Using State Indexes to Define Economic Regions in the U.S.

Abstract

When regional economists study the interaction of multi-state regions in the U.S., they typically use the regional divisions developed by the U.S. Bureau of the Census or the Bureau of Economic Analysis (BEA). The current census divisions were adopted in 1910 and divide the states into nine regional groups for the presentation of data. Since the 1950s, the BEA has grouped the states into eight regions based primarily on cross-sectional similarities in their socioeconomic characteristics. The BEA definition of regions is perhaps the most frequently used grouping of states for economic analysis. Since many economic studies of regions concentrate on similarities and differences in regional business cycles, it seems appropriate to group states into regions based on some common cyclical behavior. This paper explores the possibility of grouping states into regions based on common movements in state indexes of economic activity. These state indexes are variants of the coincident index developed by James Stock and Mark Watson for the U.S. economy. The author has applied cluster analysis to the monthly changes in these economic activity indexes to group the states into regions with similar business cycles. He has identified six distinct regions consisting of contiguous states with similar monthly changes in their economic activity indexes.