The currency ratios 1920–80: A re-examination

Conclusions

This study has analyzed the long run determinants of the ratio of currency to M1 and M2 over the period 1920–80. For both ratios, its results confirm Cagan's well-known findings on the importance of deposit yields and tax rates. Unlike Cagan, who did not attempt to construct a proxy variable for the ratio of currency-intensive to total transactions, the ratio of spending on nondurables and services to total bank clearings was used in this study as such a proxy and found to be a better explanatory variable than expected real per capita income. At times, movements in this variable were an important contributor to changes in both money ratios. In addition, movements in the ratio of teenagers to total population, a variable not considered by Cagan, had a role in explaining the 1961–73 rise in C/M1.

The finding that alterations in the tax rates change the money multiplier, and hence the money stock, in the opposite direction serves as an offset to the argument advanced recently that tax cuts, since they may increase the demand for money as well as (more familiarly) the rate of desired spending, may actually be contractionary.