A microfounded model of money demand under uncertainty, and its empirical validation using cointegration and rolling-window dynamic multiplier analysis

In this article we derive a microfounded model of money demand under uncertainty built on intertemporally optimizing risk-averse households. Deriving a complete solution of the optimization problem taking the intertemporal budget constraint into account leads to ambiguous effects w.r.t. to the impact of capital as well as inflation risk, thus contradicting standard results. We estimate both the long- and short-run model dynamics as well as potential time-variation by means of a rolling-window dynamic multiplier analysis using the error-correction framework for the U.S. economy between 1978q1 to 2013q4. The results reveal that U.S. households increase their demand for money in response to positive changes in inflation and stock market risks.