Free trade theory fails to correspond to reality
from Jeff Ferry
For the last 90 years, the United States has pursued and advocated free trade. For the last 60 of those 90 years, American workers and other observers have watched America lose high-paying jobs to imports and asked: can this really be good for the American economy?
Professional economists have answered, virtually unanimously, that yes, it is good, due to something called the Law of Comparative Advantage.
They are wrong. Their free trade theory, based on the so-called Law of Comparative Advantage, does not work for the U.S. or for many other countries. We know this because dozens of economists have published studies of the empirical results of import penetration showing that the Law of Comparative Advantage, and the modern economic theory built around it is outmoded and inapplicable to high wage nations like the U.S. Indeed, it can actually worsen the performance of high wage nations.
Economists advocate free trade theory less because they actually believe it than because of what Nobel laureate economist Paul Romer has called “a sense of academic group identity grounded in a common defense of [a] dogmatic position.”[1] In other words, economists use this dogmatic theory as a weapon to win jobs, influence, and consulting contracts.
In fact, free trade theory fails to correspond to reality, as the evidence published by economists for at least 100 years has shown.
This is not an argument that free trade is insufficiently compassionate, or that it creates short-term problems. Rather, it is an argument that the theory itself is wrong because it is outdated and fails to recognize important features of modern high-wage economies. I should add that I consider myself a conventional economist. I consider the two greatest economists of the 20th century to be John Maynard Keynes and Paul Samuelson. I believe if they were here today, they would agree with what follows.
First, a quick summary of what we mean by free trade. As first explained in 1817 by David Ricardo in his foundational text, Principles of Political Economy and Taxation[2], a free trade event, such as the abolition of tariffs between two countries, should make all workers and capitalists better off in both countries as workers and companies take advantage of the cheaper imports to move to industries where they can be more productive. In modern economics, this was generalized and mathematicised to say much the same thing: each worker will increase her “marginal product” and wages by moving into higher-productivity industries as imports provide lower-productivity goods and services.
This view of an economy was reasonably accurate in David Ricardo’s time, because workers across Europe were paid at close to subsistence wages, with specialized craft workers earning slightly more.
But this wage structure has not been true since the rise of the Industrial Revolution in the late 1800s. It is even more inaccurate today, with devastating consequences for free trade theory.
The assumption that wages are independent of a worker’s industry and depend only on something called “marginal product” (which is in turn often proxied by years of experience) is a little-appreciated but critical assumption on which all free trade economics rests. The theory asserts, and requires, that when workers change industries their wages remain the same, or rise slightly because they are supposedly moving to a more productive industry.
There are reams of evidence disproving this assumption. . . . read more
Leave a Reply
—– look inside —– $5.94 / $20.00
—– look inside —– $4.90 / $8.00
—– look inside —– $15.99
—– look inside —– $5.99 / 12.99
—– look inside —– $5.93 / $12.99
—– look inside —– $4.97 / $9.90
—— Ugarteche, Puyana and Madi ——
Gerson Lima / Maria Alejandra Madi
Edward Fullbrook and Jamie Morgan
————— Michael Hudson ————–
Maria Alejandra Madi / Jack Reardon
————- Edward Fullbrook ————-
—————— Steve Keen —————–
————— Richard Smith —————
————– Gustavo Marques————
– Victor Beker and Beniamino Moro –
————– Lars Pålsson Syll ————-
—————– Stuart Birks —————-
Edward Fullbrook and Jamie Morgan
———— Armando Ochangco ———-
Shimshon Bichler / Jonathan Nitzan
————— Mauro Gallegati ————–
————— Herman Daly —————-
————— Asad Zaman —————
—————– C. T. Kurien —————
————— Robert Locke —————-
Real-World Economics Review Blog 
After reading Mr. Ferry’s paper titled “Free Trade Theory and Reality: How Economists Have Ignored Their Own Evidence for 100 Years”, I would like to share a novel development in the field of international trade theory that he and his readers might not be aware of.
Let’s start with a piece of presumably good news for Mr. Ferry: the law of comparative advantage has been finally debunked. As a recently published paper (1) in the Journal of the History of Economic Thought explained, this alleged law is based on a fundamental misinterpretation of David Ricardo’s famous numerical example in the Principles.
But unfortunately for Mr. Ferry, this refutation of the law of comparative advantage also implies that the old critique about the unrealistic assumptions of this law does not apply to Ricardo’s case for free trade. Neither do the other arguments against free trade he mentioned in the article.
Let’s take, for example, Mr. Ferry’s claim that Ricardo’s case for free trade does not apply to high-wage countries. Ricardo was an English classical political economist. During his lifetime, wages in England were, of course, higher than in the neighboring countries. Why did he support free trade anyway? Because it was nothing other than Ricardo who refuted the old presumption that high wages lead to high prices. He showed that the distribution of the product between laborers and capitalists does not influence prices.
Ricardo neither assumed that workers were paid according to their contribution to the output of their employer’s business. Like other classical political economists, Ricardo rightly believed that wages – i.e., the market price of labor power– were determined by the supply and demand of labor power, not the workers’ individual contributions to their employer’s business. That is precisely why wages often decline after mass layoffs: in that case, the supply of labor power exceeds the demand.
