The full text on this page is automatically extracted from the file linked above and may contain errors and inconsistencies.
Authorized for public release by the FOMC Secretariat on 2/25/2020 FEDERAL RESERVE BANK OF NEW YORK Confidential March 21, 1956 To: All Members of Federal Open Market Committee, and All Presidents not now serving on Federal Open Market Committee Attached is a memorandum presenting my views of the experience of the Federal Open Market Committee with the so-called "continued operating policies", which it adopted three years ago as a result of the study of the Governent securities market and our operations in that market made by a special Ad Hoc subcommittee. My statement goes to the substance of the questions involved. The beginning of a new year of the Federal Open Market Committee seems to me to be an appropriate time to review these operating rules for the purpose of determining, by study and discussion, whether the hopes concerning the Government securities market, on which they were based, have been realized and whether they are actually serving the purpose for which they were intended, namely, promoting the most effective use of open market operations as an instrument of credit policy. I would hope that this review could begin promptly, as was suggested at our last meeting, and that it could be carried to a conclusion without more delay than is necessary for full study and debate. I would suggest that the initial study be carried forward by a working committee, such as I have already suggested be named to review the relations of credit policy and debt management, one part of this whole problem. I would hope that the discussion and debate could take place in a clearer atmosphere than three years ago, when revolt against "pegged" markets, and uncertainty as to how well individual members of the Federal Open Market Committee were meeting their responsibilities without the adoption of restrictve rules of operation, tended to obscure the economic issues and the specific question of market organization and techniques which were involved in the actions taken. We are now pretty well out from under the shadow of pegged markets, and we have pretty well resolved the practical problems of participation, by all members of the Committee, in the formulation of open market policy and of keeping individual members informed currently of policy in execution. We should be able to make an objective study of the resets of the actions taken by the Committee three years ago, and to determine what changes may be necessary or desirable. Yours faithfully, Enclosure Allan Sproul, Vice Chairnman, Federal Open Market Committee. Authorized for public release by the FOMC Secretariat on 2/25/2020 CONFIDENTIAL March 21, 1956. APPRAISAL OF CERTAIN "CONTINUING OPERATING POLICIES" OF THE FEDERAL OPEN MARKET COMMITTEE The Federal Open Market Committee has now been operating for three years within the confines of certain rules growing out of the report of the Ad Hoc Subcommittee on the Government Securities Market, dated November 12, 1952. These rules have been embedded in the three "continuing operating policies" renewed by vote of the Committee at its annual organization meeting each year. I do not dissent from the first of these "policies" - that we are not supporting any pattern of prices and yields in the Government securities market and that our transactions in that market should "effectuate the objectives of monetary and credit policy". I think the wording of this statement could be improved but, no doubt, each individual member of the Committee would have his own way of saying what we intend to say, and I suspect that the compromises represented by the present wording do not put more strain on me than on some other members of the Committee. I have voted "yes" on this "continuing operating policy" as it has come up for renewal, and I do not intend to discuss it further here. I have felt compelled to vote "no" on the other two "continuing operating policies", however, at first because they did not seem to me to serve what I conceive to be the purposes of central banking in general, and of open market operations in particular, and more recently because of what I have seen of their application in practice. Now I want to try to put before you the basis for my views and opinions. The two "policies" to which I object incorporate three rules which I think should be carefully reexamined: (1) that all of our operations at all times - except for the correction of disorderly situations in the market - should be in short term securities, which for all practical purposes has meant Treasury bills; (2) that we should not intervene in the market at times of Treasury financings by dealing in maturing issues, "when-issued" securities, or outstanding issues of comparable maturities to those being offered for exchange; and (3) that we should not engage in so-called "swaps". I think that things which have been said and done, during the three year period in which these rules have been in effect, have modified the original fundamentalist approach to the problems with which they are concerned. It has been stated more than once in the course of our discussions that no permanent pattern of open market operations has necessarily been established, and that no promise has been made to the market that the present pattern of operations will always be followed. These rules are experimental in nature and may be transient in character. In the circumstances, after the first series of debates concerning these operating rules had been concluded, it seemed to me to be desirable, and potentially useful, that they be submitted to study, over a period of time, so that we might have some idea as to how they were working and how they might be improved. After three years of attentive observation, this memorandum represents my own response to this prescription of study and report. If I present my views FOR FILES V. J. Ogilvie Authorized for public release by the FOMC Secretariat on 2/25/2020 2 starkly, it is because I believe the Committee is entitled to know exactly what I think, not what I might find it necessary to accept. The basis for a closer meeting of minds, than we have had over the past three years, can be laid only if each member of the Committee will similarly address himself to the problem and if we then seek to reconcile our differences. Basis of Current Operating Rules The principal justification of our self-denying restraints on open market operations, which grew out of the Ad Hoc Subcommittee report, was that these prohibitions would permit a new and better market for Government securities to develop and, by improving the functioning of that market, would contribute to the effectiveness of open market operations in particular and of credit policy in general. This Government securities market of the future was to be one of greater depth, breadth and resiliency; a market in which prompt arbitrage would continuously produce a yield curve reflecting accurately the net balance of forces of supply and demand throughout the maturity range; and a market in which our operations could be confined to Treasury bills in appropriate volume without fear of adverse technical repercussions. Disregarding the question of whether the report of the Ad Hoc Subcommittee, and the subsequent action of the Federal Open Market Committee, placed too much emphasis on improving the mechanical functioning of the dealer market in Government securities and too little on the detailed problems that are encountered in making credit policy effective through open market operations, what can be said about our experience with the market over the past three years? It is my observation and opinion that there has been no improvement in the depth, breadth, and resiliency of the market during our three-year trial, and that in critical periods the market appears actually to have been shallower, narrower, and more brittle than at similar times in the past. What is Meant By Depth, Breadth and Resiliency? Here it is necessary to ask what is meant by depth, breadth and resiliency. In its "Findings and Recommendations" under the general heading "Structure of the Market" Ad Hoc Subcommittee said (in paragraph 36): "In strictly market terms, the inside market, i.e., the market that is reflected on the order books of specialists and dealers, possesses depth when there are orders, either actual orders or orders that can be readily uncovered, both above and below the market. The market has breadth when these orders are in volume It is resilient and come from widely divergent investor groups. when new orders pour promptly into the market to take advantage of sharp and unexpected fluctuations in prices." (Ad Hoc report, Flanders Hearings, p. 265) In preparing for these definitions the Subcommittee indicated that it was not concerned with the size or frequency of swings in prices, and there are indications that it expected and approved somewhat more erratic movements of prices in the market which it envisaged than had been our previous experience. Authorized for public release by the FOMC Secretariat on 2/25/2020 3 It evidently expected an active trading market, with the marginally active elements (the speculators in the best economic sense of the word) keenly responsive to changes in prices and expectations, with a wide variety of investor groups repre- sented, and with continuous price changes keeping all of them active and alert. In such a market, it was said, all investors and the System Open Market Account (confining its transactions to Treasury bills) could feel assured of doing business promptly and in the volume appropriate to their purpose. To paraphrase the Subcommittee, it defined market depth as a condition in which there is a waiting queue of potential buyers, and of potential sellers, at prices below or above those prevailing for existing Government securities at any given moment. This depth gains another dimension, breadth, when these queues include among them potential buyers and sellers from most or all of the important investor groups - commercial banks, savings banks, insurance companies, other financial corporations, trust and pension funds, non-financial business corporations, and individuals. And finally the market has resiliency when these standing queues are quickly enlarged, on one side of the