Regional income inequality in the United States, 1913--2003

The starting point of this study is an ambiguity that emerges from two currents of the literature: the income inequality literature, and the regional convergence literature. In the income inequality literature, a natural decrease of inequality at later stages of economic development is predicted by Kuznets (1955) but this hypothesis does not receive strong empirical support over the long-run; the succession of U and inverted-U curves does (Piketty and Saez, 2003). In the regional convergence literature, the neoclassical economists support convergence whereas the geographers marshal evidence for divergence, at least in the case of the United States at the sub-national level. Fundamentally, both literatures study the same question, namely, the income convergence or divergence across regions and over time.; This paper sheds light on this ambiguity through the construction and use of a new homogeneous set of panel data of 4,641 income inequality observations (the 50 U.S. states plus the District of Columbia, over 91 years from 1913 to 2003). This database represents the top 10 percent of the income distribution, but data from other sources are needed to account for average income. Meanwhile, the new income database of the top decile offers an alternative estimate of average income figures used by Barro and Sala-i-Martin to study the same topic.