Explaining differences in rates of economic growth has been one of the main tasks of economics since before the time of Adam Smith. Given that rates of growth differ across nations, political subdivisions of nations, and geographical regions, there is much to explain. The explanations can be classified into two broad categories. The first category is growth accounting. Growth accounting studies focus on contributions to economic growth that can be measured directly. The most successful studies have followed a production function approach; Denison (1967Denison ( , 1985 and others have measured the contributions of various production inputs in order to make intertemporal or cross-national comparisons. Although such studies have greatly advanced our knowledge of the sources of growth, the underlying causes of differences remain unexplained.The second category of explanations for differences in growth, then, can be called underlying causes; it is much the larger category. Here, institutional, political, cultural, religious, and ethnic explanations take over. Although measurement is not abandoned -indeed, underlying cause explanations are often tested with growth accounting methods -it tends to be indirect or as part of the residual left after measurable factors have been taken into account. Recently, an institutional-political explanation has greatly enlivened the study of why growth rates differ. In a series of books and articles, Mancur Olson (1982aOlson ( , b, 1983aOlson ( , b, 1984 has argued that differences in rates of economic growth can be explained by differential accumulations of common interest groups. Most empirical tests of the Olson hypothesis have used international data on growth and productivity.1 We propose to test his hypothesis using data for the different American states. In addition to the Olson hypothesis we will test one of its leading competitors: the catch-up hypothesis.