A History of the Federal Reserve, Volume 2 (13 page)

Hayes’s response remarked that “we are all very close to agreement as to the way in which the account should actually be administered” (Memo to Special Committee, Board Records, February 4, 1958, 1). He conceded, at last, that bills-only may have helped the market to make its own adjustments, and he agreed that “while Treasury financing operations unavoidably must be taken into account, any System efforts to facilitate them should be definitely subordinated to our primary objectives of monetary policy” (ibid., 1). Alas, like the Board and its staff, he did not explain how the subordination of Treasury financing to policy objectives could be reconciled with the System’s reluctance to raise rates, steeply if necessary, to prevent Treasury financings from increasing money growth under evenkeel policy.

Hayes accepted that most operations would be in the bill market. The main point of his memo was that the System should be more flexible. He again urged rewording the operating directives by introducing less restrictive language (“usually,” “as a general rule”) and permitting the desk to engage in securities swaps with the market.

The five-year review did not change the policy guidelines. Martin, the FOMC, and the staff continued the bills-only policy despite the continued opposition among prominent members of Congress and little support from academic research.

The Board’s intellectual defense of bills-only did not claim that arbitrage was perfect or that long-term rates moved directly with short-term rates and only in response to such movements. To support the Board’s position, Riefler (1958a, 1958b) relied on four main arguments; not all were present in each discussion.

First, the main direct effect of an open market purchase or sale came from the change in reserves, not from the change in the outstanding stock of securities. Hence, purchases or sales of long-term debt would not have much additional effect.
²⁸

Second, an indirect effect on expectations amplified the System’s action. Market professionals “are not likely to operate against any trend in rates they think the System is trying to establish. . . . [W]hen System actions give rise to firm expectations among market professionals with respect to
interest rate trends, relatively small System operations may have important short-run effects on market quotations” (Riefler, 1958b, 1263).

28. As part of this argument, Riefler claimed that the multiplier of bank reserves was about 7, based on the reciprocal of the reserve requirement ratio (1958b, 1262, 1269). This calculation ignored leakages into currency. New York pointed out another error, “the continued effort to maintain the fiction that increases in reserve requirements . . . and the omission of any reference to the fact that, in every instance, the increase in reserve requirements led to increased sales of Government securities to the System” (Roelse to Sproul, Sproul papers, FOMC Correspondence, January–March 1952, February 5, 1952
, 2).

Third, operations in the long-term market, even if frequent, “would indicate a feeling on the part of Federal Reserve authorities that existing prices and yields on long-term securities were out of line” (1958b, 1264). Since long-term bonds traded infrequently relative to bills, these actions would disrupt the market.
²⁹
“Bill operations can also give rise to false or misleading expectations, but they are much less likely to do so” (ibid.).

Fourth, open market operations change the volume of securities that the market holds. Substitution and arbitrage spread this effect to securities along the entire maturity spectrum. Riefler put greater stress on substitution than on yield arbitrage. He dismissed theories that put buyers and sellers into “preferred habitat” maturities where they remained. He claimed a “high degree of actual substitutability . . . for many lenders” (Riefler, 1958a, 7).

The Board had the better analytic argument, but it did not win the point with Congress and academic critics. Many of the critics confused nominal and real rates of interest. Believing that the Federal Reserve could reduce long-term interest rates and refused to do so, the leading academic economists who prepared the
Report
on
Employment,
Growth
and
Price
Levels
for the Joint Economic Committee, chaired by Senator Paul Douglas, wrote:

Just as the preaccord policy promoted artificially low interest rates, the policies pursued, particularly since 1953, have brought interest rates to levels higher than they should or need be. [T]he full potential of monetary policy to promote stability and economic growth has not been realized. (Joint Economic Committee, 1960b, 30)

Faced with this criticism and strong opposition in Congress,
³⁰
the Board ended the bills-only policy in the fall of 1960 and in 1961 formally revoked the 1953 policy. The new administration wanted to keep short-term rates
up while lowering long-term rates. The Federal Reserve cooperated by occasionally buying long-term and selling short-term securities. Sproul was elated (letter, Sproul to Hayes, Sproul papers, Correspondence, February– March, 1961, March 14, 1961).

