companies like Boeing and Microsoft appear to have worldwide absolute and comparative advantages and to be able to increase their exports after trade liberalization. In fact, however, Boeing agreed to manufacture important parts of its airplanes in China as a condition for getting airplane orders from the Chinese government, and Microsoft opened a large technology development center in China and employed Chinese there to invent the next new thing. Whether you think those moves and others like them are good or bad for America, can we not agree that they are not predicted by or the result of comparative advantage? Is it not true that comparative advantage, if it works at all, is overwhelmed by other forces such as the increased importance and mobility of capital and technology and the governmental powers of our trading partners who subordinate market considerations to long-term strategic goals?

If comparative advantage is working for America, why have our imports increased dramatically more than our exports of goods and services? Is it not true that nations with which we have large current account deficits, such as China and Japan, have bought American debt and equity securities and American real estate, businesses and other assets instead of American goods and services? Is it not bad for America that instead of being able to balance our imports with exports of goods and services we are exporting ownership of our patrimony and have become the world's largest debtor nation? Could this be happening if comparative advantage were working as advertised?

Have you considered that you might be wrong about the efficacy of comparative advantage in particular or trade liberalization in general? Are you so certain America will be better off that you do not even need to think about what might happen if good results do not materialize when and in the amount you expect? (Incidentally, have you even made estimates of when and how much?) If you were to consider free trade and globalization without reference to comparative advantage, would you perhaps begin to give credence to observations like these?

(a) Every trade liberalization deal should be approached like a business deal in that some deals are profitable and some turn out to be the opposite. A particular transaction may or may not be "win-win."

(b) America should not willy-nilly grab every available trade liberalization deal but should bargain with knowledge, skill, leverage, and determination to try to make only those deals that are demonstrably good for America.

(c) Trade liberalization tends to raise wages and prices in low-cost nations (like China) and to reduce them in high-cost nations (like America,) and depending on how rapid and extensive the increased trade is, there can be a problematic amount and duration of inflation in one nation and/or deflation in the other; these effects should be accounted for as costs in the business case analysis or mitigated in the terms of the deal.

(d) If the gross domestic product and wealth of China, India, and other nations with which trade is liberalized rise faster than America's, then either we have not bargained for our "fair share" or our government should acknowledge that it made a bad deal or decided to confer a gift on another nation and explain why.

(e) Trade liberalization tends to increase the share of domestic income and wealth that accrues to capital and decreases the share captured by labor (while recent tax law changes and proposals decrease taxes on capital and proportionately increase the burden on labor). Capital and the owners of capital are mobile and can leave America more easily than workers. For example, as the Dutch economy collapsed in the 18th Century, Dutch money decamped to England, and Holland lost its position in the world. K. Phillips, Wealth and Democracy – A Political History of the American Rich at 183 (Broadway Books 2002). International arrangements that purport to be only about liberalizing trade also tend to facilitate other aspects of "globalization." In this larger sense, trade liberalization is also about capital flight from America and about the evolution of large and powerful stateless institutions operating like cargo ships under flags of convenience.

Is our faith in the unicorn so complete that we will risk all that and more?

Update on Monday, August 31, 2009 at 11:16AM by Registered CommenterSkeptic

The Wikipedia article on comparative advantage does a pretty good job of laying out the assumptions upon which the proof of the comparative advantage concept rests and criticisms that those assumptions are seldom if ever true in the modern world. 

Update on Tuesday, September 15, 2009 at 01:54PM by Registered CommenterSkeptic

I've been noticing a lot referrals to this page from search engines.  When I Googled "comparative advantage" today, I found this page is ranked #11 out of 25.5 million pages! 

I looked again at Professor Suranovic's online textbook, which is pretty good if you keep in mind the wise precaution of my college economics professor, Alvin Tostlebe, about the Paul Samuelson (4th edition) introduction to trade theory: In the real world it will almost always be absolute advantage that matters, but if you want to sound like an economist you have to talk about comparative advantage.  Certain incantations must be learned before admission to any priesthood.