market or the other, as sudden or unanticipated price changes occur. Presumably the Subcommittee intended to imply that these conditions would obtain through periods of credit restraint as well as through periods of credit ease, since no differentiation is made among various phases of credit policy in the report. I. What Has the Government Security Market Done During the Past Three Years? The Ad Hoc Subcommittee did well to recognize in advance that the depth, breadth and resiliency which it sought should not be equated with any tendency for price fluctuations to narrow. In fact, over the past three years, the frequency of large daily fluctuations in prices for Government securities, most of the time, has been considerably greater than in the years before our present operating rules were adopted. Nor can this change be attributed merely to the suspension of pegging operations after the accord of March 1951, but this is a point which need not be labored. What matters, so far as the aims of the Ad Hoc report are concerned, is that there have been wide and frequent price fluctuations, both daily and of long duration, during the period that the new operating rules have been in effect. This is the kind of a market in which long and varied queues of buyers and sellers were supposed to develop and flourish, sustained by the announced withdrawal of the System Open Market Account from all but the short term sector of the market (in practice, all but Treasury bills). Yet the queues which were to supply depth, breadth and resilience to the market have not appeared. So far as statistical evidence is concerned, we should have expected to find signs of greater depth, breadth and resiliency in the actual volume of transactions passing through the dealer market. And even if that measure were to prove inconclusive, we should have expected that the most sensitive of all elements in the Government securities market, the dealers themselves, would have given evidence of increased depth, breadth and resiliency by greatly enlarging their own positions, if not at all times, at least on balance over time. Authorized for public release by the FOMC Secretariat on 2/25/2020 4 Volume of Transactions With the necessary reservations pertaining to the available data, the best that can be said is that, so far as the dollar volume of transactions passing through the market is concerned, there has been no basic change in either direction since the System Account began operating under the present rules. Comparisons have been made between periods of credit ease (and rising prices), before and after the present rules came into force, and between periods of credit restraint and declining prices, before and after. To avoid the immediate postwar artificialities, only data from 1948-51, 1952, and 1953-55 have been used. During our three year-test period there has been a drop in volume of trading for longer term securities which might suggest, at first glance, that the market has actually deteriorated under the present rules. But I think that would be an unwarranted assumption. It seems to me to be more accurate to say that, after adjusting the volume data to correspond with changes in various maturities outstanding, the shrinkage of trading volume in the longer area has not been much more than might 2/ be considered proportional to the decline in outstanding issues of longer maturity. So far as the short term area is concerned, the volume of trading has generally been as large as before and, at times, it may even have been slightly larger than in earlier comparable periods. But that is not all that we were promised. The recommendations of the Ad Hoc Subcommittee were based on the claim that adherence to its rules would give new dimensions to the market, once our own operations were restricted to short term securities (in fact, Treasury bills) and the market given assurances to that effect. So far as dealer positions are concerned, the record is no better. There has been a general tendency for dealer positions to decline during the three years of testing. In periods when one might have expected long positions to increase, particularly in the longer term maturities, average dealer holdings in trading have often been strikingly smaller than those held before the present portfolios 1/ In analyzing the data three reservations must be kept in mind. First, analysis of aggregate volume or total dealer positions, when subdivided by maturity groupings, should make allowance for changes in the amount outstanding in the various maturity categories over the period studied. Second, totals for volume or for positions must be reexamined to determine whether or not they conceal significant differences among dealers - whether there is evidence that some are performing as might be expected in markets of greater depth, breadth and resiliency, while others are obscuring the statistical measurements by behaving differently. Third, one of the dealer firms, whose volume and position apparently would now place it somewhere among the largest ten, although probably below the top three or four, does not submit the daily reports on volume and position that are provided by the other dealers. The data, therefore, are not complete for the market as a whole. The Committee needs no reminder, I am sure, that the data we do have are kept in the closest confidence. We have scrupulously avoided any public reference to, or use of, even the aggregate totals for dealer volume and position. 2/ The conclusions are the same whether you look at these figures including or excluding the direct effect of System transactions on dealer volume. 3/ In addition to trading portfolios, some dealers have at times carried in their positions blocks of bonds which are, in fact, segregated in an investment acThese are not available for trading as they are held for the purpose of count. supplementing dealer income by taking capital gains, after they become eligible for capital gains treatment. Such investment accounts have been excluded from the position data which underlie this study, but their inclusion would not change the general results. Authorized for public release by the FOMC Secretariat on 2/25/2020 5 rules were adopted, even after allowance is made for variations in the amount of longer term securities outstanding. The change has not been so pronounced with respect to short sales of securities, and in 1955 there was a sharp increase in gross short positions as dealers more frequently hedged any holdings in which they happened to belong. That could scarcely be interpreted, however, as evidence of greater depth, or breadth, or resiliency. On the whole, and unfortunately, there has been a decreasing readiness on the part of dealers to make markets for their own account; that is, to buy for and sell from their own positions. The Performance of Government Security Dealers In addition to the evidence provided by statistical reports, there is another approach which helps to provide some clue to the depth, breadth and resiliency of the Government securities market during the past three years. How have the dealers performed, when confronted with specific orders to buy or sell? In judging that performance, we have not only the great variety of Federal Reserve Bank transactions (for Treasury, foreign, and member bank accounts, as well as for the System Account), but also the mass of information provided continuously to the Trading Desk by the traders at dealer firms. These comments on business done, and of possibilities for doing business that have been missed or turned aside, are not systematically logged at the Trading Desk. (There is no attempt at reconstructing any individual dealer's purchase and sales transactions or the sources of his business from these various spot reports.) However, all actual transactions of any size are noted, for they are relevent in aiding the analysis of market conditions day by day, and the senior officers of the dealer firms, who attend the morning dealer conferences, customarily mention the major transactions they have seen as market iifluences. These latter comments almost always include general statements on the nature and magnitude of the trading that has recently been going on in the market as a whole. From all of these sources, some qualitative conclusions may be drawn. In doing so, it should be remembered, of course, that no one else in the market is in a position to see as full a cross section of all that has been going on as the Trading Desk of the New York Bank. There is, therefore, no complete check on the Neveraccuracy of the information which has come to me from that vantage point. theless, it seems to me that it is information which has some claim to the attention of the other members of the Federal Open Market Committee. On the basis of all sources of information available, it appears clear that day-by-day performance of the dealer market for Government securities now shows somewhat less depth, breadth and resiliency than it had when the bulk of the System's holdings were, in effect, at least theoretically available to the trading market. There has been a change in the trading habits of the market, a change that runs from With some frequency, dealers simply the shortest through to the longest sectors. pull themselves out of the market for particular securities, and even more often show or no hesitation in making it clear that they are extremely reluctant to make little This sharp divergence from the kind of dealer behavior envisaged in the Ad markets. It should not be attributed Hoc Subcommittee report undoubtedly has many causes. wholly nor, perhaps, even in large part, to the fact that the System no longer funcBut I do tions at times throughout the maturity range of Government securities. think enough time has now elapsed, since the present operating rules were adopted, to demonstrate that a complete renunciation of System intervention, in the market area beyond one year, has not worked the improvement in dealer performance that was expected if the recommendations of the Ad Hoc report were adopted. Authorized for public release by the FOMC Secretariat on 2/25/2020 6 Even in the Treasury bill market, changes have occurred which would seem to be contrary to earlier expectations. When the Ad Hoc Subcommittee prepared its report it could say that the conditions of depth, breadth and