Document Type

Master's Culminating Experience

Publication Date



The purpose of this study is to test the convergence hypothesis that prevails the neo-classical economic literature. In light of research on cross-country economic growth, the paper is adopting the same kind of methodology in studying income growth across U.S. states. The paper starts with a survey of literature in the study of growth of convergence. Then it tries to define the term of “convergence” in various economic implications, in particular the notion of “ð-Convergence” versus “ß-Convergence”, “conditional convergence” versus “absolute convergence”, and such popular notion as “club convergence”. The paper then goes into the quantitative analysis of U.S. per capita personal income change in a time series. The time-series data is firstly divided up by U.S. census regions and the pattern of regional income change overtime is carefully identified. ð-Convergence is tested by plotting personal income dispersion across U.S. states in time series from 1958 to 1996. ß-Convergence is tested by doing a simple regression of personal income growth on personal income at its initial level. The issue of alleged divergence since 1980s is carefully addressed and the importance of State Price Index is evaluated for the sake of the accuracy of economic studies on convergence and growth. The paper ended with a multiple-regression analysis aiming at identifying some attributes and determinants of income growth at state level.