Empirics of Hills, Plateaus, Mountains and Plains: A Markov-Switching Approach to Growth

46 Pages Posted: 14 Jul 2005

See all articles by Michal Jerzmanowski

Michal Jerzmanowski

Clemson University - John E. Walker Department of Economics

Abstract

Most economic growth research has been aimed at explaining cross-country differences in average growth rates. However, it is well known that growth experiences differ over time within a country almost as much as they differ among countries. This paper builds on Pritchett's (2000) observation that the growth process can be thought of as transitions between different growth regimes and proposes a framework for systematic analysis of such regimes and the transition dynamics. I estimate a Markov-switching regression to characterize four distinct growth regimes and transitions among them. The results show that countries switch among regimes of stable growth, miracle catch-up, stagnation and crisis with the transition probabilities determined by the quality of institutions. Better institutions appear to improve long run growth by making growth episodes more persistent. Low average growth rates in countries with weak institutions are a result of these countries spending more time in stagnation regimes rather then being incapable of fast growth at all. Weak institutions do not rule out growth take-offs but limit their sustainability. The approach directs attention to shifts in growth performance instead of the average growth rates.

Keywords: Economic growth, regime switching, growth take-offs, institutions

Suggested Citation

Jerzmanowski, Michal, Empirics of Hills, Plateaus, Mountains and Plains: A Markov-Switching Approach to Growth. Available at SSRN: https://ssrn.com/abstract=755484

Michal Jerzmanowski (Contact Author)

Clemson University - John E. Walker Department of Economics ( email )

Clemson, SC 29634
United States

Do you have negative results from your research you’d like to share?

Paper statistics

Downloads
364
Abstract Views
2,901
Rank
150,248
PlumX Metrics