JOURNAL OF URBAN ECONOMICS 21,127-145 (1987) Urban Sector Income Distribution and Economic Development’ CHARLES M. BECKER Department of Economics,Vanderbilt University Nashville, Tennessee 37235 Received May 14,1982; revised May 16,1985 This paper examines the relationship between measures of urban sector in- equality and economic development for a sample of 25 developing and newly industrialized countries. A U-shaped relationship is found in which bottom urban quintiles’ income shares initially decline and then rise as per capita income in- creases. This relationship is strengthened when an estimate of urban per capita income replaces national per capita income as the development measure. The curves suggest that per capita incomes of the bottom quintiles will never decline as development proceeds, but may rise only very slowly. 0 1987 Academic press, IX The hypothesis that measuresof national income inequality initially tend to worsen and then improve as an economy develops is well-known and has some empirical support. Kuznets [15] first suggested this U-shaped relation- ship and provided the first cross-country and time-series evidence on its behalf. Oshima’s [24] early study, Ahluwalia’s [l] and Chenery and Syrquin’s [7] cross-country regressions,along with Paukert’s [25] study and Cline’s [S] work in his survey based on more extensive data have also found a U-shaped relationship between bottom groups’ income shares and per capita GDP. Fields [lo] provides an extensive survey of time-series evidence and concludes that the data neither strongly confirm nor contradict the U curve found in cross-country regressions. Recent evidence, however, sug- gests that the British economy has indeed experienced such a long-run U-shaped inequality trend (Williamson [30]). It is also of interest whether a similar U-shaped pattern can be observed within sectors of an economy. The most common explanations of the U pattern, termed “modem sector enlargement” models by Fields [lo], are based on urban/industrial sector growth that is rapid in comparison with rural/agricultural sector growth. This long-run transformation underlies the explanation suggested by Lewis [16] and refined by Robinson [26]. Such models do not stress changing relationships within economies’ urban and ‘Valuable advice and criticism have been provided by Rudolph Blitz, Malcolm Gem, W. Arthur Lewis, Samuel Morley, Mieko Nishimizu, and in particular Edwin S. Mills, with whom I have worked closely on related topics. Lee Jarrard has provided excellent research assistance. 127 0094-1190/87 $3.00 Copyright 0 19X7 by Acadermc Press. Inc. All rights of reproduction in any form reserved