How reliable are Taylor rules? A view from asymmetry in the U.S. Fed funds rate

This note raises the issue of whether asymmetry in estimated monetary-policy rules for the U.S. can be a spurious result due to model specification, rather than a robust feature of the estimated rules themselves. I estimate standard - linear - Taylor rules, and test for conditional symmetry using the procedures presented in Bai and Ng (2001a). The results cast doubt on Taylor rules providing a consistent description of the conduct of the Fed.