Metro Business Cycles

We construct monthly economic-activity indices for 51 U.S. metropolitan statistical areas for 1990 to 2014. Each index is computed via a dynamic factor model that includes 14 variables measuring various aspects of economic activity in a metro area. We estimate the dynamic factor model using the recently developed maximum-likelihood approach of Bańbura and Modugno (2014), which allows for arbitrary patterns of missing data and enables us to accommodate mixed-frequency data and differences in data-publication lags. Our indices highlight important similarities and differences in business cycles across metro areas. During the national recessions of the early 1990s and early 2000s, a number of metro areas experience sizable recessions, while other areas escape recessions altogether during one or both of these periods. In contrast, all metro areas suffer severe recessions around the time of the recent Great Recession. Nevertheless, there are significant differences in the length and depth of recent recessions across metro areas, and we find that these differences are strongly related to local housing-market conditions. We also estimate each metro area’s “beta”—its sensitivity to national economic activity—and relate the betas to metro characteristics.