Heterogeneous Convergence

We use U.S. county-level data containing 1,921 cross-sectional observations and up to 29 conditioning variables to estimate heterogeneity in convergence rates across 22 individual U.S. states. Applying GMM estimation, we find significant heterogeneity in the state-level convergence rates. For example, while the average convergence rate is 9.2 percent, the California estimate is 19.9 percent and the New York estimate is 3.3 percent. The estimated convergence rates are essentially uncorrelated with the income levels