resiliency "are most nearly characteristic of the market for Treasury bills, but even in that market reactions have been sluggish on more than one occasion since the accord". (p. 265) An Ad Hoc Subcommittee, writing today, would have to say that market reactions even for Treasury bills have been characteristically erratic and often sluggish, on the part of many of the dealers over much of the time since our present rules were adopted. There have been few occasions when we have not found some dealers - perhaps two or three, more often five or six, and at times ten or twelve, of the 18 regularly presumed to be making markets in Treasury bills - who simply decline to make markets in whatever Treasury bills may be involved in our transactions. There had been some hope that the increase in outstanding bills during the summer and fall of 1955 might help this situation, but so far it does not seem to have had any effect. In the other parts of the market, out beyond Treasury bills, there has been a continuing tendency for the regularly quoted dealer spreads to widen, and for quotations to be "good", if at all, only for small transactions. Only two of the dealers (although at times one or two more join them) still make a point of trying to give a bid or an offering for long bonds, whenever requested, for lots larger than $1 million. Even they, however, will not customarily perform for large amounts on their currently quoted markets, generally widening their spreads by several thirty-seconds in the course of assuring investors that they will perform for the larger amounts involved. It is now unusual, indeed, for an investor to think of approaching the market with a block of Government securities as large as $15 million in the intermediate area, or $10 million in the longer term area, in the expectation of finding a dealer ready to make an immediate market "on the wire." The usual reaction, most of the time, would be for the dealer to do a million (or less) for his own account, and ask to have the balance of the block left with him, in order to see what could be worked out. Only if a dealer happened to have in hand an order which could be filled on the opposite side, or if he would be interested in filling out a hedge, would he, as a rule, be ready to do business immediately on presentation of a block of this size. In other words, the thinness of the intermediate and longer term market for Government securities which concerned the Ad Hoc Subcommittee and, presumably, the Federal Open Market Committee, three years ago has in no sense been abated. The Ad Hoc Subcommittee also expressed its concern over the thinness of these longer markets by stressing the tendency apparent in 1952 for dealers to become brokers, thereby ceasing "to perform their normal functions of making markets, rendering independent advice to customers, and engaging in normal If anything, the tendency for dealers (Cf. page 300) arbitrage transactions". True, in to act as brokers has increased over the past two or three years. their capacity as brokers, dealers have at times accumulated stacks of potential orders which might have been interpreted as signs of some sort of depth in the market. But these accumulations have for the most part, so far as anything outside the short term area is concerned, consisted of so-called "tax swaps", which account for a very large part of the total trading volume, but can scarcely be considered Authorized for public release by the FOMC Secretariat on 2/25/2020 7 representative of the kind of debt that the Federal Open Market Committee wished to encourage. And even in the case of the tax swaps, it is significant that most dealers most of the time act only as brokers, preferring to keep these potential orders in a file, awaiting the possibility of working them out with other customers who may appear interested in a "swap" going the other way. There has been limited willingness on the part of dealers to take any substantial part of the tax swaps, one end or the other, into their own positions, until the swaps could be merely "passed through" their positions as various buyers and sellers could be "paired off". Moreover, there is now less readiness than formerly for dealers to make good their own quotations to other dealers, on the wire. Many of the previous arrangements, by which one dealer would agree with others always to do 100 bond lots at either side of his posted quotations, have been suspended. As a counterpart of this, the dealers themselves have come to rely more and more upon the inter-mediation of brokers - small firms, which make their living by running from one dealer to another, usually with relatively small or odd lot bids or offerings. This is not surprising. It seems only reasonable to expect that a variety of techniques will inevitably be needed in the Government securities market; and greater use of the broker mechanism is to be expected in periods of rising rates when markets generally become thinner. What can be said is that the present operating performance of the market, in handling specific transactions, provides pretty convincing evidence that there has been no improvement in the kind of depth, breadth and resilience that the Ad Hoc Subcommittee report thought it could foresee in the Government securities market if our present operating rules were adopted. II. What Should the Government Security Market Be Expected To Do? In my judgment, these expectations were unrealistic. They envisaged a The Government security market which does not exist and which we can't create. of in terms operations open market think of is for us to important thing now what we want credit policy to accomplish, rather than as something to be used in trying to create a new kind of Government securities market, in the hope that the achievement of credit policy objectives would thereby be made easier. The Government securities market has developed, in response to the nature of demands upon it, as an over-the-counter market, in which each trans- action is subject to negotiation. by investors, large and Marketable Government securities are not held small, with the expectation of trading them from day to day. Unlike the smaller stockholders who trade on the fluctuating prices of equity shares, most small investors in Government securities have been attracted 4/ One could argue that the volume figures for the past three years are not really comparable with earlier years, since they have been inflated by the relatively recent emergence of tax swaps as a dominant new element in the intermediate and long term market. Tax swapping mushroomed after Government securities had been trading for some time below par, and was given further stimulus as banks' holdings of Treasury bills shrank and as they acquired proportionately more fixed interest-bearing Government securities. Although dealers are conscious of this stimulus to their activity, and have often stated that threequarters or more of their intermediate and long trading for weeks at a time has been of this character, there are no reliable data on these transactions as such. Authorized for public release by the FOMC Secretariat on 2/25/2020 8 to the non-marketable savings bond. The larger investors with holdings of marketable securities, which they may have occasion to trade from time to time, are mostly banks and other institutional investors and, in relative terms, there are not very many of them. To be sure, there are enough smaller holdings of banks and others to give rise to a fair volume of odd lot transactions, but the bulk of the business in Government securities is done by professional investors engaged in moving blocks of some size. It is to meet the needs for transferring such blocks of securities, among investors with highly specialized and often finely calculated requirements, that the over-the-counter market for Government securities has developed. There is not the kind of supply of or demand for marketable Government securities that would call for a specialist in each of the outstanding issues or groups of issues on a trading floor, such as that of the New York Stock Exchange. Apart from a few specialized concerns and some of the institutional investors, people interested in speculation apparently choose instruments that contain some degree of credit risk, and imply judgments about some particular borrower - perhaps because of a realization that successful speculation in Government securities would involve much more knowledge and skill in judging the broader influences affecting the economy as a whole. Whether for these or other reasons, the Government securities market is principally peopled by large holders who, when they do move, are usually keenly aware of the possible influence of their own action on the market, and who are sensitive to price changes in 32nds, not in halves or full points. And the intermediary - the Government security dealer - must do more toward bringing supply and demand together, for any particular Government security, than is done by a specialist maintaining a book of bids and offers on an auction market. The Government securities market by logic and by evolution, therefore, The System should accept that fact is a negotiated, over-the-counter market. and attempt to fit its operations into the functioning of such a market. We should not assume that the existing demands for a given security on a given day will necessarily be so large that the mere posting of a substantial offering, at a price on or close to the current market, will necessarily "move" the offering, or that arbitrage between different sections of the market will always take place smoothly and quickly. There is no standing queue of buyers and sellers in the Government securities market, nor is there any reason necessarily to expect that there should be. It is not reasonable to expect that any considerable number of investors, who characteristically hold large blocks of Government securities, will become speculators and look upon Government securities as a good field for conMost investors now appear to take quite an opposite tinuous trading profits. view of Government securities trading, even after three years of the wider price fluctuations to which our present trading rules may have contributed. Yet until Government securities buyers and sellers begin lining up in queues to buy or to sell at prices