PERCENT is a lack of will."51 Certainly the will to innovate must be included along with courage and judgment if one is to justify the use of discretionary monetary policy. APPENDIX The period chosen for Figures 2 and 3 represents the expansion phase April, 1958-May, 1960 (the initial and terminal dates coinciding with the trough and peak for the subcycle as measured by the National Bureau of Economic Research's business cycle reference dates). The monthly spreads were obtained by subtracting the aver age monthly rediscount rates of all the Federal Reserve Banks during this period from the corresponding average yields on three-month Treasury bills. Such spreads were then correlated with monthly averages of daily borrowing by member banks. The numbers 1-26 found in the figures denote the chronological sequence such that number 1 stands for the initial month of the expansion phase, or April, 1958, while number 26 represents the terminal month, or May, 1960. RELATION OF MEMBER-BANK BORROWINGS TO LEAST-COST SPREAD, APRIL, 1958, THrough May, 1960 Monthly Averages of Daily Figures 1,400 THE HORIZONTAL LINES REPRESENT THE ARITHMETIC MEANS OF 1,200 1,000 800 23 14 13 .25.24 600 400 7 200 3.1 42 -1.0 -80 -.60 RELATION OF MEMBER-BANK BORROWINGS TO LEAST-COST SPREAD, THE HORIZONTAL LINES REPRESENT THE ARITHMETIC MEANS OF BORROWING MEANS - APRIL 1958 THROUGH MARCH 1959. 1,000 800 600 .26 12 The technique employed in fitting the relationship between spread and borrowings is rather crude and should be considered only as a first step in fitting the function. Nevertheless, it has the basic merit of minimizing any preconceptions concerning the underlying shape of the curves while, at the same time, suggesting their general nature. The method is the relatively simple one of computing the arithmetic means of the dependent variable (discounts and advances) for given intervals of the independent variable (least-cost spread), the hori zontal lines to be found in Figures 2 and 3 representing such arithmetic means. In Figure 3 the solid horizontal lines represent the arithmetic means of borrowing for the period April, 1958-March, 1959, while the dashed horizontal lines are the arithmetic borrowing means for April, 1959-May, 1960. So far as the use of horizontal lines to represent the means of the dependent variable for given intervals of the independent variable is concerned, it should be remembered that regression equations are basically only polynomials fitted to the means of arrays. 12. THE STRUCTURE AND USE OF VARIABLE BANK RESERVE REQUIREMENTS* By NEIL JACOBY BECAUSE ECAUSE the National Currency Act of 1863 contained the first federal legislation on commercial bank reserves, a volume commemorating its centenary is an appropriate place in which to assess the present structure of legal reserve requirements in American banking, and the use of variable reserve requirements as an instrument of economic policy. The Act with its subsequent amendments was, in fact, a landmark in the evolution of reserve legislation. Unlike most of the then-current state laws which required reserves only against notes in circulation, it required minimum reserves to be held in lawful money against deposits. It also introduced into federal law the principle of classifying banks by their geographical location for the purpose of determining minimum reserve requirements—a principle later embodied in the Federal Reserve Act of 1913.1 *The author thanks his colleagues Karl Brunner, Frank E. Norton, and Allan R. Drebin for valuable comments on the initial draft of this essay, and absolves them from all responsibility for the final product. 1 Up to December, 1960, when legal reserve ratios of central reserve city banks and reserve city banks were equalized, different ratios were required to be maintained against demand deposits by banks in each of the three types of geographical location, under the Federal Reserve Act. This Act also introduced lower legal reserve ratios for time deposits, which up to 1913 had carried the same reserve ratios as demand deposits. Legal reserve ratios were fixed by statute up to 1933, when the Federal Reserve authorities acquired emergency powers to vary them with the approval of the President. The Banking Act of 1935 granted the Federal Reserve authorities permanent administrative authority to alter legal reserve ratios within wide statutory limits, without Presidential approval. A short history of the evolution of legal reserve requirements in the U.S. banking system is contained in Frank E. Norton and Neil H. Jacoby, Bank Deposits and Legal Reserve Requirements (Los Angeles: Division of Research, Graduate School of Business Administration, University of California, Los Angeles, 1959), chap. ii. I. THE PREMISES OF THE ANALYSIS Our purpose is not to trace the history of legal reserve requirements, nor to discuss the basic question whether it is desirable to have reserve requirements at all. Our aim is restricted to analyzing the present basis of reserve requirements, and presenting an alternative basis which would enhance the contribution of monetary policy to an improved performance of the U.S. economy. Herein, we accept the principle of fractional reserve banking, and reject both of the extreme alternatives of 100 percent reserve and zero reserve requirements. If one agrees that there are no clear net social advantages to be derived from a change in either direction from the fractional reserve principle, the relevant issue is how to make that principle serve the public interest better. We assume that the basic desideratum is to design an institutional apparatus that will transmit changes in monetary policies to the economy promptly and predictably. An efficient apparatus is one that gives the monetary authority maximum control of the money supply in the hands of the public. Stated more precisely, an efficient institutional apparatus will minimize both the time required for a response, and the probability distribution of possible responses, of the money supply to any given change in the monetary base by the Federal Reserve authorities. The only theoretical justification for a legal reserve requirement is that it increases the "controllability" of the money supply in this sense. The basic premises of our analysis are three: First, that monetary policies constitute an important set of instruments for the attainment of American economic goals. Second, that along with open-market purchases and sales of securi ties and regulation of rediscount privileges, variation of legal reserve ratios by the Federal Reserve authorities can make a unique contribution to good monetary policy. Third, that the selection of the proper basis of minimum legal reserve ratios for commercial banks will sharpen this instrument of monetary policy. 2 All changes in institutions impose social costs, and changes should be made only when the social gains demonstrably exceed the costs. For reasons that cannot be discussed here fully, it is believed that there are no net gains to be made by shifting the American banking system either to 100 percent reserve requirements or to zero requirements. |