In this paper we examine whether data from business tendency surveys are useful for forecasting the macro economy in the short run. Our analyses primarily concern the growth rates of real GDP but we also evaluate forecasts of other variables such as unemployment, price and wage inflation, interest rates, and exchange-rate changes. The starting point is a so-called dynamic factor model (DFM), which is used both as a framework for dimension reduction in forecasting and as a procedure for filtering out unimportant idiosyncratic noise in the underlying survey data. In this way, it is possible to model a rather large number of noise-reduced survey variables in a parsimoniously parameterised vector autoregression (VAR). To assess the forecasting performance of the procedure, comparisons are made with VARs that either use the survey variables directly, are based on macro variables only, or use other popular summary indices of economic activity. As concerns forecasts of GDP growth, the procedure turns out to outperform the competing alternatives in most cases. For the other macro variables, the evidence is more mixed, suggesting in particular that there often is little difference between the DFM-based indicators and the popular summary indices of economic activity.
Business survey data Dynamic factor models Macroeconomic forecasting