Economic stagnation, fixed factors, and policy thresholds

Analysis of decade-long growth rates in all countries shows a striking regularity: episodes of rapid growth are limited largely to a middle range of initial income; neither very poor nor very rich countries experienced rapid growth. Episodes of negative growth are limited to low and middle-income countries. This paper develops a simple model that sheds light on this experience. The model has two familiar elements from the growth literature: (1) a Stone-Gearly utility function (saving is low at low income), and (2) fixed factors with the marginal product of capital bounded away from zero. The second property is derived by assuming an elasticity of substitution greater than one between an exogenous labor input and a broad concept of capital. The paper extends the model to consider multiple capital goods and public capital. It finds that stagnation is consistent with an array of statistical evidence because of fixed factors. Economic policies - not initial conditions - determine whether countries stagnate. Results confirm that initial income and policy variables have a different effect on whether a country stagnates than they do on the rate of growth once it starts growing, as expected from the distinction between steady-state and transitional effects. These results suggest that cross-section growth regressions may be misspecified because of the nonlinearity inherent in the possibility of steady-state stagnation.