| We evaluate the Taylor rule and investigate its stability for the period 1963Q2 to 1999Q4. Using a benchmark model, we demonstrate that the equation cannot be evaluated over this period without taking into account parameter instability and structural changes, which reflect changing monetary policy preferences. Neglecting to allow for at least one shift in the equation can lead to artificial results by ignoring the heterogeneity of the long-run relationship, while it is not capturing the changes in monetary policy preferences. To estimate the equation over the 1963 Q2--1999Q4 period, we follow Bai and Perron's (1998) recent methodology, with which we find evidence for up to five breaks. |
