Economic Downturn, Change in Unemployment, and the Midwest: A Quantile Regression Approach

Abstract

Although every area of the U.S. was hit by the Great Recession, the change in unemployment rates varied widely across locations. We use a quantile regression approach to examine the tails of the distribution of change in unemployment rates between 2006 and 2009 across all counties in the Midwest. We find consistent evidence that manufacturing was the largest contributor to the change in unemployment across the three conditional distributions examined, with the impact increasing as we move from the 25th quantile to the mean (OLS) and 75th quantile. Likewise, local labor mobility has amplifying effects on the change in unemployment rates, while educational attainment has a moderating effect. Our results suggest that human capital and location-and industry-targeted policies are important in promoting recession-resilient local economies.