August 10, 2009
Jon Huntley and Eric Miller
We compare the performance of a subset of CBO's economic forecasts against that of an unrestricted vector autoregression (VAR) model. We evaluate forecasts of real economic indicators as well as budget-related nominal statistics. We find that under most specifications, the VAR performs competitively with, if slightly worse than, the corresponding CBO forecasts at up to 20 quarters. Therefore, a simple VAR is unlikely to be able to contribute directly to improving budget forecasts. The only series for which the VAR outperforms the CBO forecast is the growth in real consumption.