above and below the day's prevailing price, we can never, by definition, achieve the depth, breadth and resiliency promised by the Ad Hoc Subcommittee report. Some dealers, with an interest in generating market activity (particularly in the face of relatively high borrowing costs that prevent a regular profit on the carry of securities), may have hoped for such a market, but investors have shown no inclination to reconstruct the market in this image. Authorized for public release by the FOMC Secretariat on 2/25/2020 9 What about the dealers themselves? The Ad Hoc Subcommittee had expected that they, too, would be joining the queues of buyers and sellers. In fact, the summary of dealer interviews included in the report said that if the "..dealers, generally, stated they would be conAd Hoc rules were adopted: siderably more willing to carry positions and operate more actively in the long-term sector of the market" (p. 299). The evidence on that is disillusioning, but not an indictment of the dealers' behavior. Can dealers really be expected to do anything different, by and large, from adding somewhat to long positions when they expect prices to rise, or running down their positions and going short when prices are expected to decline, or hedging themselves (with only limited and offsetting holdings of both longs and shorts) when the outlook is particularly uncertain? That is what they have continued to do during the past three years, and that is the way they probably will behave in the future, whether we change our rules or not. What I am suggesting in this review of market traits and practices is that the System should not have based its operating rules on a market which does not and, I believe, cannot exist, but should accept the market for what it is, and must be, and make its rules only for its own purposes of credit policy. In doing so, I suspect that a place will then be more clearly recognized for System operations outside the short term area of the market. My own view, of course, is that an even stronger and more positive case can be made for such operations. If the System Account as a whole were to be potentially available for trading in the market, under the close direction of the Federal Open Market Committee, I think we would find ourselves actually stimulating some growth in dealer positions, and certainly in trading volume. What we have done under the current operating rules is to remove entirely from the potential trading market the largest blocks of holdings of many of the outstanding issues. Freeing these blocks for use at suitable times and under suitably cir- cumscribed arrangements, would seem to me to provide more "stock in trade" for the market, and aid in the consummation of other transactions being attempted in the market. But I see no way of proving what is admittedly a supposition, and I would not want to rest my case for some change in the thinking of the Committee upon mere hopes or expectations no matter how attractive they might be to me. III. The Functioning of the Money Market In seeking some revision in the thinking of the Committee, I do not suggest that the performance of the money market, in responding to changes in credit policy, has not been pretty good over the past three years. The position of marginal dependence of the market upon System releases of Federal Reserve credit has been well maintained, contributing to the effectiveness of policy during this period. The paradox in the situation, however, is that these happy results have been obtained while the Treasury bill market has behaved at times quite differently from what was envisaged in the Ad Hoc Subcommittee report. Other changes in the functioning of the money market, not fully visible or recognized at the time the Ad Hoc report was prepared, have had much more to do with the successful performance of the market than has the behavior of the Treasury bill market and the confinement of our operations to that market. Three matters of a fundamental nature deserve attention: (1) the shifting significance of the Treasury bill; (2) the money market role of the nonbank dealer; and (3) the System's procedures in making repurchase agreements. Authorized for public release by the FOMC Secretariat on 2/25/2020 10 The Shifting Significance of the Treasury Bill In 1952 when the Ad Hoc report developed its theory of the Treasury bill, stressing the supposed unique suitability of bills for central banking transactions, it was still possible to assume that the Treasury bill was the predominant money market instrument. Certainly it had been through much of the postwar period up to that time. In terms of the dollar volume of outstanding amounts, that is still true. But the nature of that predominance was changing at the time the Ad Hoc study was made, and changes have continued right up to the present. Most Treasury bills are now held by nonbank investors (corporate and foreign) who, taken as a whole, keep the bills they hold out of active circulation in the money market more or less permanently. Thus, the floating supply of Treasury bills available for money market adjustments, either by their corporate or foreign holders or by banks, is much smaller than the total volume outstanding. Along with this absorption of Treasury bills by corporations or foreign accounts, there has been a concurrent decline in the volume of commercial bank holdings of Treasury bills. While the current low level of bank holdings of bills may be attributed, in part, to the System's restrictive credit policy of the past year, such holdings were quite small even before that policy was adopted. The extent to which banks or others have relied upon bills to meet short run swings in bank reserve positions, or in the money market situation, has been greatly reduced. Some corporations and other large investors have turned to direct financing of Government security dealers, through repurchase agreements with them, in order to supplement reliance upon the Treasury bill itself as a short term lodgment for fluctuating balances. The same has been done by many of the banks outside New York City which were not formerly, and are not now regularly, in a position to make dealer loans on a call loan basis. And by far the greater volume of short term reserve adjustments by the banking system have come to be effected, not through Treasury bills, but through a combination of adjustments in short term street loans (including loans to Government security dealers), the averaging of reserve deficiencies and excesses, purchases and sales of other short and intermediate term Government securities, transactions in Federal funds and, of course, borrowing at the Federal Reserve Banks. Because so much of the demand for Treasury bills has been of a continuing character and unresponsive to money market conditions, much of the time the marginal influences, which operate to set the effective auction rate, have only weakly reflected current money market conditions in a sensitive, significant way. There is a tendency, of course, for the Treasury bill rate, in time, to go along with the prevailing degree of tightness in the money market and the prevailing direction of System policy, but the factors at work influencing rates have so many unique characteristics of their own that the Treasury bill rate has lost much of its short run reliability as a guide to current policy developments. Thus a part of the logical market structure developed by the Ad Hoc Subcommittee It is not now possible report has not fitted into the place intended for it. to think of the Treasury bill as a security which, by its consistent behavior, always accurately reflects the forces at work in the money market. The Money Market Role of Nonbank Dealers It may be argued that the unavailability of funds to finance dealer positions, at reasonable rates, has itself been a major deterrent to the development of depth, breadth and resilience in the Government securities market. The Authorized for public release by the FOMC Secretariat on 2/25/2020 11 high lending rates of the money market banks in New York and Chicago, particularly during the phases of restrictive credit policy, could be pointed to as the obstacle which prevented the full flowering of the kind of Government securities market which was talked about three years ago. There is a serious problem, or a collection of such problems, in the area of dealer financing. I strongly doubt, however, that the high rates of the New York banks (or others) have been a major factor preventing depth, breadth and resiliency, for two reasons: first, as has already been suggested, the nature of the market itself is the basic deterrent to realizing the ideals pictured by the Ad Hoc Subcommittee; and second, the dealers have developed a wide variety of other sources of loans at more favorable rates than those offered by the New York and Chicago banks. On a very rough basis, since we receive no data on sources of funds in the regular dealer reports, it appears that the dealer loans (on Governments) of the weekly reporting banks in New York City have consistently accounted for less than one-half, and generally much less than one-quarter, of the total funds borrowed against Governments by the reporting dealers. It also appears, from the accumulation of oral comments received, that much of this financing with "out-of-town" banks (excluding Chicago) and all of the repurchases and "buybacks" arranged with other financial and nonfinancial corporations, is done at rates below the posted lending rates In many cases, these other rates are below of the New York (or Chicago) banks. the discount rate. Other lenders are willing to receive less than the discount rate in return for avoidance of market risk in employing short term funds, in recognition of the convenience of tying maturities precisely to days of money needs and, in the case of banks, as a concession for avoiding limitations on the size of loans to any one borrower (repurchase agreements are in most cases considered as "sales"). There is an added consideration impelling dealers to locate funds outside New York. That is, such funds, since they must come in over the System's wire facilities, are automatically Federal funds and the dealer is thereby spared the expense, common in New York City bank borrowing, of having to purchase Federal funds over and above the proceeds