Abstract:
While trade integration is often regarded as a principal determinant of economic growth, the empirical evidence for a causal linkage between trade and growth is ambiguous. This paper argues that the effect of trade in dynamic panel estimations depends crucially on the specification of trade. Both from a theoretical as well as an empirical point of view one specification is preferred: the volume of exports and imports as a share of lagged total GDP. For this trade measure, a positive and highly significant impact on economic growth can be found.