A History of the Federal Reserve, Volume 2 (28 page)

OTHER ISSUES

During the years 1971–73, the Treasury, at last, began to auction securities with more than a year until maturity, reducing the importance of even keel. Innovation in banking continued. The Board had to decide what to do about close substitutes for demand deposits such as negotiable orders of withdrawal (NOW accounts) that circumvented interest rate regulations. Financial markets increasingly accepted the challenge of regulation to find ways to legally avoid restrictions. Slowly the Board learned that lawyers and bureaucrats make regulations but markets decide how to circumvent them.

Personnel
Changes

Three Board members and four presidents changed in 1971–73, and Peter Sternlight replaced Alan Holmes as manager of the open market account in 1973. Table 6.9 shows the changes in Board members and presidents. Board members especially changed more frequently than in the past. Their real income declined with inflation.

Burns wanted replacements to the Board who would support his positions. His often close relationship with President Nixon gave him more influence on appointments to the Board than Martin had, but his fluctuating relationship with the president limited his influence at times. By May 1973, the majority of the Board owed their positions either to President Nixon or to Burns. This was particularly true of Robert Holland, who had served on Burns’s staff. He was the first of several staff members promoted to the Board. None of the new members stayed very long.

Burns used his office, personal standing, and approval power to influence the choice of reserve bank presidents. In the first instance, the reserve bank directors chose the president, but the Board had to approve the choice. Although there is no way to measure his influence, he seems to have been more involved in the choices than either Martin or some of his own successors. One way of increasing his influence was to have the directors submit several names to the Board instead of a single candidate.

Delegation
of
Authority

Legislation during the 1960s increased the Board’s responsibilities. In addition to new regulations affecting consumers, the Board had to decide a greatly increased number of mergers, acquisitions, and holding company actions. The approval or rejection process slowed. By late 1971, half the applications took more than the ninety days of review that the Board had set as a standard for the staff (Board Minutes, January 4, 1972, 3).

A staff memo proposed that the Board delegate to the reserve banks
authority to approve acquisition of mergers, membership, and acquisition of new banks by holding companies. Its letter to the reserve banks set out some uniform standards to apply. The Board’s staff continued to review decisions and supporting documents (letter, Board to Reserve banks, Board Minutes, January 17, 1972). In 1973, the Board broadened the authority of the reserve banks over holding companies.

Securities
Auctions
and
Even
Keel

Changing to a monetary or aggregate target required the System to reconsider even keel policy. The commitment assured “reasonable stability in narrowly defined money market conditions such as the Federal funds rate and net free or borrowed reserves” 94 (memo, Staff to FOMC, Board Records, July 17, 1970, 2). Its origin was the 1951 Accord that separated the responsibilities of the Treasury and the Federal Reserve for debt management.

An earlier study by Stephen Axilrod found that in the past even keel sometimes “enlarged [the] rate of increase in the monetary aggregates” (ibid., 4). After the shift to monetary targets, the desk and the FOMC “sought to achieve target growth rates in the aggregates only on an average basis over periods of several months” (ibid., 7). The desk did not get estimates of the monetary aggregates for a week, so the manager “tended to maintain day-to-day money market conditions within fairly narrow ranges for a number of days at a time, shifting to new target ranges only at relatively discrete intervals after firm data on the aggregates showed persistent deviations from their target paths” (ibid., 9). Hence, the desk
could continue even keel actions much as before. This helped the dealers to avoid losses during the period between securities sales and their distribution to portfolios. The memo recognized, however, that if the Treasury auctioned bonds and notes, the System could apply even-keel in a less rigid way.

94. Usually even keel did not apply to bill auctions, but the FOMC made exceptions during large cash offerings. The time span for even keel varied with the offering and the market environment (memo, Staff to FOMC, Board Records, July 17, 1970, n. 2). The Board’s staff and the FOMC never considered how much the Treasury paid for even keel. The market knew that the FOMC waited until after a Treasury sale to raise interest rates. There should be information in the asymmetry between periods of rising and falling rates.

The memo recognized that neither the market nor the staff could estimate accurately future values of market interest rates, reserves, and money growth. It proposed to retain the old procedure for even keel. The major change was the adjustment of interest rates once it became clear that the reserve and monetary targets would not be met (ibid., 13).

