Estimating potential output using business survey data in a SVAR framework

Potential output and the related concept of output gap play a central role in the macroeconomic policy interventions and evaluations. In particular, the output gap, defined as the difference between actual and potential output, conveys useful information on the cyclical position of a given economy. The aim of this paper is to propose estimates of the Italian potential GDP based on structural VAR models. With respect to other techniques, like the univariate filters (i.e. the Hodrick-Prescott filter), the estimates obtained through the SVAR methodology are free from end-of-sample problems, thus resulting particularly useful for short-term analysis. In order to provide information on the economic fluctuations, data coming from business surveys are considered in the model. This kind of data, given their cyclical profile, are particularly useful for detrending purposes, as they allow to include information concerning the business cycle activity. To assess the estimate reliability, an end-of-sample revision evaluation is performed. The ability of the cyclical GDP component to detect business cycle turning points is then performed by comparing the estimated output gaps, extracted with different detrending methods, over the expansion and recession phases of the Italian business cycle chronology.