ABSTRACT
That the Federal Reserve did not add enough reserves to the banking system through open market operations to stem the recurring banking crises of the early 1930s and prevent a decline in the money supply is widely acknowledged. But banks should have been able to obtain ample emergency credit through the discount window. This paper explores whether the decline in discount window borrowing in the early 1930s resulted from a change in Federal Reserve credit policy. The methodology employed in the paper is to test for a structural break in the demand for borrowed reserves using monthly data from New York City, Other Reserve City, and Country member banks from the interwar period. The paper presents evidence that a large structural shift in the demand for borrowed reserves occurred during the late 1920s and early 1930s that is consistent with an increase in the perceived administrative costs of borrowing. Specifically, it finds that member banks borrowed substantially less during the Great Depression than they had in the 1920s given spreads between market rates and the discount rate and the cost of liquidating secondary reserve assets (bonds). The estimates draw on a new monthly data series on bid-ask spreads on corporate bonds as measure of the transactions costs involved in raising funds by selling assets. These results support statements by Federal Reserve officials at the time that they had, indeed, tightened credit standards for borrowing from the discount window. Concerns about the solvency of borrowing banks and the availability of eligible collateral appear to have been behind the tightening of credit policy.
Brian Gendreau