BTW, I'm still waiting for an example of where comparative advantage works as advertised in the real world, where resources are never fully employed. Another thing I'd like to see is an economic model of what "will happen" or "can happen" if trade is liberalized in an environment where resources are not fully employed and one trading partner allocates internal resources to "inappropriate" industries for strategic reasons or to serve oligarchic interests.  Since that's what we are actually doing most of the time, that model might generate some useful insights instead of misleading ones. 

Update on Tuesday, September 15, 2009 at 05:39PM by Registered CommenterSkeptic

Rummaging through old posts, I found this one pointing out that Ricardian trade theory also assumes balanced trade, whereas the US has had a trade deficit for 27 years, averaging almost 3% of GDP since 1981.  Link.  Also in the first linked post are links to Paul Krugman statements and papers affirming the balanced trade assumption. 

Update on Tuesday, September 29, 2009 at 09:30PM by Registered CommenterSkeptic

Ralph Gomory agrees that the reality is much different from the theory:

Manufacturing should not be given up but rather rebuilt, as G.E. CEO Jeffrey Immelt has recently advocated. We cannot afford to get out of manufacturing unless and until there are new things that we are good at and that add up to the same scale. But today that condition is nowhere near being met.

Update on Friday, October 23, 2009 at 10:48PM by Registered CommenterSkeptic

Here's a counterargument from Paul Krugman at a time when he could refer to a 1995 publication as "recent." I say again that the idea is clearly correct and clearly not working in the real world because the required conditions do not exist and countervailing forces are also at work.

Update on Friday, October 23, 2009 at 11:09PM by Registered CommenterSkeptic

It turns out I was not the first to use the unicorn metaphor with free trade.  Pat Buchanan did it in a 1998 speech: 


Address to the Chicago Council on Foreign Relations

November 18, 1998

To [the] new corporate elite, putting America first betrays a lack of loyalty to the company. Some among our political elite share this view. Here is Strobe Talbott, Clinton's roommate at Oxford and architect of his Russian policy: "All countries," said Talbott in 1991, "are basically social arrangements...No matter how permanent and even sacred they may seem at any one time, in fact they are all artificial and temporary...within the next hundred years...nationhood as we know it will be obsolete; all states will recognize a single, global authority".... This is the transnational elite, our new Masters of the Universe.

This is a prestigious forum; and I appreciate the opportunity to address it. As my subject, I have chosen what I believe is the coming and irrepressible conflict between the claims of a new American nationalism and the commands of the Global Economy.

As you may have heard in my last campaign, I am called by many names. "Protectionist" is one of the nicer ones; but it is inexact. I am an economic nationalist. To me, the country comes before the economy; and the economy exists for the people. I believe in free markets, but I do not worship them. In the proper hierarchy of things, it is the market that must be harnessed to work for man - and not the other way around.

As for the Global Economy, like the unicorn, it is a mythical beast that exists only in the imagination. In the real world, there are only national economies -- Japan's that has lost its animal spirits, South Korea's that is deep in recession, China's which is headed for trouble, Brazil's which is falling, Indonesia and Russia's which are in collapse.


Update on Tuesday, December 15, 2009 at 10:47AM by Registered CommenterSkeptic

In this tribute to Paul Samuelson, Paul Krugman quotes Samuelson’s 1964 paper Theoretical Notes on Trade Problems:

“With employment less than full and Net National Product suboptimal, all the debunked mercantilist arguments turn out to be valid.” And he went on to mention the appendix to the latest edition of his Economics, “pointing out the genuine problems for free-trade apologetics raised by overvaluation”. The solution, of course, was to end the overvaluation rather than restrict trade; Samuelson understood that good macroeconomic policies are a prerequisite for good microeconomic policies.

So, isn’t less-than-full employment and suboptimal NNP pretty much always true?  And isn’t some major country (currently China) pretty much always manipulating exchange rates to its advantage? 

The whole Samuelson paper is here (pay wall) and here [link broken as of 6/26/2011], and it contains this gem on page 145:

Currency Overvaluation.  In 1948 I shocked at least one of my teachers by saying that the theory of comparative advantage does not guarantee a country against balance-of-payments difficulties, nor does it even keep a country from being undersold in terms of every good.  Then it was a question of dollar shortage rather than of American gold loss, and I am not displeased to reread what was said there.

