Factor Endowments, Factor Proportions, and the Allocative Efficiency of U.S. Interregional Trade
利用美国区际贸易数据检验赫克歇尔-俄林理论，发现要素比例差异显著，并探讨商品贸易对要素流动的替代作用及其对配置效率的影响。
THE Heckscher-Ohlin (H-O) theory of trade has been subjected to numerous international tests.' Based upon the empirical evidence presented in the literature, it appears that the theory's explanatory power of the actual pattern of international trade is somewhat limited. This may not be surprising, however, when one considers a host of extraneous factors such as tariffs, quotas, and other policy and institutional barriers to trade which distort the pattern of trade. On the other hand, some of the requisites of the H-O theory are more nearly satisfied in interregional trade within a nation. The extent of policy distortions on the overall pattern of commodity trade should be far less in interregional trade than in international trade. Likewise, both production technologies and demand conditions, assumed to be identical between nations in the H-O theory, should be more nearly so among regions within a nation, while the theoretical basis of regional trade is, of course, the same as that of international trade. It is primarily for these reasons that a number of researchers have more recently turned to interregional trade for the purpose of testing the H-O theory. The current literature on the U.S. regional comparative advantage includes the contributions by Moroney and Walker (1966), Moroney (1972, 1975), Estle (1967), and Klaasen (1973). Whereas these studies focused primarily on the pattern of regional production as revealed by value-added concentration ratios for the U.S. South, the major objectives of this paper are (i) to design and perform factor-proportion tests based upon the actual interregional pattern of trade, and (ii) to investigate one neglected empirical aspect of the H-O theory bearing upon the substitutive role that commodity trade plays for direct factor mobility. In section II, we present estimates of the regional distribution of major input supplies in the United States. The input classes considered are capital equipment, human capital, labor, and renewable and nonrenewable natural resources. In particular, we have found that there are fairly distinct differences among U.S. census regions pertaining to relative aggregate input supplies defined by physical capital/labor ratio, human capital/labor ratio, renewable natural resource! labor ratio, and nonrenewable resource/labor ratio. Section III presents the methodology and major empirical findings of the factor-proportions test as adapted to interregional trade. Section IV extends the analysis to consider an alternative hypothesis of the H-O theory based on the substitutive relationship between direct factor movements and commodity trade as embedded in the original H-O theory. This aspect of the theory has not been explicitly dealt with in the empirical literature, and we consider its practical implication concerning the allocative efficiency of trade. A summary section concludes the paper.