received on a bank loan made in clearing house funds. It is nonetheless true that present dealer financing arrangements, even taken as a whole, are a drag on the market. It is my intention, again, to bring this situation to the attention of the New York financial community, and to encourage a study by the New York Clearing House Banks and the Government securities dealers looking toward its improvement. Help could come, too, at least collaterally, from a System-wide study of the Federal funds market, which I recommend. System Procedures in Making Repurchase Agreements The Ad Hoc Subcommittee report, in Appendix C on "Ground Rules", concluded that repurchase agreements had a place in relieving "purely temporary It was recognized that experience stringencies in the money market" (p. 302). was not yet sufficient "to warrant the adoption of a specific formula under which repurchase agreements would be made available to dealers" (ibid.), but the hope was implicit that some formula could be found. I think this is a vain hope. Nor can I conceive of a version of "the line of credit theory" (p. 296), giving Government security dealers assured access to the repurchase window, that would not also mean a loss of control over reserves. In a highly organized national money market, marginal swings of $100-$200 million may often be the cutting edge of System policy during a Authorized for public release by the FOMC Secretariat on 2/25/2020 12 restrictive phase. If the dealers are guaranteed access to a comparable aggregate of Federal Reserve credit, at their own initiative, they will have in their hands the power to create or to extinguish enough reserves to offset the adjustments being attempted as part of the System's credit control. The cardinal aim of System help to the dealers should be promotion of the ends of credit policy. Help that defeats or curtails the effectiveness of credit policy would seem to represent a confusion between means and ends. Finally, recent experience (Nov.-Dec. 1955) does not suggest an important place for giving "advance notice of intention to make repurchases available" (p. 296). The problem is to retain initiative within the System, whenever actual contracts are being made, without appearing to default on the spirit and purpose intended by the advance assurances. In November-December, 1955, just as would probably happen during any period for which assurances of some type were given at the start, days did arrive when other factors had added greatly to reserves, and further repurchase agreements were not only unnecessary but would have been excessive. At such times contracts eventually had to be refused, with explanations, but dealers still felt they were being "let down". The only approach dealers really seem to understand - and perhaps this is all they should understand - is that of keeping us continually informed each day concerning their progress in finding money, and then waiting each day for us to decide on the basis of overall credit policy (partly based on a critical appraisal of these dealer progress reports) whether or not more reserves are needed in the market, and should be put in through repurchases. Experience does endorse the final alternative mentioned in the Ad Hoc Subcommittee report. "that the open-market account should be prepared to terminate or reinstitute repurchases from day to day based on the open-market management's judgment of the needs of the market" (p. 296). There may also be possibilities (provided legal questions can be resolved) of improving performance by permitting substitution of collateral under repurchase agreements without terminating the original contract, a procedure which was also favorably mentioned in Appendix B (p. 296) of the report. Our approach even here, however, should be on the basis of retaining the initiative in deciding, case by case, whether substitution would be in line with credit policy at the time. More important than the use of repurchase agreements to promote the better functioning of the Government securities market has been their use to test the state of the money market. The repurchase technique has become a most valuable means of testing the development of money market pressures, as these are reflected in dealer transactions and borrowing needs. Any by varying the rate, both above and below the discount rate, we have a further means of adapting mar- ginal releases or absorption of funds to the changing balance of bank reserve needs and money market pressures. The result of all these developments in the use of repurchases, where the initiative has been retained throughout in System hands, has been to assure their place as a needed and important part of open market operations. Authorized for public release by the FOMC Secretariat on 2/25/2020 13 IV. System Account Operations Over the Full Maturity Range of Government Securities The Reasons Given for Confining System Operations to the Short Term Market The prohibition on System transactions outside the short-term area rests basically on three arguments: 1. Intervention in the longer term market prevents depth, breadth and resiliency in the market, because it keeps dealers and investors so uncertain over where the System's giant portfolio will strike as to make trading on the basis of their own judgments unduly risky. 2. Intervention in the longer term market cannot be regulated. Once started, it either degenerates into pegging, or the Management of the System Open Market Account has to resort to pressures and ruses in order to get prices to change without meeting all the supply or demand called out by the change. 3. a. If pegging is the result, there is a consequent loss of control over reserves and destruction of the price flexibility essential for free markets. b. If pressure and ruses are tried, they create artificial markets, impose undue burdens on the dealers who must respond to suasion or deception rather than price, and will eventually break down anyway, with pegging the result. Intervention in the longer term market is not needed to inject the appropriate influence of credit policy into the maturity range outside the short area, for in a deep, broad, resilient market, arbitrage will promptly transmit the effects of action affecting bank reserves into all sections of the market. Moreover, the way in which that transmission occurs, and the extent to which it affects any particular issue, will be the freely determined result of market forces, reacting to the impulse provided at the short end. No intervention by an official body or agency can match the flexibility, nor the appropriateness, of rates determined by the complex of all factors at work in a free market. A brief indication of answers to each of these three arguments is now possible on the basis of experience with our present operating rules: 1. Removal of System "intervention", as shown earlier in this memorandum, has not brought increased depth, breadth and resiliency to the market for Government securities. More important, it has not relieved dealers or investors of concern over what the System may do, nor of speculation Authorized for public release by the FOMC Secretariat on 2/25/2020 14 over how hard the System may press through the short market toward unknown objectives in the long. After the MayJune episode in 1953, the long market reacted only slowly until the massive easing of System policy by way of a reduction in reserve requirements; and during the tightening phase of credit policy in 1955, reactions were often sluggish or perverse. a. The frequently recurring failures of the long market to reflect the easing or the restrictive effects being attempted through the short market illustrate a paradox in the logic of the Ad Hoc Subcommittee Report. Dealers continue to be so greatly influenced by expectations of Federal Reserve policy that they can, in acting on these expectations, reduce or intensify the intended effects of credit policy outside the short area, so long as the only visible evidence of policy intentions is given by changes in bank reserves, where temporary factors often obscure the goal of policy. Moreover, when the market persists in incorrect expectations of System policy, the effort ultimately taken to bring it around may result, for a time, in even greater tightness or ease than that intended by policy. b. Perhaps it is partly because of the System's inability to clarify the meaning of its policy, by some direct action outside the shortest area, that dealers have been less ready to make markets than they were formerly, with a consequent tendency for the longer markets to be "thin" most of the time. 2. There has been considerable "intervention." during 1955, for Treasury accounts at times of Treasury financing. It has not degenerated into pegging. At no time have purchases been carried through with rigid price targets to be maintained; instead, the effort has been to provide some additional buying simply to provide resistance against transitory psychological reactions. Established market procedures have been followed in all cases, and there has been no question of using ruses or pressure. The result has merely been to exert some steadying influence, during a brief period of large scale shifting of securities, helping the free market to make its own adjustments to a sustainable equilibrium level of prices. 