GLOBAL FACTOR TRADE: Theory 2
Hicks-Neutral Technical Differences A wide body of literature, both in productivity and in trade, suggests that there are systematic cross-country differences in productivity, even among the richest countries [e.g. Jorgenson and Kuroda (1990)]. This is very likely an important reason why Trefler (1995) found that the data suggests poor countries are “abundant” in all factors and vice versa for the rich countries. Bowen, et al. (1987) and Trefler (1995) have focused attention on Hicks-Neutral technical differences as a parsimonious way to capture these effects. Under this hypothesis, the technologies of countries differ only by a Hicks-neutral shift term. This can be characterized via country-specific technology shifts 8c:
Bc 8 = 8cB V c
In order to implement an amended HOV equation, it is convenient to think of the productivity differences as reflecting efficiency differences of the factors themselves (rather than technology per se). For example, if we take the US as a base and US factors are twice as productive as Italian factors, then 8Italy E = 2. In general, we can express a country’s endowments in efficiency terms:
All succeeding models and the associated empirical specifications will be in efficiency units, although we will henceforth suppress the superscript E for simplicity.
So far we have allowed differences in input coefficients across countries only as a Hicks-neutral shift. For cases of adjusted FPE, this implies that capital to labor ratios are fixed by industry across countries. However, there is good reason to believe this is not the case. The simple Rybczynski relation suggests that countries with a relatively large stock of capital should have an output mix shifted toward relatively capital intensive goods, but with FPE they should not use different input coefficients within any individual sector. Baumol, Dollar and Wolff (1988) estimated cross-country differences in capital to labor usage and found this was correlated with country capital abundance.
They interpreted this as evidence against the FPE model, although they recognized that aggregation might be a problem. We develop this insight in a model that accounts for the positive relation between country and industry capital to labor usage, yet preserves (approximate) FPE and the simple HOV prediction. This is valuable in that it will provide a first set of theoretical insights that help us to understand why the mystery of the missing trade might arise in previous data exercises, even if the HOV prediction is being met. In the following section we will go on to consider the question of how to pursue the problem if indeed FPE has broken down.
In order to make our discussion compact, we will provide only a sketch of the model that provides the essential insights. Consider as a starting point the Dornbusch-Fischer-Samuelson (1980) continuum Heckscher-Ohlin model. Goods are arrayed on the unit interval with continuous and strictly increasing capital to labor ratios by sector. We consider first the integrated equilibrium (cf. Helpman and Krugman 1985). The FPE set is depicted in Figure 1 as a “Deardorff lens,” reflecting the factor intensities and usages for the corresponding sectors. Electronic Payday Loans Online