This paper aims to test the existence of different growth regimes, that is of different relationships between growth rate and income level. We propose a simple nonlinear growth model and test its empirical implications by estimating Markov transition matrices and stochastic kernels. We show that growth is indeed nonlinear: a first phase of slow or zero growth is followed by a take-off and, finally, by a phase of acceleration.We discuss the relevance of these results with respect to the issue of convergence and reversibility of development, in thelight of models of structural change and technological diffusion.
|Rivista||JOURNAL OF DEVELOPMENT ECONOMICS|
|Stato di pubblicazione||Published - 2007|
All Science Journal Classification (ASJC) codes
- Economics and Econometrics