Unless there are uniform required reserve ratios, shifts in deposits from one reserve requirement class to another affect the amount of reserves that banks are required to hold, and thus necessitate that Federal Reserve policy actions offset any undesired change in the overall issue of deposits or bank credit that might result. It has been demonstrated that such effects are readily estimated so that the Federal Reserve could control the monetary supply process in spite of shifts of deposits from one requirement class to another. I also append an article that demonstrates that the average required reserve ratio is a highly predictable magnitude based on the predictability of the deposit distribution among required reserve classes. 3 One particular aspect of the existing requirements machinery is a likely cause of part of the poor monetary control record of the Federal Reserve. It is that requirements for a current settlement week are based on the weekly average of deposits subject to reserves two weeks earlier. Consequently, required reserves are absolutely fixed in the current settlement period. Even though banks can carry forward deficits up to 2 percent of the requirements, if the Federal Reserve were so oriented, it conceivably could determine an amount of bank reserves too small to meet the reserve requirements. But in point of fact the accommodating posture of Federal Reserve open market operations typically works in just the opposite way. Banks can generate deposits and credit 3The Required Reserve Ratio for Member Banks," Bulletin of Business Research, The Ohio State University, Vol. XLVIII, No. 9 (September, 1972), without a requirements constraint in the current weekly settlement period. The Federal Reserve tends to validate the change two weeks hence by supplying sufficient reserves to meet requirements. No single action would do more to endow the Federal Reserve with the capability of controlling monetary growth rates than the restoration of reserve requirements based on current deposits.* The only factor more important would be a willingness of the Federal Reserve authorities to use their powerful instruments for such a purpose. * actually lagged one day. By WILLIAM G. DEWALD Federal Reserve Policy and the Money Supply more emphasis to limiting erratic variation in the rate of growth in the money supply or even to providir, a steady rate of monetary growth to keep pace with real output. This article considers, first, the question whether recent policy of the Federal Reserve has been aimed at controlling the growth of the money supply. Second, it considers whether the Federal Reserve could control the money supply if it wished to. The conclusion briefly is that the Federal Reserve has not tried to control short-term variation in monetary growth, but it could if it tried. A subsequent article will consider the probable consequences of a policy designed to provide steady growth in the supply of money. The uncertainties of the nation's current economic situation, accentuated by the long deadlock between the Administration and Congress over federal tax and spending policy, give new prominence to the nation's monetary policy. Monetary policy actions by the Federal Reserve to heighten or ease credit conditions have long been a center of controversy. The record suggests that actions by the Federal Reserve have often been mistaken or ill-timed. Many critics argue that the monetary authorities have tried too hard to limit short-run fluctuations in interest rates, and have measured their policies in terms of changes in interest rates. These critics hold that the Federal Reserve could make a greater contribution to economic stability by giving Has the Federal Reserve Controlled Monetary Growth? To the question whether the Federal Reserve has consciously sought to limit variation in monetary growth, the answer is no. The evidence is that monetary growth has been very erratic. Money (defined as private demand deposits and currency in the hands of the public) increased 7.2 per cent during 1967. If that were the desired rate, there would have been a very strong tendency for weeks (Continued on page 6) 6 The Committee has tended in recent years in its directives to hedge its statements about money market pressures. For example, the Committee has directed that desired con-, ditions be attained subject to particular developments that might occur between meetings. Shocks related to Treasury financing, bank credit, money, and liquidity developments have been referred to in this way. The directive has been unclear about what precisely would have to happen to change desired market conditions and by how much. Federal Reserve Policy and the Money Supply (Continued from page 1) Observed deviations from average monetary growth over even longer periods give further evidence that the Federal Reserve does not control monetary growth. The average annual rate of increase in money was 2.6 per cent from 1957 through 1967. Relative to that historical trend, monetary growth accelerated to a 3.6 per cent annual rate from August 1962 through August 1965. This acceleration in the rate of monetary growth probably made sense in terms of economic policy goals, though it could have come earlier. But as the economy approached capacity utilization the monetary growth rate, rather than decelerating, accelerated further to 7.6 per cent from August 1965 through April 1966. Through the rest of 1966-during the "credit crunch"-there was no growth at all. As mentioned, monetary growth accelerated to 7.2 per cent in 1967. A similar on-again, off-again monetary growth is shown in money broadly defined to include commercial bank time deposits. The directives of the Federal Open Market Committee to the Manager of the Open Market Account in New York offer the best testimony of what it is that the Federal Reserve tries to do. The directives are usually phrased in such terms as reserve "positions" or "availability," money market "conditions" or "pressures." The key measure is net borrowed reserves (negative free reserves)-the arithmetic difference between member bank borrowings from the Federal Reserve and excess reserves. "Long experience has shown