3. Because there has been no improvement in the depth, breadth and resiliency of the market, and because the market has at times suffered from undue uncertainty over the intended implications of credit policy for the longer-term area, arbitrage has not provided the prompt or direct response to System action that the Ad Hoc Subcommittee envisaged. Authorized for public release by the FOMC Secretariat on 2/25/2020 15 Shouldthe System Free Itself to Operate, at times, in the Longer Term Market? In view of the character and performance of the Government securities market over the past three years, and of the experience that has been gained, through Treasury operations, in testing the feasibility of limited intervention in the longer term area of that market, is it not time to consider whether the System should not again free itself to operate anywhere in the maturity range? Otherwise I am afraid that the longer we persist in our self-imposed limitations, and the more we say and do to indicate that operations in anything ouside Treasury bills are an exception, the greater our difficulties will be when "exceptions" are forced upon us. At least should we not consider operating more broadly in the short term area, working out to the two or three year range, without public or private commitments, and letting experience determine our later course? I suspect that a preponderant share of all System operations would, in fact, be in this part of the maturity range, even if there were no restrictions on operations in longer maturities. For my part, however, whether this suspicion is valid or not, I find the evidence persuasive, on the basis of repeated experience in executing Treasury transactions, that System operations in the long term area could be handled so as to further the objectives of credit policy without either degenerating into pegging, or causing the System to relinquish control, exercised at its own initiative, over bank reserves and general credit conditions. Since there is already general agreement within the Federal Open Market Committee that the System's responsibility does include concern over changes in interest rates and the availability of credit in all maturity sectors, it should follow that, unless there are compelling reasons to the contrary, the System should be free to operate in a manner most likely to assure the effectiveness of policy over the whole range. This need not and should not mean It does mean that the imposing on the market a predetermined schedule of rates. System should at times directly nudge those sectors of the market that are not responding to the influences which the System intends to exert on general credit conditions. Probably the most important, among the many aspects of the need for flexibility in System operations, arises from the fact that we do not have the kind of arbitrage in the Government securities market that formed the basis of a good deal of the hopeful speculation in the Ad Hoc Subcommittee report. Consequently, so long as the System recognizes the existence of a need, at times, for exerting some influence upon the longer-term markets, and thus upon longer-term rates of interest, there will inevitably be times when action confined to the short end (in practice, Treasury bills) must take an extreme form, if it is to exert even a remote influence at the longer end. This could raise painful conflicts between credit policy objectives and open market techniques. Essentially, the need is to keep the largest single portfolio of Government securities (that of the System Account) available for trading in the market so that such conflicts can be avoided. Direct action by the System Account, even for small amounts, can provide important help to the dealers and the rest of the money market, in sensing the direction of change intended by System policy. There is no need for the System to meddle in the market every time prices begin to move; quite the contrary. But prices that are standing still from inertia, or persistently drifting in a direction opposite to the longer term trend implied by credit policy intentions, This would rarely if ever should at least be subject to testing from time to time. Authorized for public release by the FOMC Secretariat on 2/25/2020 16 involve large sums; small bids or offers would test out the readiness of the market to make good on quotations. Probably without exception, the amounts involved would be considerably less than those which would have to be used to gain a comparable effect through transactions in Treasury bills. From an even more practical standpoint, freedom to operate in any part of the maturity range would make it possible to get the greatest desired effect on markets and expectations, with the smallest outright release or absorption of reserves, whenever the need for positive influence occurs at times when there is little or no occasion for action to change the outstanding volume of bank reserves. And, at times when the need is essentially one for influencing bank reserves, rather than a particular part of the market, the System can often best achieve its aims if it can buy or sell the particular securities that happen to be available or in demand. There have been a number of occasions since the present operating rules were adopted, when the fulfillment of reserve objectives had to be accomplished by additional purchases of Treasury bills at times when Treasury bill rates had already dropped far out of line with money market conditions, or any levels consistent with the overall objectives of System policy. Some distribution of System purchases among the various kinds of securities then available, would have minimized the distorting impact of System transactions upon the existing rate structure. Similarly, there have been times when System selling to absorb reserves might better have been distributed among various issues, instead of concentrating upon the Treasury bill market with sharply distorting effects. How Should Operations Be Conducted Over the Full Maturity Range of Government Securities? I am not suggesting that all we need to do is to withdraw the present operating rules. There are difficult problems involved in working out satisfactory arrangements for conducting System operations outside the short term area. One such problem concerns the difficulty in which the Open Market Committee finds itself in formulating directives that permit needed flexibility in quick response to changing market circumstances, while retaining full control over the objectives being sought. Other difficulties would arise in specifying the kinds of transactions that may be undertaken, and the way in which the Management of the Account may alternate from one kind of transaction to another, in fulfilling the Committee's particular objectives under varying conditions. And the actual operating procedures would have to be carefully studied, and continually reviewed, to assure conformity with the "best price" criteria, appropriate for all System operations, in determining the distribution of business among dealers. The answers to all of these questions would have to be developed through experience. It should be possible to establish certain general guides, however, which would serve as a satisfactory basis for initiating System operations outside the short term area. 1. Formulating Committee Directives and Assuring Committee Control It seems to me that the general approach can proceed by analogy with that already being followed. That is, the published directives would take roughly the same form as they have in recent months. The Manager of the Account would have to rely for specific guidance on the full context of Committee discussion, and the final record of such discussion. Even the discussion, however, Authorized for public release by the FOMC Secretariat on 2/25/2020 17 would not have to take the form of detailed consideration of specific interest rates, nor of particular issues of Government securities. Instead, policy guidance could be provided in terms of the need for some influence to test out the prevailing conditions in the intermediate or long sectors of the market, possibly with a view to providing the market with some clearer indication than had been provided by Treasury bill operations alone, that System policy was aimed at somewhat tighter or somewhat easier conditions. There would probably be need at most meetings for some comment (there usually is now) on the direction of change apparent in the rate structure as a whole, as a symptom of the degree of restraint or ease prevailing in the various sectors of the market. Attention would have to be given, too, to prevailing market tone and to the need, if any, for exerting influence upon, or providing some guidance to, the expectations of dealers and investors. It could then be indicated whether a particular objective ought to be achieved solely through an immediate release or absorption of reserves by purchase or sale of short term securities, or whether some substantial part of such release or absorption should be accomplished by directing purchases or sales of securities to sectors of the market where some direct rate influence was also desired. Or, the Committee might wish to have its reserve objectives accomplished while exerting the least possible influence on prevailing interest rates, and could instruct the Management of the Account to buy, or sell, whichever securities would make that possible, in the light of day-by-day market conditions. With this kind of guidance, the Manager would have a sufficient basis for day-to-day judgments in executing policy. Questions would still remain as to whether or not the Manager had been given too much discretion, or whether his exercise of judgment in practice corresponded with the general intentions implied by the Committee discussions. Again, in these matters, we already have the basis for an answer in the procedures that have developed over recent years. Present arrangements for reporting promptly on all aspects of operations to the senior officials throughout the System ought to provide an adequate basis for continuous Committee review of the Manager's judgment and action. 