Empirical models of trade, direct investment and growth in developed economies

Kernohan, David (2002) Empirical models of trade, direct investment and growth in developed economies. Doctoral thesis, University of London: London Business School.

Abstract

The importance of direct investment in the global economy seems clear. Yet surprisingly little agreement exists in economics as to what role FDI plays in international economic adjustment. This uncertainty is partly due to a lack of theoretical guidance, however the empirical literature has also failed to reach a conclusive view on the relationship between trade and investment. While an influential WTO (2001) study has concluded that trade and investment were complements, other reviews of the econometric literature contend that the issue is far from resolved. Clearly, both trade and investment correlate over time but a key issue is what endogenous assistance an individual country gains to growth, if any, from success in attracting/supplying FDI. The present study looks at FDI as a topic in the multi-country modelling literature. Here we find that existing models of FDI use predominantly aggregate measures, set in a partial equilibrium framework, which offer little scope for feedback effects from the supply side to investment demand. The approach adopted here has been to construct a small, three-country bilateral flow data-set for the US, Japan and UK, and to use the VAR methodology in which notions of causation are suppressed in the initial modelling. The problem then becomes one of isolating and identifying a convincing FDI vector from a system with initial estimates of multiple cointegrating vectors. A number of novel techniques are adopted to test for single equation integrity, within the VECM system, and six convincing bilateral FDI equations are eventually isolated. Subsequently a mixture of exogeneity testing and the use of the general-tospecific method within the systems setting follow, before six adequate FDI equations are formally identified within the system. The six long-run FDI equations that result are discussed in the context of the FDI literature. Building on three important strands in the literature, on trade, investment, and growth it appears that although FDI may not be linked to productivity growth within countries, it may play a part in linking growth rates - by linking supplyside innovation channels internationally. Whilst, at the same time, FDI may play a part in linking productivity improvements between countries, by stimulating world demand.