Cap'n Arbyte's

A Direct Reply to Roberts on Free Trade

Paul Craig Roberts has published a response to the comments of Reisman and Salerno, who in turn were commenting on the Schumer/Roberts NYT Op-Ed about free trade. (I've commented on Roberts's views a couple times myself.)

Roberts's central argument is that the classic case for free trade, based on Ricardian comparative advantage, is decreasingly applicable due to the increasing mobility of factors of production in a knowledge-based economy. From the Op-Ed:

Comparative advantage is undermined if the factors of production can relocate to wherever they are most productive: in today's case, to a relatively few countries with abundant cheap labor. In this situation, there are no longer shared gains — some countries win and others lose.

I am a microprocessor validation engineer, a high-tech, knowledge-based job. I also work in the United States. I am a member of one of the most "at-risk" groups of workers. There are many people in foreign countries already performing the sort of work I do, and I have personally met several of them. I'm on the front lines of globalization and have a first-hand view of its actual operation. I am also, unlikely as it may seem from my circumstances, an unabashed supporter of globalization.

In his latest response, Roberts ends by stating:

The proper way to answer the argument that Schumer and I have made is to make a case that free trade is mutually advantageous in the absence of comparative advantage. Alternatively, make a convincing case that comparative advantage does not require at least some factors of production to be immobile.

This is an impossible but also irrelevant challenge. Comparative advantage is the raison d'être for production intended for trade. Without comparative advantage there would be no production intended for trade, and no trade. The entities involved, be they individuals or nations, would produce and consume independently of each other.

A world without comparative advantage is a fantasy. Even if all physical capital were fully and costlessly mobile, natural and unavoidable differences between people create comparative advantage. The young have a comparative advantage over the old in the performance of physical tasks due to their physical vigor. Adults have a comparative advantage over children in the performance of intellectual tasks due to their greater knowledge and experience. Even among adults, specialization along lines of study must occur because it is impossible for any single individual to learn the entire corpus of human knowledge. Specialization creates comparative advantage and therefore an incentive to trade.

The inevitability of specialization allows us to imagine two groups of people, each specializing in a different field, who will therefore trade with each other. It makes no difference if each group's members reside in a different nation, or if everyone resides in the same nation, or if the people are evenly mixed between nations. In the first case a quantity called "international trade" will exist, in the others it will not.

The international mobility of knowledge jobs effectively joins together what had been separate labor markets into a larger world labor market. This increases opportunities for specialization and enhances the division of labor. Whether this results in greater international trade or not depends on the geographic distribution of the new specialists, but it definitely increases real wages regardless of its impact on international trade. Protectionist policies would keep labor markets separate and prevent these opportunities for specialization.


Roberts argues, correctly, that the increasing mobility of factors of production (jobs in this case) make it easier for them to leave the United States and move into nations that have an absolute advantage due to cheaper labor costs. He attributes the availability of cheap labor to the "collapse of world socialism", and this socialist legacy is the reason why he attaches importance to political borders that would otherwise be irrelevant from an economic point of view.

As factors of production move to where production has an absolute advantage, comparative advantage is lessened, and trade is correspondingly diminished, Considered narrowly, Roberts is correct to say that the absolute advantage nation gains while the other nation loses. The gains from trade are reduced as the volume of trade is reduced.

In the present controversy over jobs, the amount of emigration and physical capital moving to other nations is negligible. The physical productive capacity of the United States is undiminished, so the total losses to the United States are limited to the losses due to the reduced volume of trade. Displaced domestic workers must find new jobs, but their wages will be reduced. The reduction in trade could be permanent.

To the disaffected people in that situation, solace can only be found by recognizing that while not every person gains through every act of economic competition, the risk of occasional competitive loss is dramatically outweighed by the certainty of continuous gain in all those areas where the person is simply a consumer. Economic competition must be accepted or rejected as a principle. The tremendous benefits of economic competition exist precisely because it favors the more productive over the less productive. The more productive become wealthy, and the less productive are incentivized to improve.

