| Sumario: | A bank's demand function for excess reserves is derived using inventory theory. In the model, banks hold excess reserves as a means of reducing the cost of meeting their reserve requirements in a world in which they are faced with random reserve flows and transaction costs. The resulting demand curve is kinked at very low interest rates (estimated to be between 0.3 and 0.5 percent for the Treasury bill rate). This kinked demand curve offers an explanation of the large accumulation of excess reserves during the 1930s that compares favorably with alternative hypotheses. The difference in behavior is also consistent with the adjustment cost. hypothesis. Three things should be noted about the Canadian banking system during the 1930s. First, there were fewer than twenty banks in Canada, each having many branches and serving a wide area. Consequently, their adjustment costs should be similar in magnitude to those of the New York banks. Second, prior to 1935, the Canadian banks were not subject to legal reserve requirements. Instead, an informal reserve requirement of approximately 10 percent was enforced by the Canadian Bankers' Association. This was a very flexible requirement that apparently had to be met only on the last day of each month. Morrison notes that the banks "window dressed" their month-end reports. The flexible reserve requirements in Canada certainly lowered the yield from holding excess reserves. The kinked demand curve hypothesis has the advantage over the shock effect and inertia effect hypotheses in that policy makers can predict an accumulation of excess reserves in response to low interest rates much more accurately than they can predict an accumulation due to a psychological reaction on the part. of bankers.
|
|---|