THE STRUCTURAL SHIFT IN THE DEMAND FOR MONEY

Abstract

RECENTLY several articles have provided evidence that a structural shift occurred in the market for savings deposits during the early 1960's. Specifically, Modigliani [12] demonstrated that prior to 1962 the commercial bank time deposit rate did not affect the demand for nonbank deposits nor did nonbank rates affect the public's demand for time deposits at commercial banks. Bank and nonbank deposits apparently became strong substitutes only after 1961. Slovin [18] found that this shift in structure also characterized the deposit rate setting behavior at both bank and nonbank institutions. The explanation for this structural shift may be that the- low level of Regulation Q ceilings prior to 1962 may have seriously curtailed the ability of commercial banks to compete for savings deposits during the 1950's. During 1962 Regulation Q ceiling rates on time deposits were raised sharply.1 This triggered a substantial rise in commercial bank time deposit yields and, as a result, the differential between the deposit rate at savings and loan associations, RSL, and the time deposit rate, RTP, narrowed dramatically. The differential between these rates is graphed in Figure 1. Overall, this evidence suggests that the pattern of substitution between various types of liquid liabilities has changed over the postwar period. During the 1950's, interest sensitive funds were apparently diverted into nonbank intermediaries instead of commercial banks. The substantial rise in Regulation Q ceilings in 1962, however, significantly enhanced the ability of commercial banks to compete for deposits. It is our contention that these financial developments in the early 1960's induced a shift in the structure of the public's demand for money. Our approach is to disaggregate the demand for money into the demand for demand deposits and the demand for currency. First, we consider a standard demand for money specification and investigate whether there has been a structural shift in the pattern of substitution between money and near monies. Second, we reestimate the money demand function in order to incorporate the effect of developments in the savings market.