It is senseless to advocate economic competition except when one's job or one's neighbors' jobs are at stake. The system only works because those jobs are at stake — remove the incentives of the system and you destroy the benefits of the system. (I wrote a more complete discussion of why even the disaffected should support economic competition in my pro-globalization essay.)

Is it better to scrap economic competition when it becomes uncomfortable, or to keep it and adapt to the new business conditions? If Roberts would like to propose an alternative to or a specific exception from voluntary economic competition, I would entertain it, but it must be forthright about the hazards of disrupting economic competition. It must also be clear about its exact scope, i.e., why should it apply only internationally but not domestically?


I suspect this belief is responsible for Roberts's race-to-the-bottom attitude about American wages:

The huge excess supplies of labor in countries such as China and India ensure that it will be many years before labor in those countries, both skilled and unskilled, will be paid the value of its marginal product.

There is no such thing as an excess supply of labor. Human desires are unlimited. There is always more work to be done. A world with no further need for labor would be an utter paradise with no possible improvement. I would very much like to live in such a fantastic world, but it bears no resemblance to the world that exists today.

A large and inexpensive labor force would not simply replace the American work force and leave them unemployed. Rather, both groups of people would continue to work and would engage in many more projects than the American work force was able to on its own. A change in trading patterns would cause different people to be affected in different ways, but increased specialization will bring about higher real wages as the overall result.

To adopt Reisman's formulation [Capitalism p. 618], real wages are determined primarily by the productivity of labor. Low present wages in China and India are a result of their low productivity, but will rise rapidly in response to increasing productivity. If Roberts disagrees with the productivity theory of wages, he should clearly say so, and the overall argument should be recast in those terms.

(I acknowledge that this may be dismissed as hearsay, but I've been told confidentially by a source with first-hand knowledge of wage rates in nations gaining jobs through offshoring that wages in those nations are, in fact, rising rapidly — at least among those segments of the population who have the skills we want to employ. This is anecdotal confirmation of the productivity theory of wages.)


It is important to acknowledge that the economic advantages of foreign low-cost labor are real and that the competitive process cannot be stopped. Even an outright ban on offshoring would not prevent foreigners from obtaining knowledge-worker jobs — it would merely prevent them from being hired by American companies. Companies from other nations, perhaps even from the low-cost nations themselves, would still eagerly take advantage of low-cost labor and thereby gain a competitive advantage over American companies frozen out of the process by protectionist policies.

This is a familiar pattern in economics: Competitors must always strive to match the cost reductions of their peers, or they will be driven out of business. Protectionism will not prevent those precious American knowledge-worker jobs from being lost. In fact, more jobs would be lost, because entire companies would be made uncompetitive instead of just a few workers!

The alternative is not: High-paying American jobs versus the loss of those jobs to foreign workers. A shift in international job distribution is unavoidable, it cannot be prevented by political fiat. The alternative is: Should American companies be free to profit from cost reduction opportunities, or should the American government intervene on the behalf of foreign competitors by prohibiting those cost reductions?


I suggest, with some hesitation, the following as a broad outline of the changes in the world economy today. It is easiest to think about this in terms of an invariable money shared by all nations.

There is a tendency for each nation to have a proportion of the world money supply equal to its proportion of world economic output. As nations like India and China grow more quickly than nations like the United States, the former will gain money from the latter, causing a general increase in prices and nominal wages in the former and a decrease in the latter.

Imports and exports (broadly defined, including securities) will not balance until each nation's proportion of world money supply matches its proportion of world economic output. This flow of money tends to equalize prices and nominal wages across the world.

Real wages in the nation losing money do not decrease because its productive capacity is unchanged. Both prices and nominal wages fall there in tandem. Real wages actually increase for the duration of the adjustment period because a permanent loss of money means that some imports do not need to be balanced by exports. This additional wealth should be judged against the cost of restructuring to conform to the new conditions of trade.

Seen in this way, falling prices and nominal wages in the United States and rising prices and nominal wages in places like India and China are a natural effect of the former's lower relative growth rate. These price changes are an inevitable part of the motion toward a new global equilibrium. They do not have a sinister character.

Integrating this outline with the facts of fiat currencies and currency pegs and central bank interventions is a task I leave to someone else.

Tiny Island