The stabilisation problem: the case of New Zealand

This paper examines stabilisation bias - the difference between the inferior macroeconomic outcomes attained with discretionary monetary policy relative to the ideal that could be attained with commitment policy. The paper works within the linear-quadratic framework and represents the monetary policy problem for the central bank as setting the interest rate in order to minimise an explicit loss function for macroeconomic variables. The government's problem is one of "optimal negotiation", whereby the government, representing society, joins with the central bank to search for the optimal set of loss function parameters to be embedded in a contract with the central bank. The framework, due to Rogoff (1985), is usefully applied to the case of New Zealand where recent Policy Target Agreements - contracts between the government and the central bank - are interpreted as representing society's preferences between inflation, output and other dimensions of macroeconomic stability. Within the context of an estimated, small open economy model, a sizeable stabilisation bias is found for New Zealand. Substantial reductions in the stabilisation bias can be achieved by strategic optimal delegation behaviour on the part of the government. It transpires that the weight the central bank should have on the variance of the output gap is considerably lower than the weight society places on the variance of the output gap.