We consider whether disaggregated data enhances the efficiency of aggregate employment forecasts. We find that incorporating spatial interaction into a disaggregated forecasting model lowers the out-of-sample mean-squared-error from a univariate aggregate model by 70 percent at a two-year horizon

This paper examines the spatial interaction of neighboring cities over their employment cycles. The cycles of neighboring cities tend to be more similar to one another than are those of non-neighboring cities, although this is due primarily to neighbors' tendency to be in the same state. In...