that any departure from a relatively steady ratio between bank credit expansion and the reserves supplied at Federal Reserve initiative sets forces into operation that tend to encourage bank credit expansion when free reserves exist and to restrain bank credit expansion when net borrowed reserves exist." Net borrowed reserves and market interest rates are correlated; and it is to one or both of these that the Committee usually refers. In the terminology of the Committee, easing conditions are measured by declines in interest rates or net borrowed reserves, while tightening or firming conditions are measured by the comparable increases. When conditions in New York differ from those elsewhere, the Manager may indicate that the "feel of the market" is tight, even though aggregate measures indicate the contrary. The Federal Reserve and the Treasury Answers to Questions from the Commission on Money and Credit, Englewood Cliffs, N.J.: PrenticeHall, Inc., 1963, p. 9. According to the record for the December 12, 1967, meeting, the Committee directed the Manager to conduct operations for the purpose of moving slightly beyond the firmer conditions that have developed in money markets partly as a result of the increase in Federal Reserve discount rates provided, however, that operations shall be modified as needed to moderate any apparent significant deviations of bank credit from current expectations or any unusual liquidity pressures. During the intervening period until the January 9, 1968, meeting, bank credit, estimated by total bank deposits, increased at a 3 per cent annual rate, but money narrowly defined increased at an 11 per cent annual rate. It is presumably not a coincidence that free reserves did decrease as directed and were widely interpreted as an indicator of tightening policy despite the fact that monetary growth had proceeded at such a rapid rate. A similar directive was issued by the Committee at its next meeting. During the following four weeks free reserves fell further; the rate of bank credit growth was about the same; and monetary growth proceeded at about a 1.5 per cent annual rate. As this episode illustrates, it is clear that the Federal Reserve has not tried to control monetary growth, at least not directly. The proximate targets at which the Federal Reserve has aimed have typically been achieved. The timing of changes in desired money market conditions reveals that the Federal Reserve has been quick to pick up evidence of a need for action. But actual policy actions and money supply changes have often been in the wrong direction and of inappropriate magnitude. The policy of manipulating money market conditions or interest rates can be likened to a baseball player who can't hit curve balls. The policy is satisfactory if market conditions are at an equilibrium associated with achievement of objectives. But otherwise, when the economy throws curves, tardy adjustments in desired money market conditions lead to strikeouts, as the monetary managers swing behind the economy, where it was rather than where it is. Could the Federal Reserve Limit Variation in the Rate of Generations of American university students have learned that Federal Reserve open market operations can Federal Reserve Bulletin, March 1968, p. 306.. BULLETIN OF BUSINESS RESEARCH The Ohio State University JUNE be used to control bank reserves and other money that is issued by the Treasury or the Federal Reserve. This socalled "high-powered" or "base" money in turn has been interpreted as the cornerstone on which the money supply depends. The quantity of money is determined within a supply and demand or market framework. But the operation of this market process is subject to important policy constraints, including the amount of base money and the legal requirements imposed on banks to hold base money. A number of recent studies have examined the determination of the quantity of money within a market framework. In general, they show that more than 80 per cent of the quarterly variation in the money supply can be traced to readily identifiable market forces. But these statistical results are not altogether relevant from the point of view of actual monetary control. It is not necessary to fix open market policy or reserve requirements over an entire three-month period, as is implied in the use of quarterly models. The Federal Reserve obtains weekly money supply statistics with a lag of one week. It is promptly alerted to significant deviations in monetary growth from the desired rate. The Federal Reserve must of course take into account various non-controlled factors that affect the supply of or demand for base money. It presently makes day-to-day and week-to-week projections of likely changes in these 1968 non-controlled factors. A large part of the variability in non-controlled factors that affects average bank reserves and other base money could readily be offset over a week or two by open market operations of sufficient magnitude. Changes in the ratio of the money supply to the volume of bank reserves and other base money are accountable to changes in the distribution of money among deposits subject to different reserve requirements, between monetary and non-monetary deposits, between base money reserve holdings of banks and currency holdings of the public, and, finally, between bank required and excess reserves. These changes reflect both supply and demand factors in the money market. There is a relatively strong seasonal pattern in variation with respect to some of these noncontrolled distributional factors; and there is knowledge with respect to their response to market interest rates and to spending. Though non-policy factors are important, a large part of the quarterly changes in money are accountable to changes in reserves (and other base money) and in reserve requirements. At but one remove from the money supply, more than two-thirds of the variation in changes in net deposits of member banks over half-monthly periods was accountable to changes in bank reserves, changes in required reserve ratios, and predictable changes in distribution of deposits subject to different reserve requirements. (Continued on page 8) |