Beginning in 1970, the Treasury began auctioning notes and later bonds. Previous attempts in 1935 and 1963 failed, so the Treasury introduced auctions slowly, starting with shorter-term coupon securities and keeping procedures close to the familiar Treasury bill auction.
⁹⁵
As the variance of Treasury yields increased, it became much harder for the Treasury to price the bonds or notes (Garbade, 2004, 35).

The first auction in November 1970 successfully sold eighteen-month notes. After several similar note offerings in 1971, the Treasury auctioned two-, three-, and four-year notes in 1972 and twenty- and twenty-fi veyear bonds in 1973. By the middle of 1973, the auction had become the established method for selling notes and bonds. At first the Treasury announced the coupon and let the market set the price. In September 1974, it allowed the market to bid on the basis of yield. After the auction, the Treasury set the coupon to meet the yield. To facilitate bidding, in 1975 the Treasury permitted “when issued” trading. Bidders would know the approximate yield at the time they placed their bids.

Auctions reduced the use of even keel procedures but did not, at first, eliminate them. The January 1973 auction of twenty-year bonds prevented “the desk from exercising as much restraint on the growth of reserves as otherwise would have been desirable” (memo, Fred Struble to Axilrod, Burns papers, B_B81, August 2, 1973). Gradually, even keel disappeared as a major constraint on Federal Reserve actions to control inflation.

Revision
of
FOMC
Procedures

After Sherman Maisel left, Andrew Brimmer took up his efforts to focus more on long-term objectives. Brimmer wrote that “our present proce
dures do not integrate adequately our short-run policy decisions with a strategy that adapts monetary policy to the longer-run needs of the national economy. . . . [W]hen the principal focus of the Committee’s attention is on the probable effects of its policy decisions on financial variables in the immediate weeks ahead—as is often the case—a grave danger exists that the long-run course of monetary policy will turn out to be almost an incidental by-product of short-run decisions” (memo, Brimmer to Holland, Board Records, April 14, 1972, 1).

95. Friedman (1960, 64–65) criticized the Treasury’s use of fixed-price offerings as inefficient. Later, he showed that in fact the Treasury at times offered substantial premiums because it misjudged market yields. He proposed a single price sale so that small investors could get the same price as large investors. The inefficiency arose because the Treasury did not know the market-clearing price, so it often paid a premium.

Brimmer’s memo recognized that the FOMC had to develop strategies capable of reducing inflation. Concentration on control of short-term monetary changes did not assure that longer-term goals would be met. In fact, Brimmer claimed that the opposite had occurred; concentration on shortterm changes had come at the cost of reduced attention to long-term goals. He proposed that the FOMC hold three “outlook meetings” a year.

Congressman Henry Reuss also wanted changes. In a January 14, 1972, letter, he urged the System to adopt seven guidelines for monetary policy. His aim was to improve control of inflation and unemployment or growth. His proposal, like Brimmer’s, would have shifted attention to longer-term objectives.

Burns’s reply emphasized the importance of discretionary changes, avoidance of a rule for money growth, and the need to take account of resource utilization, lags in response, fiscal policy, and other factors that Reuss mentioned. Reuss suggested setting three-month monetary targets. Burns replied: monetary targets “should be evaluated over a period of at least six months” (letter, Burns to Reuss, Board Minutes, February 7, 1972, 2).

Reuss also stated that the Federal Reserve had to assist the housing industry by purchasing issues of the housing agencies and by buying longterm securities at times of monetary stringency. He neglected to say that credit and money are fungible. Burns’s reply noted that the FOMC had purchased agency issues since September 1971 with little effect on the housing industry. He did not point out that the supply of housing depended on real resources. Burns also was skeptical of the System’s ability to change the shape of the yield curve. This was an improvement, perhaps reflecting experience in the 1960s, when the System tried to lower longterm rates while raising short-term rates.

Burns ended his response by again citing the importance of “confidence.” Confidence affected not just spending but also the demand for liquidity. This, he said, contributed importantly to the length of the lag in monetary policy (ibid., 4). Burns seemed to think of confidence as an autonomous influence.

