Working Paper: NBER ID: w5625
Authors: Donald R. Davis; David E. Weinstein; Scott C. Bradford; Kazushige Shimpo
Abstract: The Heckscher-Ohlin-Vanek model of factor service trade is a central construct in international economics. Empirically, though, it is a flop. This warrants a new approach. Using Japanese regional data we are able to test the HOV model by independently examining its component production and consumption elements. The strict HOV model performs poorly because it cannot explain the international location of production. However, relaxing the assumption of universal factor price equalization yields a dramatic improvement. We also solve most of what Trefler (1995) calls the mystery of the missing trade. In sum, the HOV model performs remarkably well.
Keywords: Heckscher-Ohlin model; factor service trade; international trade; regional data
JEL Codes: F11; F14
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| strict Heckscher-Ohlin-Vanek model (F11) | poor explanation of international location of production (F29) |
| relaxing the assumption of universal factor price equalization (F16) | improvement in model performance (C52) |
| regional characteristics (R11) | applicability of HOV model in Japan (R48) |
| regional production characteristics (R32) | international trade patterns (F10) |