Two's not company: mis-aggregation and "supply-induced" unemployment increases

The seminal work by Blanchard and Quah (1989) identifying long-run shocks finds that the unemployment rate increases following a “supply” shock. This puzzling result engenders a lively debate between leading schools of macroeconomic thought. But should we employ this model to study macroeconomic fluctuations? Faust and Leeper (1997) warn that the model is not reliable for structural inference, nor is the related Bayoumi and Eichengreen (1993) model. In this paper, I revisit the disputed result from the viewpoint of IRF’s bias. Using a novel methodology to parse the sources of IRF’s bias, I find that “supply-induced” unemployment increases reflect mis-aggregation of technology and labor-supply shocks.