The Empirics of Comparative Advantage: Overcoming the Tyranny of Nonrefutability

Abstract

I assess the empirical evidence on comparative advantage. I argue that the Heckscher–Ohlin–Vanek (HOV) relationship is not a refutable general-equilibrium proposition. Consequently, the empirical Heckscher–Ohlin literature has been suffering from the tyranny of nonrefutability. The trade-governing principle of comparative advantage, the Ricardo–Haberler–Deardorff (RHD) theorem, yields a refutable general-equilibrium prediction about the pattern of international trade and allows for a theory-based assessment of the magnitude of the gains from trade. The recent experimental evidence on Japan's nineteenth-century opening-up to world trade provides a strong case for the hypothesis that comparative advantage governed Japan's international trade in its early trading years. The aggregate gains from that trade are estimated to be no larger than 9% of Japan's GDP.