Why Low Inequality Spurs Growth: Savings and Investment by the Poor

the Inter-American Development Bank, and in an IMF-World Bank sponsored conference in Escorial, Spain for helpful comments on a related paper. The usual disclaimer applies. 2 INTRODUCTION New empirical evidence on the relationship between inequality and growth contradicts an important dimension of the conventional wisdom regarding the nature of the process of transforming low income into high income economies. The conventional wisdom: high rates of economic growth are likely to be associated with high levels of inequality in the distribution of income. This prediction, made decades ago, was largely based on economic theory. Kuznets (1955) saw rising inequality as the by-product of growth and structural change. As workers shift from a low to a high productivity sector there is a tendency for inequality to increase at first. Kaldor (1978) and-more specifically for developing countries-Galenson and Leibenstein (1955) saw the causality running the other way, from high inequality to rapid growth. If the rich have a higher marginal propensity to save than the poor, greater concentration of income results in higher savings in the aggregate, hence in more rapid capital accumulation and growth. There is evidence, however, that countries which, in 1960, had relatively low levels of inequality grew faster over the subsequent three decades than countries in which the distribution of income was more skewed. 1 Rather than being associated with more rapid growth, high levels of inequality appear to be a constraint on growth. Thus, higher levels of inequality help to explain Latin America=s slower growth relative to East Asia. 1 For estimates of growth rate functions in which the impact of initial inequality is econometrically assessed for a large number of developing countries while controlling for other determinants of growth, see Birdsall, Ross, and Sabot (1995) and Clarke (1995). Figure 1 relates percentage growth in GNP for the period 1965 to 1990, and income inequality, as measure by the ratio of the income share of the top and bottom quintiles. Latin American countries are concentrated in the southeast corner, indicating that they experienced slow growth and high inequality, while 3 East Asian countries, having achieved both low inequality and rapid growth, stand alone in the northwest corner. Birdsall, Ross and Sabot (1995) assess econometrically the relationship between inequality and growth, controlling for other determinants of growth. The addition of a measure of inequality, in 1970 or earlier, to the basic Barro function explaining variations in growth rates among …