Rethinking the Value of Initial Claims as a Forecasting Tool

Do such episodes occur randomly, or is there a pattern to claims’ failure to predict employment fluctuations? This edition of Current Issues compares the accuracy of a simple forecasting model that incorporates initial claims with the accuracy of a forecasting model that excludes initial claims. Interestingly, we find that the reliability of claims in predicting employment fluctuations depends on the state of the business cycle. In our model, claims serve as a useful tool for forecasting employment during recessions. Very early on in economic expansions, however, our model loses its predictive power, and claims actually worsen forecast accuracy.