Working Paper, Federal Reserve Bank of Atlanta 97-8
This paper examines whether financial aggregates provide information useful for predicting real output growth and inflation, extending the inquiry conducted in Tallman and Chandra (1996). First, we investigate whether perfect knowledge of the future values of financial aggregates helps improve significantly the forecasting accuracy of output and inflation in a simple vector autoregression framework. The results display only one notable improvement to the forecasts with the addition of perfect information on the financial aggregates—future information on credit growth helps improve the prediction accuracy of real output growth. The improvement is most noticeable during the early 1990s recession. Second, we test whether the financial aggregates are important explanators within single-equation models that are more rigorously fitted to the data. We find only one instance in which an aggregate helps explain the variation in either real output growth or inflation—that is, the growth in credit helps explain the growth in real output in a particular specification of the output model. This finding, though, is sensitive to the choice of foreign output proxy. In sum, we conclude that while credit may have some useful information in times of financial restructuring it is unlikely that there is information in financial aggregates that is exploitable systematically for predicting either real output growth or inflation.
Australia Credit Financial markets Vector autoregression