The Effect of Macroeconomic Announcements on Credit Markets: An Autometric General-to-Specific Analysis of the Greenspan Era

I show that a congruent, parsimonious, encompassing model discovered using David Hendry’s econometric modelling approach and Autometrics can overcome the many inadequacies of the typical static models of US Treasury returns regressed on macroeconomic announcements. The typical specification tends to fail most, if not all, specification tests. Further, the techniques employed are able to expand our knowledge of time varying risk premia and asymmetric news responses in financial markets. Previously studied within a GARCH framework, such methods offered little evidence as to the precise sources of the asymmetries. Asymmetric effects are shown to be concentrated in a handful of announcements, such as the Employment Cost Index and Core PPI. Results suggest a place for general-to-specific modelling in financial economics, a place where it has only recently begun to be employed. These results underscore the contributions of David F. Hendry and his collaborators in econometric modelling, they also and demonstrate the need for better models in finance that may be alleviated by employing modelling practices advocated by econometricians doing research in the LSE/Oxford tradition.