In lieu of an abstract, here is a brief excerpt of the content:

A P P E N D I X C Interpreting the Values of RCAs in the Presence of Intraindustry Trade In Chapter 2 and in Section 4.1, the properties of revealed comparative advantages were examined within the context of the standard Heckscher-Ohlin model of trade. However, HO theory describes only one type of flow: interindustry trade due to differences in relative resource endowments. This appendix examines the consequences of intraindustry trade for the interpretation of RCAS. The product-differentiation and economies-of-scale theory (the PE theory) reviewed in Section 3.2 views trade flows as comprising two components (Helpman and Krugman 1985, pp. 141-42). Intraindustry flows increase the number of varieties available for consumption, when the number efficiently produced domestically is limited by economies of scale; interindustry trade compensates for differences in resource endowments across countries. One can view the first component as balanced for each good, while any imbalance in the second reflects relative resource endowments.1 The second component is the focus of interest in attempts to draw conclusions about the factor endowments of a country. The way in which a particular RCA reflects differences in resource endowments depends on the size of intraindustry trade in the countries in question, but there is no way of knowing the amount of intraindustry trade precisely. Theory, however, helps. Helpman and Krugman (1985, p. 173) predict that countries with higher per capita incomes have larger shares of intraindustry trade. Balassa and Bauwens's (1984) empirical study strongly confirms this prediction. Furthermore, one can predict the types of goods in which intraindustry trade is particularly important—goods for which style and quality are important and differ between manufacturers. Thus, for example, one would expect intraindustry trade to be important for cars and clothes, but not for crude oil or raw cotton. Finally, and perhaps most importantly, theory tells 1 Of course, the flows that make up the first and second components are in the same products, so the distinction between the two types of trade in a single industry is theoretical rather than anything that can be observed empirically. 244 Appendix C us that any trade imbalance in any specific commodity category reflects resource endowment differences, not product differentiation. Armed with the lessons of theory, one can proceed to examine the possibilities for the RCAs. Let us begin with an example. Countries A and B trade in a multicammodity, multicountry world. Country A conducts more intraindustry trade than country B (perhaps Country A has a higher per capita income than B). Good i is subject to a larger-than-average amount of intraindustry trade. Assume that interindustry trade in i reflects imbalances in the endowments of a specific resource, which both A and B have in relative abundance. Then A will certainly export more than an average amount of i, because both interindustry trade and intraindustry trade will be high. As assumed, country B has an insignificant amount of intraindustry trade. Thus, one can conclude that xIB > 1, but no condition on xlb can be established. For the import RCAs, the reverse case holds. B imports little of /, but A has large intraindustry imports. Then, m,b is certainly less than unity, but one cannot predict the size of mm. The argument is simpler for the export-import measures. Intraindustry exports and imports are equal for each commodity. The export figure is in the denominator, the import figure is in the numerator. Thus, the size of intraindustry trade does not affect the ordering of wv and unity. In the present example, both w,„ and W1 J, are greater than one. However, the presence of intraindustry trade affects the relative ordering of the two countries' indices. Because intraindustry trade is more important in country A, its index will be closer to 1 than B's is. Hence, wia 1 X / b 7 1 X,a 7 X,b Large Large m , a 7 1 mlb mlb W,a > 1 w,b > 1 wia 1 x,b rnlb W,a > 1 W/i> = 1 Wia > Wib X,a > 1 X,b 1 Mfo w,b x«?1 x,b ?1 x,a ? Xib Average Large mla? 1 mib m,b w,a = 1 IV,b > 1 W,a Xib Average Average m,a7 1 mib m,b W,a = 1 w,b = 1 Wia = Wlb Xlb W,b X , t o 7 1 Xia ? X,b Small Large m,a > 1 mlb m,b IV,a 1 w,a 1...


Additional Information

MARC Record
Launched on MUSE
Open Access
Back To Top

This website uses cookies to ensure you get the best experience on our website. Without cookies your experience may not be seamless.