Alternative identifying restrictions in a model of UK real output and prices

Abstract

This paper presents some simulation results for a small Structural Vector Autoregression model of the interaction between real output and prices for the UK economy. The model is estimated using quarterly data over the period 1966:2 to 1995:4. The effects of alternative identifying restrictions on the simulation properties of the model are considered and it is shown that the restriction of long run neutrality of real output with respect to demand shocks has a major impact. Even if long run neutrality is not imposed the results still indicate that most of the variance of real output over long forecasting horizons can be attributed to supply side disturbances. The model is used to analyse the role of demand and supply disturbances in explaining the three major recessions of the period and it is shown that each of these exhibits unique characteristics.