2. Types of Transactions As a general rule, three broad types of transactions can be visualized among which the Committee might be expected to indicate some preferences, from meeting to meeting, in clarifying its intentions. One of these is "swaps", through which the System itself can, in effect, transmit pressures from one part of the market to another as a substitute for market arbitrage, without any net release or absorption of reserves. which the Committee viewed as Second, there would be outright transactions it would expect temporary in character, and which to have reversed in the market within a relatively short time. In effect, this would be a means of doing, outside the short term area, what is now done inside that area by the use of repurchase agreements, or by turnaround purchases and sales of Treasury bills. Third, there would be the cases in which outright transactions should be undertaken on a more nearly "permanent" basis, and joined with the reserve objectives currently being sought. This latter type of transaction might arise either under conditions in which the Committee wished simply to have its reserve objectives achieved by effecting transactions in whatever part of the market offered the readiest possibilities for accomplishing them, or in conditions where the Committee saw an opportunity to combine effects intended upon reserves with those specifically desired in certain parts of the market. Authorized for public release by the FOMC Secretariat on 2/25/2020 3. Operating Procedures There is no reason why the "best price" principles that have characterized all System transactions, since the end of the war and early postwar period, cannot be applied in transactions executed anywhere over the maturity range. The form might vary, but the intent, in providing an equitable basis for the distribution of business among dealers, could be preserved. Dealers are characteristically reluctant to offer to buy or sell longer term securities on a firm basis for the time period (twenty minutes or more) that is necessarily involved in the "go-around" procedure. However, because there are fewer dealers who actively make markets in the long term area, at least at present, it may be possible at times to use the "go-around" technique. That will depend, too, upon the particular objective being sought. The "go-around" approach may, in some conditions, so strike market psychology as to put prices down; in others, pull them up. Other techniques can have similarly variable effects. It is quite possible that dealers will come into the Trading Desk frequently on their own initiative, as they often do now in checking for possible orders from other accounts than the System Account, so that occasions for the System to take the initiative in "showing an order around" might not occur too often. There is no point, therefore, in trying to standardize procedures. Given the objectives, the Account Management must vary the method with the circumstances in attaining the desired result, and then report fully to the Committee on the "how" and the "why". It probably would be necessary, more often than not, to do business rather promptly, "on the wire", on the basis of testing comparative prices among four or five dealers and selecting those that best meet the immediate need. By maintaining accurate records and check lists, there should be no difficulty in assuring a fair distribution of opportunities to do business among the various dealers who make markets outside the short term area. There will undoubtedly be some occasions when the objectives can best be achieved by remaining anonymous, just as there will be others when the System will wish to have it known that it is buying or selling in a certain sector of the market. There will be some circumstances when swap transactions are preferable, and others when outright transactions are to be preferred. The details 5/ It was in connection with swaps that we encountered criticism on the last routine operation outside the Treasury bill market (December 1953) - the question being, in effect, "who gets favored", because a large swap appeared to benefit or facilitate a private operation, and embarrassed at least one dealer firm which had apparently envisaged a profitable role for swapping, through its own position, at wider spreads than the System accepted. The System's answer to that criticism has, thus far, been to ban swaps of any shape or form. While fully mindful of the risks of occasional collisions of this kind, I think we have denied ourselves too much, and that there is a different answer: let decisions to effect swaps reflect the judgment of the Committee that they would, in given circumstances, best serve to effect credit policy, or be useful in the practical administration of the Account (especially in improving the maturity distribution of the Account); let specific swaps as contemplated be reported to the Committee, as a further confirmation that the actual step is in line with the Committee's general instruction; and let swaps be handled on a broadly competitive basis, unless the Committee directs otherwise in an unusual case. Authorized for public release by the FOMC Secretariat on 2/25/2020 19 of our procedures would have to be developed, and reviewed by the Federal Open Market Committee, if and when we again permit ourselves to operate over a wider range of maturities, at times when such operations will promote the aims and objectives of credit policy, or give appropriate assistance to Treasury debt management without endangering credit policy. I have no fixed prescription for these wider operations, but I am sure that we do not have to let operating considerations keep us from renewing System touch with the whole of the Government securities market. V. Credit Policy and Treasury Financing Anxious to consolidate the achievement of the Accord of 1951, in relieving the System of a continuing obligation to support Treasury issues, the Ad Hoc Subcommittee report veered to the opposite extreme in its treatment of our responsibility toward debt management. The Ad Hoc Subcommittee did not recognize in 1952, and the full Committee since then has been reluctant to recognize the need of the Federal Government for some underwriting assistance, through the initial offering phase of its securities, such as is characteristically expected in connection with important market offerings by private borrowers. There has, to be sure, been a continuing recognition of the desirability of keeping an "even keel" in the money market through Treasury financing periods, particularly during the various phases of a restrictive credit policy, but with the limited exception of last November-December, our aid has been given without buying "rights", or "when-issueds", or any other securities except Treasury bills - although, of course, repurchase agreements have been made against rights and other outstanding securities within the fifteen-month range. The exception to our present operating rules which was made on November 30, 1955, must have raised some doubts, however, as to whether this should only be an exception to prove the rule. Perhaps it would be better to view this test development as the culmination of a series of developments - pointing toward the need for some underwriting aid to Treasury financings whenever a general program of credit restraint continues over a period of time, even if the Treasury conscientiously does its best to price its offerings in an appropriate relationship to the investing market and tailor the terms to investor needs. The comparative tranquillity of Treasury operations during much of the period since we ruled ourselves out of the role of possible underwriter has been more apparent than real, and this has been particularly so during the past year of credit restraint. Every Treasury market operation during 1955, apart from two of the issues of Tax Anticipation obligations, has had some Treasury underwriting usually not recognized in the market, but enough to spare the market (and the Treasury) the full consequences of adherence to the theory of a completely "free market". In connection with each of the four refunding operations in 1955, Treasury accounts have invested amounts ranging from $50 million to nearly $300 million in underwriting support. These relatively moderate amounts were effective, along with direct System help in the last of these episodes, in averting disruptive market developments, which in each instance had begun to appear. Although it must be recognized that these Treasury purchases came on top of System action aimed at steadying the money market through the financing period through buying Treasury bills, or making repurchase agreements, or allowing market factors to supply reserves - the general success of the efforts indicates that a little help, directed at the particular issues exhibiting greatest strain during a financing period, can accomplish a great deal without impairing the essential character of the market or preventing price flexibility. Authorized for public release by the FOMC Secretariat on 2/25/2020 20 It is not sufficient to reply that the Treasury should be expected to handle all of its own underwriting problems, for many of the problems are rooted in aspects of market performance or psychology that involve credit policy as well as debt management. And there is also the risk that future Treasury administrations may regard a System policy of indifference, concerning the behavior of the Government securities market during financing periods, as leaving the way wide open for maneuvers to stage-manage the market - maneuvers that might result in serious outright collision with credit policy. If the System were, instead, in a position to consider jointly with the Treasury, as a matter of course, the there would seem to be much greater hope for maintaining a check on the nature and magnitude of any substantial underwritmarketing impact of Treasury issues, ing support that might be undertaken. The counter to this hope is the fear that the door can never be opened just a crack, and that once we again begin giving direct assistance, the door will be swung wide open, the System will be back where it was before the Accord, and the market will be subjected to a succession of rigging operations. While this, at any time, would depend on the character of the men bearing responsibility, both in the Treasury and in the System, it appears doubtful indeed that these relationships can, in the long run, be left on an all or nothing basis. The greater risk, as I see it, lies in an attempt to achieve complete separation, for sooner or later a combination of events, and Treasury determination, will make what the System regards as neutrality appear to others as a stubbornly doctrinaire attitude. That sort of situation could invite action of a drastic character, perhaps by Congress itself, to assure System attention to the market problems which are necessarily connected with large Treasury borrowing operations, and which can have disturbing influences on credit markets, as well as for the Treasury, if some moderate underwriting assistance is not available. Having fought Treasury domination of System credit policy as vigorously as any one,I am not suggesting that we again enter that trap. But the idea that "good fences make good neighbors" is an over-simplification of Treasury-Federal Reserve relations. We cannot be, and in fact we are not, indifferent to the success, or the influence upon the credit markets, of the Treasury's debt management program. It seems to me