Patterns of Economic Growth: Hills, Plateaus, Mountains, and Plains

43 Pages Posted: 20 Apr 2016

See all articles by Lant Pritchett

Lant Pritchett

Harvard University - Harvard Kennedy School (HKS); Center for Global Development

Date Written: July 1998


The recent growth literature has underestimated the importance - and ignored the implications - of the instability and volatility of growth rates. In particular, the use of panel data to investigate the effects of long-term growth in developing countries - especially with fixed effects estimates - is potentially more problematic than helpful.

Except during the Great Depression, the historical path for per capita GDP in the United States has been reasonably stable exponential trend growth, with modest cyclical deviation. Graphically, growth in the United States displays as a modestly sloping, only slightly bumpy, hill. But almost nothing that is true about per capita GDP for the United States (or for other OECD countries) is true for developing countries.

First, per capita GDP in most developing countries does not follow a single time trend: For a given country, there is great instability in growth rates over time, relative to both average level of growth and to cross-sectional variance.

These shifts in growth rates lead to distinct patterns. Some countries have had steady growth (hills and steep hills); others have had rapid growth followed by stagnation (plateaus); others have had rapid growth followed by declines (mountains) or even catastrophic declines (cliffs); still others have experienced continuous stagnation (plains) or even steady decline (valleys).

Second, volatility - however measured - is much greater in developing than in industrial countries.

These stylized observations about growth rates, Pritchett concludes, suggest that it may be useless to use panel data to investigate long-term growth rates in developing countries. Perhaps more can be learned about developing countries by investigating what initiates (or halts) episodes of growth.

There is something of a professional split in growth literature, Pritchett observes. Macroeconomists studying industrial countries discuss steady-state growth and ponder whether all countries in the convergence club will reach the same happy level in the end. Development economists, on the other hand, are the pathologists of economics, having discovered that developing countries are most emphatically not all alike. Developing countries have found ways to be ecstatic but they have also discovered many different ways to be unhappy.

This paper - a product of the Development Research Group - is part of a larger effort in the group to understand the determinants of economic growth.

Suggested Citation

Pritchett, Lant, Patterns of Economic Growth: Hills, Plateaus, Mountains, and Plains (July 1998). Available at SSRN:

Lant Pritchett (Contact Author)

Harvard University - Harvard Kennedy School (HKS) ( email )

79 John F. Kennedy Street
Cambridge, MA 02138
United States
617-496-4562 (Phone)
617-496-2554 (Fax)


Center for Global Development

2055 L St. NW
5th floor
Washington, DC 20036
United States

Here is the Coronavirus
related research on SSRN

Paper statistics

Abstract Views
PlumX Metrics