The Information Content ofthe Paper-Bill Spread: The Case of Canada

Abstract

Considerable attention has been given to examining the information content of the paper-bill spread in terms of predicting U.S. real economic activity. However, the results have been mixed. This paper examines the robustness of the paper-bill spread as an indicator variable ofreal output growth by studying Canadian data. This study uses the newly developed generalized forecast error variance decomposition technique [Koop, et al. (1996), Pesaran and Shin (1998)]. The results indicate that the information content of the paper-bill spread in Canada is non-trivial even when the performance of the U.S. economy is taken into account. Moreover, Canadian economic activity may not be tied directly to how well the U.S. is performing per se, but is instead significantly effected by unexpected changes to the growth rate of U.S. real output. (EO, GI)