only proper, in terms of the kind of coordination among agencies of Government that helps the effectiveness of the whole, to utilize the System Account as a means of facilitating Treasury operations whenever our objectives are parallel and our steps can be synchronized. Quite apart from these matters of grand strategy, there are a number of operating considerations that have significance, whether or not they constitute an irresistible case for dropping our present rules altogether (or for making more frequent exceptions, while we keep them). (1) As has already been indicated in periods such as the recent brief period of steadily increasing restraint, the magnitudes needed to provide temporary underwriting assistance may become too large for the Treasury accounts to handle. This is not to imply that any attempt at assistance must be so handled as to give the whole market a wide-open invitation to unload securities on the official accounts, nor that vast sums will always be required once any direct assistance is begun. In Authorized for public release by the FOMC Secretariat on 2/25/2020 21 my judgment, there was no such invitation involved in the recent November-December episode and the total purchases were much smaller than the total loose supply that was overhanging the market. (2) Even if the Treasury had virtually unlimited resources, (possibly through borrowing directly from the however System on special certificates),there is the question of reserve effects. Reliance upon the Treasury's judgments alone, when underwriting was needed, certainly might in some circumstances mean an abnormally large release of reserves - making "sure" so far as the Government securities market is concerned, but not mindful of the reserve consequences. A synchronized approach, which included System reserve operations as a normal part of the program, would assure maximum effectiveness from the minimum use of funds, and assure suitable control over the total release of reserve funds. For the Treasury just to spend as it sees fit during one borrowing operation, and then to come quickly to the market again to borrow new money (and perhaps repay special borrowing from the System) might have the virtue of simplicity, but would not be likely to ease market strains - which is, after all, the objective of the underwriting assistance. (3) Subsequent reversal in the market of purchases made during an issuing period (or any offsets or swaps carried out during the period) should be carried out in conformity with the steps then being taken to implement credit policy. (4) Wholly apart from problems of magnitude or those related to the possible desirability of offsets or subsequent reversals, there may be difficult mechanical or legal problems in using the various Treasury accounts for the underwriting that is needed. In several instances during 1955, it was clear that the Treasury faced mechanical difficulties in arranging authorizations to purchase the relevant securities promptly, even in small amounts. The Treasury's use of funds available in the accounts must conform to the statutory earnings requirements applicable to each, as well as to the current flow of investable funds that each may have available. Because the Treasury has, up to the present at least, been conscientious in its handling of the trust accounts, it has been hampered in helping its own financing operations. To be sure, no one can write a formula for the amount of underwriting assistance that is appropriate or will be needed in all circumstances. There is no substitute in this matter, any more than in other critical aspects of System Authorized for public release by the FOMC Secretariat on 2/25/2020 22 operations, for reliance upon the judgment and competence of the officials involved. In the long run, forthright recognition of that principle will do much more for the System than any attempt to follow rigid rules in meeting problems created by the infinite variety of market circumstance and psychological reactions when large borrowing and refunding operations occur. My own opinion is that one or more of the following kinds of operations should be contemplated as possibilities in connection with Treasury financing operations, and that we should not prejudice their use, in advance, by enunciat- ing a general rule which seems to deny their validity. (1) The System Account should be permitted to make swaps, offering any of the securities out of its portfolio for which there may be a market demand, against the purchase of maturing rights or when-issued securities involved in a Treasury financing operation. Clearance to engage in swaps would carry with it a mechanical protection against any undue release of Federal Reserve credit as part of the underwriting assistance, since the end result would merely be an exchange among securities without any outright buying for the System Account. (2) The System Account should be authorized, subject in this instance as in all others to specific review at the time by the Federal Open Market Committee, to enter exchange subscriptions for any maturity being offered by the Treasury in connection with a refunding operation. (3) The System Account should be authorized to purchase outright other securities in the short term area, if not throughout the maturity range, whenever that method of releasing needed reserves would, collaterally, provide indirect underwriting assistance that would promote or smooth market re- sponse to a Treasury offering. (4) The System Account, subject to review at the time by the Federal Open Market Committee, should be authorized to buy rights. (5) The System Account, subject to review at the time by the Federal Open Market Committee, should be authorized to purchase when-issued securities on an outright basis. Credit policy and market events at the time should determine the priorities among the five possibilities listed, with specific questions concerning one or the other being raised directly with the Committee whenever serious problems arise in connection with future Treasury borrowing operations. The overriding principle, as I see it, is that the Committee be free to make the best choice in the circumstances of each borrowing operation, and not be bound by rigid rules. CONCLUSION This memorandum is necessarily longer than it otherwise would be because it raises questions about the continuance of operating rules of the Federal Authorized for public release by the FOMC Secretariat on 2/25/2020 23 Open Market Committee which have been firmly espoused by a substantial majority of the Committee over a considerable period of time. By the same token we have now had enough experience with the operation of these rules to warrant a considered re-examination of their validity and their efficacy in promoting the aims and objectives of credit policy, and in discharging our secondary responsibilities with respect to Treasury debt management. My own study of the problem convinces me that our experience suggests, if it does not demonstrate, that the hopes engendered by the report of the Ad Hoc Subcommittee, which provided the basis for the adoption of the present operating rules of the Federal Open Market Committee, have not been realized in practice, either in terms of improvement of the depth, breadth and resiliency of the Government securities market or in terms of our ability to operate through that market in attempting to obtain the objectives of open market operations. The reasons presented for denying ourselves, in advance, full freedom of action in these operations, therefore, have not been confirmed, even though it may be claimed that our experience with present operating rules, on the whole, has not been unfavorable. More important than this partly fortuitous result, however, is the fact that there have been times, particularly during periods of credit restraint, when we might have done a better job than we did. In addition there is the danger to which we expose ourselves so long as we retain rules on the books which may have to be breached in particular circumstances in the future. Flexible operation is inhibited by unnecessary prohibitions, and necessary departures from these prohibitions subject us to criticism and complaint - criticism that we have abandoned principle; complaint that we have misled the market. The longer these rigid rules are kept in force the greater the difficulty in deviating from them, or abandoning them entirely, and the greater the danger of having to break them down piecemeal in our actual operations. This suggests to me that the time has now come to consider relieving ourselves of the unnecessary restraints on our freedom of action which these rules imply and enforce, despite the fact that they are supposed only to be in effect until superseded by further action of the Committee. I am encouraged in presenting this view by the fact that improvements in the functioning of the Federal Open Market Committee, and in communication between the operating bank and the Committee, during the past three years, should have removed concern of individual members of the Committee that they were not fully sharing its responsibilities or participating in its decisions, and that certain fixed rules of operation were necessary for their protection in the discharge of their duties. We are all now in a position to contribute to the making of policy on the basis of current economic conditions, and to keep ourselves informed as to how those policies are being executed, from day to day and week to week. Whatever basis there may have been in the past, for the adoption of the present rules of operation,seems to me to have been reduced or to have disappeared. My individual suggestion and recommendation would be that we now remove these "continuing operating policies" from the books and that we discuss the matters with which they are concerned at each meeting of the Committee, or between meetings of the Committee, whenever any member of the Committee, or the Manager of the System Open Market Account, brings to the attention of the Committee circumstances and situations which may warrant deviations from the In that way we practices which we have followed during the past three years. could restore to our techniques of operation the same flexibility which was re- stored to us in matters of policy by the "Accord" of 1951. Allan Sproul
@FedFRASER