Econometric Estimations and Test Results, Empirically Valuable Insights About Reality and Consulting in Economic Policy

This paper critically assesses the current policy consensus that greater integration into the world economy is beneficial for poor countries. Both the theoretical and empirical basis is examined, concentrating on a) modern growth theory in an open-economy setting, and b) a wide range of newer empirical studies on the link between openness and economic performance. Throughout great emphasis is placed on dynamic gains and losses, and the importance of knowledge diffusion and technical progress. Although endogenous growth theory is often used to highlight the dynamic gains from economic integration, there is a surprising amount of ambiguity. The literature suggests a number of possibilities for divergence, due to linkage effects and endogenous learning. In contrast, open-economy extensions of the neo-classical growth model provide a much clearer case for greater openness and economic integration. Empirically, there is little evidence of convergence in per-capita incomes between rich and poor countries, irrespective of whether one considers longer time series or the post-war period only. The last two decades have even witnessed an acceleration in the speed of divergence. The empirical evidence on the link between openness and economic performance does not provide unambiguous support for the policy consensus. For instance, there is no robust econometric evidence linking faster economic growth to trade liberalisation. However, there is some evidence at the aggregate level for technological and knowledge spillovers between rich and poor countries. It is difficult to pinpoint the reasons though. Microeconomic studies find that exporting activities do not seem to make firms more efficient, and foreign direct investment does not raise diffusion of knowledge to developing country firms. We are thus left with the conclusion that the potential benefits of greater openness and, by implication, increased globalisation for developing countries have been significantly overstated. There simply is neither the empirical evidence to substantiate the claims, nor are there unambiguous theoretical results