Money demand, structural drift, and equity returns
Recent evidence that the stock of money may transmit monetary policy into the real economy, independently of short-term interest rates, has resulted in an increased focus on monetary aggregates and money demand. I take a portfolio approach to money demand and estimate the demand for money with two opportunity cost measures---traditional bond returns and expected equity returns.
Expected equity returns are found to be important determinants of money demand in the long run. The effects of bond returns on the long-run money demand have become increasingly insignificant in the last decade and a half. I solve the 'puzzle' of bond return insignificance by explicitly allowing for structural change. Once expected equity returns and structural drifts are formally included in the specification of money demand, the interest elasticity with respect to bond return is negative and statistically significant. The elasticity of money demand with respect to expected equity returns has a significant positive drift and has become positive recently. The consumption elasticity has a strong negative drift and may be a source of the observed money demand instability.
A conventional specification of money demand cannot accommodate either multiple returns or a positive elasticity of money demand with respect to equity returns. I first show that the inclusion of heterogeneous agents or assuming uncertain asset returns allow multiple asset returns to enter the money demand equation. Then I develop a simple theory, based around the idea that multiple assets may provide liquidity services, and show that it can support both positive and negative elasticities with respect to equity returns.
Return on investment;