Working paper No. 99
By Ben Martin and Chris Salmon
This paper examines the empirical importance of parameter uncertainty for monetary policy-making in the United Kingdom, following the method used
by Brian Sack of the US Federal Reserve. Using a VAR model of the UK economy and an assumed quadratic loss function for the policy-maker, we calculate an optimal interest rate rule first ignoring parameter uncertainty, then assuming that the parameter uncertainty is given by the estimated standard errors on the VAR coefficients. We contrast these rules with the estimated interest rate equation from the VAR. The optimal rule accounting for parameter uncertainty results in a less aggressive path for official interest rates than when parameter uncertainty is ignored. However, the estimates of parameter uncertainty are not so large that the optimal rule matches all the characteristics of the actual path of official rates.