29. Remembering the pegging period, Martin (1959, 20) wrote: “If an attempt were made to lower long-term interest rates by System purchases of bonds and to offset the effect on reserves by accompanying sales of short-term issues, market holdings of participants would shift by a corresponding amount from long-term securities to short ones. This process would continue until the System’s portfolio consisted largely of long-term securities.” This would not be true if the market expected rates to fall. Two years later, Martin endorsed the policy he rejected here. The written argument was likely the work of the staff.

30. In 1959, Congress refused the administration’s request to remove the 4.25 percent ceiling rate on government bonds with more than five years’ initial maturity until the Federal Reserve abandoned the bills-only policy. In 1960, the Joint Economic Committee made as its major recommendation for monetary policy that the Federal Reserve “abandon its discredited ‘bills only’ policy and acquire long-term bonds for the portfolio” (Joint Economic Committ
ee, 1960a, 16).

The bills-only policy may have contributed slightly to strengthening the market for long-term Treasury securities, but the evidence is mixed. After seven years, only a few government security dealers held positions in the long-term market or made markets in those issues.
³¹

In retrospect, much of the criticism of bills-only seems misdirected. If there were gains for stabilization policy that the Federal Reserve could, but would not, achieve, the Treasury could have changed its offerings to supply more (fewer) bills and fewer (more) long-term securities. Until 1959 debt management by the Treasury remained fully capable of maintaining the composition of the debt that the Treasury wanted (or that was socially optimal). For a brief period in 1959–60, the 4.25 percent ceiling restricted what the Treasury could do.

Bills-only did not achieve the System’s principal aim—to free itself from responsibility for the success of debt management and support of Treasury offerings. Treasury operations were a continuing concern. The Federal Reserve held interest rates unchanged before, during, and after Treasury sales. It intervened in the bill market frequently and delayed policy changes to provide enough reserves so that banks could buy new or replacement issues. It no longer counseled the Treasury on what it should issue, but it estimated the reserves it would have to supply to make the issues succeed.

For the System, the Treasury was not just another borrower. Some members suggested that if many holders redeemed for cash (called attrition), the main effect would be that the Treasury would have to borrow again. The Treasury and the Federal Reserve did not accept this view. Attrition was evidence of failure, a cause of embarrassment that both wanted to avoid.

The Federal Reserve described its role as independent within the government, but independence within government included some support of Treasury offerings.
³²
Its policy, called “even keel,” later severely restricted
the number of days on which the Federal Reserve could change policy. All members did not always agree with this commitment, but a large majority did. Chapter 3 has a table showing “even keel” activity during 1966 to 1971.

31. In 1960 Congressman Wright Patman of the Joint Economic Committee commissioned Gert von der Linde and me to study the dealer market. The conclusion in the text is based on that study (see Joint Economic Committee, 1960). The Treasury supported or did not criticize bills-only during the 1950s, but members of the FOMC began to tire of the criticism and to urge greater flexibility (memo, Mangels [San Francisco] to FOMC and Bank Presidents, Board Correspondence, February 24, 1960).

32. This was not new. In the 1930s, Treasury Secretary Henry Morgenthau held the Federal Reserve accountable for any change in market rates no matter how small. It took until
the 1970s before the System recognized that its problems with the pricing of Treasury issues would be reduced if the Treasury auctioned its debt instead of fixing the price. Auctioning debt gradually freed the Federal Reserve from “even keel” operations. It became possible to slow inflation despite persistent, large budget deficits, although it did not have that effect for several years.

Transforming
the
FOMC

The Banking Act of 1935 mandated four meetings of the FOMC in Washington each year. In the early 1950s these meetings set policy guidelines. The chairman and vice chairman of the FOMC met every two weeks with three other members of the Executive Committee to decide how much to buy or sell to achieve the full committee’s objective and to instruct the manager.

