By Mikael Bask and João Madeira
We outline a dynamic stochastic general equilibrium (DSGE) model with trend extrapolation in asset pricing that we fit to quarterly U.S. macroeconomic time series with Bayesian techniques. To be more precise, we modify the DSGE model in Smets and Wouters (2007) by incorporating asset traders who use a mix of fundamental analysis and trend extrapolation in asset pricing. We conclude that trend extrapolation in asset pricing is quantitatively relevant, statistically significant and results in a substantial improvement of the model’s fit to the data. We also find that the strength in trend extrapolation is much stronger during the Great Moderation than it was prior to this period. Moreover, allowing for asset mispricing leads to more pronounced hump-shaped dynamics of the asset price and investment. Thus, asset price misalignments should be an important ingredient in DSGE models that aim to understand business cycles dynamics in general and the interaction between the real and financial sectors in particular.
DSGE models tend to assume away mispricing, at least beyond Lucas-type real/nominal confusion. This paper shows it actually matters and can help understand current economies.