Does Government Spending Affect Money Demand in the United States?

This paper elucidates the effect of government spending on money demand in the United States using quarterly data over the period 1973Q1–2013Q4. The related conventional literature has been developed with money demand defined as a function of income, interest rate, exchange rate, and inflation. I propose a new method of income decomposition to the public sector and private sector following Barro’s (1990) spending model. I apply the autoregressive distributed lag cointegration approach (ARDL) and include government spending in the conventional money demand function to investigate its impact on the demand for money. The results confirm the long-run, significant effect of government spending on money demand, finding the elasticity of money demand with respect to government spending to be 0.62. In addition, I find that money demand tends to be unstable and shifts toward the edge of a structural break during recessions. The results do not support Friedman’s (1969) idea that the demand for money is “highly stable.” Instead, the findings suggest that money demand is “slightly stable” during a recession. This confirms that the current evidence is insufficient to switch to interest rate targeting instead of money targeting as an appropriate monetary policy in the United States.