News Driven Business Cycles and data on asset prices in estimated DSGE models

• We estimate an NDBC model using data on macro variables and asset prices.
• We examine a long-run news specification and a short-run news specification.
• Inference about drivers of fluctuations is sensitive to structure of news shocks.
• Long-run news shocks fit data better than traditional short-run news shocks.
• Long-run news shocks are main drivers of stock prices, but not of macro variables.

We demonstrate that inference from estimated structural News Driven Business Cycle (NDBC) models about the main drivers of fluctuations in macroeconomic variables and asset prices is sensitive to assumptions about the structure of the news shock processes. We show that, when data on asset prices are used in the estimation, a long-run news shock specification has a better fit than the short-run news shock specification which is prevalent the existing literature. The variance decompositions from the former model specification reveal that long-run news shocks are not the main drivers of macroeconomic variables, but do account for the majority of aggregate stock market fluctuations.