A Reluctant Purist: Bhagwati on Trade
Review of In Defense of Globalization by Jagdish Bhagwati (Oxford University Press, 2004) and Free Trade Todayby Jagdish Bhagwati (Princeton University Press, 2002).
Neoclassical economists often make matters more complicated than necessary; but, fortunately, the best of them manage to stumble close to the truth. Jagdish Bhagwati is by no means a committed supporter of the free market. For him, restrictions on the right of labor unions to strike are little better than slavery. Further, he began his career as an international trade theorist by devising models in which free trade was not the best policy. Nonetheless, he has worked his way to becoming a committed supporter of free trade; and, in the two books we have here to consider, he refutes a large number of fallacious arguments in favor of trade restrictions.
The prima faciecase for free trade is straightforward: is it not vastly more productive to specialize in one's best products, and then exchange with the specialties of others, than to try to produce everything for oneself? The argument does not differ whether we take individuals or countries as the trading partners.
Further, both partners can benefit from trade even if one of them is better than the other in producing every good exchanged. As David Ricardo argued, it is to the advantage of the nation more efficient in producing all goods to concentrate in the area its edge is greatest. The poorer producer will then specialize in the product at which it is least bad, i.e., where it has a comparative advantage
As our author notes, many people find this elementary truth hard to grasp. "When asked by the mathematician Stanislaw Ulam . . . which proposition in the social sciences was the most counterintuitive yet compelling, Paul Samuelson chose the law of comparative advantage. . . . Most people think it intuitively sound that you should do most things that you do better than others, not specialize" (Free Trade, p. 5).
Austrians find this truth very easy to grasp, and Mises extended Ricardo's principle into a general law of association. But neoclassicals are not so easily satisfied. Although Bhagwati accepts the general purport of Ricardo's results, he tells us that "the analytically satisfactory proofs of free trade that we modern economists demand are the handiwork of theorists working in the twentieth century" (Free Trade, p. 4). 1
Why are Bhagwati and his fellow neoclassicals so demanding? For Austrians, the market as actually created by producers and consumers suffices: not so for neoclassicals. Let the slightest hint of "monopoly," by their definitions, arise, and they are overcome with anguish. The conditions for efficiency set forward in their models are not perfectly fulfilled; we have, then, a case of "market failure" and the benign state must step in to secure an optimal outcome.
Such reasoning did not await the twentieth century. Thus, Robert Torrens opposed Sir Robert Peel's call to end the Corn Laws. Because of British monopoly power in trade of the goods concerned, an optimal tariff would excel in efficiency the results of full free trade. Again, John Stuart Mill used another instance of "market failure" to justify tariffs. Do not infant industries constitute an exception to the generally desirable rule of free trade? (To my surprise, Bhagwati accepts Mill's argument as theoretically sound. I should have thought the analysis by Mises in Human Action[Scholar's Edition, pp.505–07] tells decisively to the contrary.)
Austrians, who reject the neoclassical view of market failure, find it easy to dismiss arguments that presuppose what they reject; but Bhagwati's work has great value for those who prefer to take matters the hard way. 2 He shows that even if one accepts the neoclassical view, free trade is usually the order of the day.
He raises three principal points that radically reduce the cases in which deviations from free trade count as efficient, on the neoclassical view. First, various models may show that free trade is not perfectly efficient; but the question then arises, how much do these deviations matter? Such questions arouse our high-powered neoclassicals to furious calculations; and it transpires that often the efforts to "correct" the market cost more than they are worth. "As it happens, the younger free trade theorists of imperfect competition in product markets. . . came back to free trade precisely on the ground that the gains to be had from pursuing a policy of optimal departures from free trade in industries characterized by imperfect competition were not large enough to justify intervention" (Free Trade, p. 30).
The next point I shall pass over briefly, since readers of this journal will I suspect find it of little interest. Cases in which domestic problems allegedly dictate a departure from free trade are often better handled by a different measure of government interference that preserves free trade. Keynes once suggested that tariffs are desirable in cases of mass unemployment; why not instead leave trade free and increase aggregate demand directly? Concerning this, I can only echo Lincoln: "If you like this sort of thing, this is the sort of thing you'll like."