Update on Tuesday, February 2, 2010 at 01:14PM by Registered CommenterSkeptic

Paul Krugman is continuing today with posting of “class notes,” I assume for a course he’s teaching.  In this 14-page PDF he presents typical crossing lines and curves to illustrate several theories about the effects of international trade.  On page 10, he uses an example to show that although there are positive effects from trade, the impact is rather small.  He summarizes with these "two big lessons:"

  1.  Major growth effects from trade policy, if they exist, must come from unconventional channels.  Conventional trade theory DOES NOT justify claims of huge positive payoffs from free trade. 
  2. In the politics of trade policy, distributional effects can easily swamp concerns about efficiency.

(Emphasis in original.)

Update on Thursday, October 7, 2010 at 12:25PM by Registered CommenterSkeptic

John Duffield claims he has devised "a definitive refutation of Comparative Advantage on its own terms." PDF here. Errata and comments are here. I don't understand the details of the argument, but was able to discern that it exploits what is claimed to be an internal inconsistency in the Ricardian argument which makes it fail as a matter of math/logic.  

Update on Monday, December 6, 2010 at 08:27AM by Registered CommenterSkeptic

In response to the Obama Administration's claim it will boost the US economy by opening up foreign trade, Paul Krugman uses standard economic theory to demonstrate that increasing exports doesn't help the domestic economy and can hurt it if they are matched by increased imports. It's an increase in net exports we need.  PK ends his post with this paragraph:

If you want a trade policy that helps employment, it has to be a policy that induces other countries to run bigger deficits or smaller surpluses. A countervailing duty on Chinese exports would be job-creating; a deal with South Korea, not. If you want the Korea deal, fine; but don’t claim virtues for it that it doesn’t possess.

Update on Tuesday, December 21, 2010 at 09:37PM by Registered CommenterSkeptic

Ian Fletcher discusses here 7 critical assumptions required to make comparative advantage work as claimed, many or all of which assumptions are absent in important current US trading relationships.

Update on Tuesday, July 19, 2011 at 01:43PM by Registered CommenterSkeptic

Although comparative advantage still seems to dominate political discussions, professional economists have developed other theories to explain and/or defend international trade.  Sunanda Sen at the Levy Economics Institute of Bard College surveys the principal thought currents in International Trade Theory and Policy: A Review of the Literature (November 2010).  

Abstract: This paper provides a survey of the literature on trade theory, from the classical example of comparative advantage to the New Trade theories currently used by many advanced countries to direct industrial policy and trade. An account is provided of the neo-classical brand of reciprocal demand and resource endowment theories, along with their usual empirical verifications and logical critiques. A useful supplement is provided in terms of Staffan Linder’s theory of "overlapping demand," which provides an explanation of trade structure in terms of aggregate demand. Attention is drawn to new developments in trade theory, with strategic trade providing inputs to industrial policy. Issues relating to trade, growth, and development are dealt with separately, supplemented by an account of the neo-Marxist versions of trade and underdevelopment.

Update on Saturday, October 1, 2011 at 10:27AM by Registered CommenterSkeptic

Steve Keen points out another flaw in the Ricardian argument for free trade: It ignores the fact that converting from autarchy to free trade destroys the value of capital equipment in both nations. (Wine presses can't be converted to spinning jennies or vice versa.) Keen also refers to the 2001 work of Dani Rodkrik finding that the utopian vision of trade liberalization also misleads developing nations. 

These and many other failings that explain why, when Dani Rodrik took a careful look at the empirical record of trade liberalisation, he found that it had frequently reduced material welfare rather than increasing it. Writing back in 2001, he summarised his findings for Foreign Policy magazine with the statement that:

“Advocates of global economic integration hold out utopian visions of the prosperity that developing countries will reap if they open their borders to commerce and capital. This hollow promise diverts poor nations’ attention and resources from the key domestic innovations needed to spur economic growth.”

Article originally appeared on realitybase (
See website for complete article licensing information.