Predicting a recession in real time is difficult, which is why one can make good money with a good forecast. Here, FRED offers one of many such forecasts: a recession probability index computed by Marcelle Chauvet and Jeremy Piger. This forecast is backed up by research the authors have published in the peer-reviewed journals International Economic Review and the Journal of Business and Economic Statistics, with an early St. Louis Fed working paper added here for good measure. As the graph above shows, their forecasting method’s past performance is impressive; the predicted recession dates align well with the official NBER recession dates. Of course, it is difficult to compute any forecast in a timely fashion: One has to wait for the appropriate data to be released, and only then can one compute the forecast. In this case, that translates into a delay of about three months.
How this graph was created: Search for “recession,” and the first series shown should be “Smoothed U.S. Recession Probabilities.”
Suggested by Christian Zimmermann
View on FRED, series used in this post: