Rethinking Ricardo's Comparative Advantage
No industrial policy and no global "convergence"--just a correlation?
Since the 1980s, the field of economics has considered the principles of free trade as immutable laws, indispensable for economic growth. With certain recent populist exceptions, US economists have elevated free trade theory to a national theology. New York Times readers’ favorite economist Paul Krugman wrote, “if there were an Economist’s Creed, it would surely contain the affirmations [including] ‘I advocate Free Trade.’” Gregory Mankiw, author of the most widely used textbook in economics, explained, “although economists often disagree on questions of policy, they are united in their support of free trade.”
All students of economics learn and re-learn the basic justification for free trade: the notion of comparative advantage. David Ricardo developed the classical theory of comparative advantage in Principles of Political Economy, published in 1819, and Ricardo remains taught in every introductory course on economics. Ricardo showed that a higher level of overall output could be achieved if countries specialize production in activities in which they have the lowest relative “opportunity cost,” and trade with other countries to fulfill their consumption demands. In other words, if Portugal can produce wine more cheaply than England, but England can produce textiles more cheaply than Portugal, then to maximize overall quantities of both goods Portugal should only produce wine and England should only produce textiles. Then, free trade allows the two countries to trade textiles for wine and reach their optimal levels of both goods. Sounds great, right?
According to estimates from the Maddison Project database, Portugal and the UK had relatively similar levels of wealth in the late 17th century (measured using GDP per capita in 2011 prices). But as countries industrialized and their mercantilist tariffs fell, the UK prospered while Portugal stagnated. By 1900, the UK’s GDP per capita was nearly three times Portugal’s. Even today, despite Portugal’s recent economic “catch-up,” there remains a significant wealth gap between two countries. The World Bank estimates that the UK is still nearly twice as rich as Portugal. Ricardo said both countries would grow by specializing—what happened?
What Ricardo missed is that economic development is “activity-specific,” or, in other words, some specializations are more conducive to economic growth than others. While this seems obvious, comparative advantage ignores that specializing in computer chips is better than potato chips, and that specializing in textiles was better than wine in the 18th century. At the time, textiles were the leading sector in the world economy, and their production had close linkages with other sectors. Perhaps most importantly, there was room for innovation in the textile industry—production originally done by individuals was transitioned into large water-powered mills. The increase in productivity generated an economic surplus which the UK invested in other productive industries. I am not certain about the extent to which Portugal actually specialized in wine, but regardless their economic activities had lower returns to scale and lower profits than early UK industries.
The neoclassical view of free trade also obscures relations of power. In traditional models of comparative advantage, trade is assumed to be taking place between equal partners with similar levels of development—political factors are ignored. But this is not a realistic view of global political economy. 20th century British economist Joan Robinson wrote, “Portugal was dependent on British naval support, and it was for this reason that she was obliged to accept conditions of trade which wiped out her production of textiles and inhibited industrial development.” Given power imbalances between trading partners, is free trade truly “free”?
Today, trade liberalization is ubiquitous. World trade increased by five times between 1993 and 2013 as developing countries followed the logic of free trade and integrated into global value chains. In many cases, developing countries entered formal bilateral or regional trade agreements with developed countries, pursuing mutual benefits. These agreements often contained “rules of origin” and “tariff escalation” clauses which restricted developing countries’ ability to do industrial policy. In effect, the trade agreements prevented developing countries from supporting their strategically important industries, forcing them to specialize in their respective comparative advantages. If we follow traditional Ricardian logic, this should be no problem. In fact, the integration of non-western countries and firms should be a major driver of development. As many mainstream economists have argued, reduced developing country protectionism and subsequent greater participation in global value chains should generate higher profits, greater investment, and productivity growth.
But, once again, economic development is activity-specific. And developing countries found they often enjoyed a comparative advantage in low wage, labor-intensive manufacturing. This activity is low “value-added,” meaning that a small portion of the profits earned by the corporation remain in the developing country. High value-added activities like marketing and R&D have historically been performed exclusively by developed countries. Free trade also ensures that labor-intensive manufacturing remains a low value-added activity—countries in this sector are under competitive pressure to minimize costs (and, in effect, value-added) or else corporations will build their factories elsewhere. In this way, free trade does not produce mutual benefits—instead, it keeps developing countries poor while mobilizing a cheap workforce for Western companies.
And the second point—global trade is based on relations of power. The owners of Western capital have pursued trade liberalization because it falls squarely in their self-interest. They want unrestricted access to cheap labor and natural resources in developing countries to keep their production costs low and their products competitive. The IMF and World Bank—institutions created and still dominated by Western countries—have included trade liberalization as a necessary condition of their financial assistance. In particular, the IMF’s structural adjustment programs in the 1980s demanded that their borrowers (developing countries) adopt free market trade regimes. The mission of the WTO is to provide a common framework for trade agreements, which typically include reducing or eliminating tariffs, quotas, and any remaining industrial policy.
Since 1960, very few countries have “developed.” Those that have are primarily in two regions: the European periphery (Portugal is one) and East Asia. While the World Bank and IMF tried to take credit for the “East Asian tigers,” research suggests that these countries did not follow free market policy prescriptions. Instead, East Asian countries adopted industrial policy, including large-scale state intervention in the market and protections for strategic domestic industries. Similarly, scholars have documented how the now-developed countries once practiced trade and industrial strategies they now consider “bad,” like heavy protectionism. Dr. Chang finds that both the UK and the US depended on tariffs for their industrial development.
Last week, I was at a talk with a senior macroeconomist from the IMF who said something along the lines of “…given the lack of global [wealth] convergence, we are rethinking our bias against industrial policy…” (certainly a misquote, but you get the idea). If even the IMF is reconsidering its stance on industrial policy, why should students of economics still be subject to Ricardo? Economics prides itself on nifty models like comparative advantage. But development is complex—industrial policy is a tool like any other, and one that shouldn’t be disregarded.