Bhagwati's last point seems to me much more significant. If tariffs and other trade restrictions are imposed, people trying to divert these gains to themselves may spend immense resources. Such resources make no apparent contribution to production. Bhagwati anxiously forestalls readers inclined to see this as a case of rent seeking: what we have here, he assures us, is directly unproductive profit seeking, a concept of his own devising that is entirely different. Bhagwati is not afflicted by the vice of false modesty; and, I regret to say, he is always alert to stress his own importance, though in an amiable way. Those who do not share our author's investment in his own reputation may find the two ideas hard to distinguish.
Be that as it may, the important issue is this. Such rent seeking, or unproductive profit seeking, may be so extensive as to rule out tariffs as a measure of efficient economic policy. Anne Krueger estimated that these expenditures for Turkey were "of the order of 40 percent of GNP" (Free Trade, p. 40).
If Bhagwati is right, neoclassicals can in good conscience embrace free trade. Well and good, but what about the rest of us? Why should those not beguiled by neoclassical siren songs be interested in Bhagwati's books? Our author has a ready answer. He analyzes a large number of objections to free trade, and his responses to these, and his various illuminating remarks on related subjects, usually do not depend on controversial neoclassical assumptions.
A common objection to free trade rests on an undeniable fact. In the free market, every exchange takes place because all the parties involved expect to benefit; but it by no means follows that everything that happens in the market benefits everybody. If lower prices by the restaurant across the street put your establishment out of business, you have certainly not gained.
If so, are not proofs of the market's efficiency of limited value? Will it help American workers sent into unemployment or reduced to penury by low wage competition from abroad to be told that the overall efficiency of the world's economy has risen? The Stolper-Samuelson theorem offers a proof that under certain conditions, the low wages of foreign workers can indeed drive down domestic wages.
Bhagwati maintains that this line of thought fails to fit reality, so far as America is concerned. The effect in question requires that the goods that undercut high American wages be labor intensive. Foreign competition that drives American firms out of business for other reasons, such as technological superiority, is not here in point. But imported labor-intensive goods will lower domestic wages only if their relative prices fall; otherwise, there is no increased pressure on the wages of those who produce competing American goods.
Thus the key issue is whether the prices of such goods have actually been falling, triggering the declining real wages. Here the evidence does not support the assertion. During the 1980s, when the real wages of American workers were stagnant, the prices of the labor-intensive goods as a group actually rose relative to the prices of the set of all goods in world trade. (Globalization, p. 124)
Sometimes opponents of free trade raise a different sort of objection. Here what concerns them is not the plight of workers in rich countries. Quite the contrary, it is alleged that free trade is part of a strategy by predatory multinational corporations to exploit the Third World. Do not these companies make enormous profits by paying a pittance to the impoverished denizens of the Third World?
Bhagwati replies with a point that is no less effective because it is obvious. If workers accept what we regard as extremely low wages, they must regard these wages as improvements. "[S]everal empirical studies do find that multinationals pay what economists now call a 'wage premium'; they pay an average wage that exceeds the going rate, mostly up to 10 percent and exceeding it in some cases, with affiliates of U.S. multinationals sometimes paying a premium that ranges from 40 to 100 percent" (Globalization, p. 172).
Those who find convincing the case for free trade must avoid a pitfall that our alert author, like Lew Rockwell in Speaking of Liberty, is careful to emphasize. The advantages of free trade stem from the genuine article, not counterfeits that claim the name of freedom. In particular, so-called free-trade areas are not at all what supporters of free enterprise should want. As "the great economist Jacob Viner pointed out . . . free trade areas (FTAs) are not free trade. While they remove tariffs for member countries, they also increase the handicap (for any given external tariff) that nonmembers suffer vis-à-vis member-country producers in the markets of the member countries, implying therefore protection against them" (Free Trade, p. 107). Does not this consideration vindicate free trade opponents of Nafta?
- 1. Bhagwati is a trade theorist of international standing; but he never mentions the argument, claimed by Paul Craig Roberts to be at the heart of international trade theory, that Ricardo's law shows that free trade is advantageous only if capital goods are immobile. Small wonder, as Roberts's claim is false. The law is indeed dependent on the restriction Roberts mentions, but it hardly follows that free trade is advantageous only if one trading partner has an absolute advantage in all goods traded, the condition in which Ricardo's law applies. If factors of production move to where they have an absolute advantage, will not mutually advantageous trade take place in accord with the new conditions?
- 2. In addition to the standard Austrian literature, Joseph Schumpeter, <em>Capitalism, Socialism, and Democracy</em> (New York, 1942) gives an immensely valuable criticism of the standard analysis of monopoly and competition.