On competing theories of economic growth: cross-country evidence

In the recent literature on economic growth, four approaches compete to explain growth trends in per capita GDP of the world set of countries: (1) the Solovian approach; (2) the New Theory of Endogenous Growth; (3) the Abramovitz–Baumol ‘catching up’ approach; (4) the Kaldor approach of increasing returns and cumulative causation. Approaches (3) and (4) are formalized here in a single model and then tested against approaches (1) and (2) on a common database and sample of countries. The main conclusions are the following: that labour is not the limiting factor on growth, as implied by (1) and (2); that catching up is an important and a non-declining process; that increasing returns to output rather than to capital are very important and accelerate the catch-up with the US trend; that a lack of social capabilities and technological push may reverse catching up into falling behind with respect to the US productivity level; and that the pattern of growth changes between 1960–73 and 1973–88 because physical investment becomes less important while R&D becomes more important, and because cumulative causation through export growth ceases.