The Observational Equivalence of Taylor Rule and Taylor-Type Rules

In the past few years the view has commonly been expressed that central banks follow `Taylor Rules' (as first promulgated by Henderson and McKibbin (1993)). We show that the appearance of such an interest rate rule – a ‘pseudo-Taylor rule’ – can be created by a standard macro model in which actually a money supply rule is operating with no interest rate feedback – i.e, where there is in fact no Taylor rule operating at all. Hence an interest equation does not identify a (structural) Taylor rule; a Taylor rule and a pseudo-rule, though corresponding to different structural models, are ‘observationally equivalent’ to use the expression coined by Thomas Sargent (1976). It remains an open question whether Taylor rules or money supply rules are appropriate from a welfare viewpoint.