© 2018 Elsevier Inc. Forecasting economic activity has attracted a great deal of econometric work, while mixed evidence has been found concerning the ability of the yield spread to forecast gross domestic product (GDP). This paper uses a meta-analysis framework to deal with the heterogeneity in the results seen in the literature. Our findings suggest that nonlinearities, as well as the role of monetary policy, should be considered when modeling this relationship. We also find that the forecasting ability of the yield spread has become much stronger over the last twenty years. Moreover, we argue that the yield spread is a useful tool in predicting economic activity in many major world economies, particularly those of the US, Canada, and Europe and, more importantly, especially during financial stress periods. Last, we find that improvements in the stock market reduce the usefulness of the yield spread in predicting future economic activity.