I am further surprised that Mr. Ferry suggests that Ricardo believed that the abolition of tariffs would make all workers and capitalists better off in both countries. In truth, Ricardo wrote a whole chapter in the Principles about the negative consequences of a sudden influx of cheaper foreign goods on specific industries (chapter XIX: On Sudden Changes in the Channels of Trade).
Contrary to what Mr. Ferry believes (p. 86), the impact of imports is very similar to technological change. Both replace jobs in specific sectors in the home country with jobs in another sector of the home country (remember, “all imports must ultimately be paid for with exports” p. 86). Likewise, both allow countries to obtain a certain amount of commodities (or a similar service in the case of advertising) with less quantity of labor. That is precisely why free trade and technological change lead to cheaper products, whereas protectionism and technological stagnation lead to dearer products and services.
I hope that Mr. Ferry will take note of the new development in the field of international trade theory and try to present new arguments against free trade, as the ones in his article no longer apply to the classical case for free trade.
(1) See Morales Meoqui, Jorge. 2021. Overcoming Absolute and Comparative Advantage: A Reappraisal of the Relative Cheapness of Foreign Commodities as the Basis of International Trade. Journal of the History of Economic Thought, 43 (3): 433–449. doi:10.1017/S1053837220000401.
Hmm. From what I recall, the theory also requires that there is no unemployment in either of the economies being compared, that there are no international transfers of investment capital, … Oops.
https://www.bis.org/publ/work890.htm “How does international capital flow?”
《Gross capital flows play a central role in today’s policy debates. Yet current theory largely relies on net flow models of saving and current accounts. This limits the scope of policy advice.》
Geoff,
you are right. All four generations of mainstream trade theories assume that all countries employ their labor force at maximum. As long as they employ general (or partial) equilibrium framework, it will be difficult to make a theory that admits unemployment.
Keynes, in his General Theory, used an equilibrium framework. Its logical coherence was doubtful. This ambiguity was a reason of quick success of his doctrine, but became the deep cause of counter-Keynesian revolution in 1970’s. As I have noted in my post below (December 4, 2022 at 8:39 am. Please see the next post as well, because I made a mistake to use < and > in my mathematical expressions), trade theories or international microeconomics were insensitive to Keynesian revolution.
As for international transfer of capital, we should distinguish two aspects: indirect (or financial) investment without control of the business and direct investment that control the business (production, procurement, and marketing). Indirect investment may change the purchasing power of two countries, i.e. between investing country and receiving country. However, in the case of indirect investment, there is no change of production techniques and value relations (wage rates and prices). Hecksher-Ohlin(-Samuelson) theory claims that the change of capital-labor ratio changes factor prices (i.e. wage and profit rates), but it is a wrong theory based on neoclassical economics idea. Moreover, empirical researches in 1990’s revealed that its generalized theory called Heckscher-Ohlin-Vanek theory performed very badly.
Direct investment is quite different, because the investment accompanies improvement of not only management but of production techniques. We should examine those effects in order to see what happens after the investment.
yoshinorishiozawa,
How does your model incorporate something like Musk’s recent Twitter buyout? Was that direct investment funded by indirect investment, much of which flowed from international sources (Dubai, Saudi Arabia, Japan, UK, France, etc.)?
Thus, can indirect investment pretty easily affect prices? Was Musk able to charge for blue checkmarks, due to how much indirect foreign capital funding of his direct domestic investment?
Dear rms,
my theory is not almighty. It is concerned only with productions and trade of goods and services. I do not know the deep reason of Musk’s buyout. If it is motivated by a political reason, such as excluding people like Dean Baker from Twitter, it is impossible to explain such an act from economics.
A direct investment is restrained by two conditions: (1) a prospect of raising enough profit from the investment, and (2) the possibility to fund the investment. As long as the real economy (say production, transaction, and consumption) is not affected from money funds, the prices of goods and services are determined mainly by technological conditions (plus the conditions of competition). The prices remain unaffected from the amount of money for example.
A possible exceptions would be the change of effective demand. If the affluent money fund is poured into the housing, then the aggregate demand increases and it may bring a tight labour market and the wages may be pulled up. If a series of these chains is realized, it may cause inflation.
I can make this kind of suggestions, but this is out of the proper range of my theory.
With regards to the interpretation of David Ricardo, I support Morales Meoqui. What is interpreted as Ricardo’s theory of trade is in reality an interpretation from the neoclassical viewpoint. Let me give another paper that gives a new view of David Ricardo:
Gilbert Faccarello (2017) A calm investigation into Mr Ricardo’s principles of international trade. Chapter 6 (pp.85-119) in Senga, Fujimoto and Tabuchi (eds.) Ricardo and International Trade. Routledge, London and New York.
Jeff Ferry’s complaint on international trade theory is almost right. The history of trade policy of the USA is much more complex to claim that it was a promoter of free trade system since its foundation. Alexander Hamilton was symbolic with this regard. He claimed that American industrial development would be impossible without protective tariffs. For these 80 years, when it was the industrial strongest, most of the American economists continued to support free trade policy but it is doubtful if this is a correct conclusion. The policy
is a kind of conclusions that are logically deduced by the structure of their theories. They (roughly speaking there are four generations) were astonishingly restricted and imperfect and contained many implicit assumptions. They were no good tools of analyses. For example, practically all trade theories (including those sympathetic to New Keynesians like Paul Krugman) assumed full employment of all countries. Some economists commented that trade theorists were insensitive to the unemployment problem even after the Keynesian revolution. This is true even today. Four generations of (mainstream) trade theories exclude unemployment by assumption. I know only a book in trade theory that includes unemployment in its title.
Ferry seems, however, short-sighted, because there is now a trade theory that does not assume full employment. See my paper:
The new theory of international values: an overview, which was published as a chapter of a book A New construction of Ricardian Theory of International Values.
A more dynamic version of the theory is sketched in
“A new framework for analyzing technological change,” Journal of Evolutionary Economics 30(4): 989-1034, October 2020.
Although the new theory is far from perfect, it is a theory that holds in a sufficiently complex situation: many-country, many-good economy with input trade and choice of production techniques. A new definition of regular values is found without relying on the world production frontier (the set of points where all countries enjoy full employment).
The most important point in the international competition is the dynamic capability (Teece and others) of nations. Competition in a globalized economy is the race of labor productivity with wage differential as handicaps. (See Shiozawa and Fujimoto 2018 The nature of international competition among firms).
If the labor productivity of production techniques in an industry K of country A (assumed to be advanced country) remain stagnant where as that of country B (say a developing country) is rapidly progressing, industry K of country A loses its competitiveness vis-à-vis the same industry of country B. A simple criterion is which of production costs
uA wA + and uB wB +
gives the cheaper cost. Here I assumed that uA and uB are labor input coefficients (the inverse of labor productivity), wA and wB wage rates, aA and aB (vectors of) input coefficients in industry K, respectively for countries A and B, and finally p is the prices of goods that are equal for all countries (Here I implicitly assume that here is no transport and transaction costs). In other words, uA wA and uB wB are labor costs and and are material input costs for the industry K in two countries.
If as the first approximation we assume = , the criterion is reduced to the comparison between uA wA and uB wB. This implies that industry K of country A must satisfy the condition
uA wA ≤ uB wB,
or equivalently
uA/uB ≤ wB/wA.
When the wage discrepancy wA/wB between A and B is big, this is a very sever condition for industry K of country A. For example, around the year 2000, the ratio wA/wC (wage ratio between the USA and China) was greater than 20. It indicated that country A must have an extremely high productiveness of industry K in order to compete imported goods from country B.
The trouble occurs not only from wage discrepancy wA/wB. The productiveness of developing countries 1/uB increases as well. It is easier to catch up when the discrepancy wB/wA is large when a certain level of technology is attained. Then the competition becomes a race between a turtle and a rabbit. This race occurs in all industries. It is no wonder even if many (or the most) industries of country A loses the race. Then, unless the wage rate wB rises rapidly or the wage rate wA decreases substantially, it is inevitable that may industries lose its competitiveness. Firms go bankrupt and workers discharged. In this situation, there are no good replacement and it is natural that workers have to satisfy with lower wages as Ferry reports.
If once we start to analyze the economy with an appropriate theory, it is not difficult to understand what are happening in the USA and in Japan.
It is a shame that Ferry is not searching a good theory outside of mainstream economics. Even if mainstream economics is everywhere powerful and giving practical advice, the absence of a good theory in the mainstream does not mean there is no hopeful alternative among heterodox economics. International economics is one of such conspicuous examples.
I am sorry that some mathematical forms in two paragraphs after “If the labor productivity of production techniques” did not appear as I intended. I replaced the symbols and I hope the expressions will be more sensible. I reproduce here only two paragraphs:
If the labor productivity of production techniques in an industry K of country A (assumed to be advanced country) remain stagnant where as that of country B (say a developing country) is rapidly progressing, industry K of country A loses its competitiveness vis-à-vis the same industry of country B. A simple criterion is which of production costs
uA wA + (aA, p) and uB wB + (aB, p)
gives the cheaper cost. Here I assumed that uA and uB are labor input coefficients (the inverse of labor productivity), wA and wB wage rates, aA and aB (vectors of) input coefficients in industry K, respectively for countries A and B, and finally p is the prices of goods that are equal for all countries (Here I implicitly assume that here is no transport and transaction costs). In other words, uA wA and uB wB are labor costs and (aA, p) and (aB, p) are material input costs for the industry K in two countries.
If as the first approximation we assume (aA, p) = (aB, p), the criterion is reduced to the comparison between uA wA and uB wB. This implies that industry K of country A must satisfy the condition
uA wA ≤ uB wB,
or equivalently
uA/uB ≤ wB/wA.
When the wage discrepancy wA/wB between A and B is big, this is a very sever condition for industry K of country A. For example, around the year 2000, the ratio wA/wC (wage ratio between the USA and China) was greater than 20. It indicated that country A must have an extremely high productiveness of industry K in order to compete imported goods from country B.