In practice, New York dominated the Executive Committee. Sproul and the account manager, Robert Rouse, had much more information than the other members about the current and near-term position of the money market, prospective Treasury offerings, and other technical details. Consequently, New York could usually get its way. Although Martin and Sproul often agreed on policy, Martin wanted to increase Board control and the participation and influence of other FOMC members. He proposed to discuss termination of the Executive Committee at the June 1955 FOMC meeting and to include discussion of discount policy, changes in reserve requirement ratios, and margin requirements at FOMC meetings. The FOMC was the “heart of the System,” and it should discuss all of these decisions, not leave some to the Executive Committee and others to the Board (Executive Committee Minutes, April 12, 1955, 5). New York’s advantage weakened in later decades, so the regional banks became more important participants, but the Chairman dominated decisions.

In June, the time for a showdown with New York had come. In advance of the June 22 FOMC meeting, Riefler mailed a proposal to all the presidents that would abolish the Executive Committee and substitute full committee meetings every three weeks. The general counsel prepared a memo stating the required changes in by-laws, regulations, and procedures to adjust to the change. Martin repeated his earlier statement about increasing the FOMC’s responsibility.
³³
Sproul argued that the Executive Committee
could meet quickly in an emergency and that the telephone was not a close substitute for “a face to face meeting at which ideas can be developed and debated, and the reaction of your associates to those ideas can be observed and taken into account” (FOMC Minut
es, June 22, 1955, 7).

33. In 1953 Martin considered a legislative change to reduce the Board to five members (each with a ten-year term) and the FOMC to nine members, five from the Board and four from the reserve banks. New York would continue to hold a permanent seat. In testimony at
the 1952 Patman hearings, Martin had also proposed ending the prohibition against reappointment after a full term and limiting terms to six years. The Board staff prepared legislation. Martin discussed the proposal with Treasury Secretary Humphrey and Undersecretary Burgess and hoped that the new organization would be in place by March 1954 (letter with attachments, Martin to Burgess, Martin papers, April 8, 1953). Eccles had proposed a
fiveperson board in 1935.

Several members pointed out that almost all members of the full committee had attended Executive Committee meetings that spring. New York was clearly on the losing side, just as it had been in 1930, when the Board urged an expansion of the committee to include all twelve banks instead of five. With little further discussion, the committee voted unanimously to abolish the Executive Committee and transfer its responsibilities to the full committee. The FOMC issued a press statement the following day, June 24, 1955. Slowly the regional bank presidents increased their role and appointed advisers to inform them about policy actions.

The vote settled an issue that was older than the Federal Reserve System: Where should control of monetary policy reside, in Washington, the political capital, or in the regional banks, particularly New York, where merchants and bankers had more influence? The Banking Act of 1935 had shifted power to Washington by giving the Board a majority on the FOMC, by putting the appointment of reserve bank presidents under Board control, and in other ways.
³⁴

To give a larger voice to the regional banks and the Board, while retaining influence over decisions, Martin instituted the “go around.” FOMC members discussed conditions in the economy or in their districts, stated their concerns, and recommended policy action. Martin then summarized the views, described the consensus, and put his statement to a vote only if there was a change in the directive or substantial division. Members could dissent, but generally they supported Martin’s consensus in the 1950s. Initially, this was a more substantial change in procedure than in practice, since many of the presidents did not have strong views. In time, new people with more training in economics or experience in financial markets became presidents of the reserve banks.

The change had two im
mediate effects. One was to lessen New York’s
dominance. Martin usually asked Sproul to speak first, after the staff briefing. Sproul always had prepared remarks and stated his position forcefully but was always courteous.
³⁵
His descriptions of market conditions, current and prospective Treasury borrowing, and the economic outlook were carefully done and respected by the other members. Although the procedural changes reduced his discretion and independence, he remained influential. With the passage of time, the FOMC developed procedures for keeping regional presidents and their staffs informed about economic and market conditions and policy actions to a greater extent than under the older arrangements.

34. In the 1930s, Chairman Eccles tried to keep the presidents from voting at the FOMC and made New York alternate with Boston as a member of FOMC. The latter effort failed, and New York was made permanent vice chair in 1942. Gradually, New York recovered some of the influence lost in 1935–36.