In June 1973, the staff reviewed experience with RPD control. It con
cluded the results were “mixed and difficult to interpret” (memo, Axilrod to FOMC, Board Records, June 8, 1973, 2). Control would be improved if the RPD target range was narrow and the federal funds constraint was wider. The memo showed that the RPD targets were almost never hit. The procedure had a much better record of controlling longer-term money growth than short-term ranges (ibid., 14, 28).

Surprisingly, the account manager was more positive than the staff about the experiment. He disagreed, however, about the desirability of widening the bands on the federal funds rate. Unless the bands became very wide, “one could not expect the portfolio adjustments of banks and their customers to proceed so rapidly as to offset unforeseen shifts in the demand for money within a four to five week interval” (memo, Alan Holmes to FOMC, Board Records, June 3, 1973, 2). As before, the manager put the interests of the banks ahead of inflation control.

An October memo from Arthur Broida to Governor Sheehan explained the way the procedures worked in practice. Its summary explained that “the Committee has not felt rigidly bound to the terms of some precisely defined experiment. Rather, it has felt free both to modify the general framework of the experiment on the basis of experience, and to adapt the approach to the special circumstances that arise from time to time” (memo, Arthur Broida to Governor Sheehan, Board Records, October 26, 1973, 6).

A System committee reported on the effects of lagged reserve accounting on monetary control. The members agreed on the direction of effect but differed on the magnitude. The Committee concluded that lagged reserve accounting:

(a) significantly reduces the ability to hit a total reserve or RPD target . . . ;

(b) is a less significant limitation on the System’s ability to control reserves and monetary aggregates over the longer run;

(c) adds to the tendency for day-to-day money market variability; and

(d) increases somewhat the range over which the Federal funds rate needs to fluctuate if monetary aggregates are to be controlled by use of a reserve handle.(First Report of the Staff Committee on Lagged Reserve Accounting, Board Records, August 10, 1973, 2)

The Committee estimated that, short-term, a two-week lagged reserve system caused the federal funds rate to vary from ten to twenty-five basis points more than contemporaneous reserve accounting (ibid., 10). The report emphasized the Committee’s uncertainty about the estimate, but it recommended returning to contemporary reserve accounting. The princi
pal objection cited in the report was bank relations. Some banks preferred the lagged arrangement because they had better control of required reserves and could hold smaller balances on average. The banks’ interest prevailed.

The FOMC reinstituted control of reserves in 1979. Despite the 1973, report it retained lagged reserve requirements and did not restore contemporaneous reserve accounting until after returning to control of money market conditions, when it no longer mattered for monetary control.

The memos make clear that senior staff understood the problems with System efforts to increase control of monetary aggregates. The System would neither permit wider variation in the federal funds rate nor adopt procedures that made reserve control easier.

NOW
Accounts

The combination of inflation and regulation created opportunities and heightened incentives to circumvent regulation. The Board responded by trying to limit the exceptions. By the end of the decade, there were numerous exceptions. Soon after, interest rate regulation ended.

In July 1970, a Worcester, Massachusetts, savings bank asked the state regulator to permit its customers to issue negotiable orders of withdrawal (NOW). The orders drew on the customer’s savings account and were payable through a commercial bank.
⁹⁶

The banking commissioner denied the request. After almost two years, the Massachusetts courts ruled in favor of the savings bank. Beginning in June 1972, savings banks opened NOW accounts. Deposits in NOW accounts rose rapidly but remained less than one percent of total savings deposits after eight months. New Hampshire savings banks followed soon after.

The new accounts posed competitive problems for commercial banks. Savings banks were not members of the Federal Reserve System, so they had lower reserve requirements and could hold reserves in interestbearing Treasury bills. Also, they could pay higher interest rates on savings accounts than commercial bankers. And regulation Q prohibited interest payments on demand deposits. Technically, NOW accounts were not demand deposits, but they were available on demand. The NOW orders transferred the balances as if by check.

Congress approved legislation in September 1973 that permitted the ex
ception. The Board limited NOW accounts to individuals in Massachusetts and New Hampshire, two states with many savings banks, and it limited interest payments to 5 percent a year. To restrict use, the Board limited the number of NOW orders on an account to 150 a year, and it restricted advertising of NOW accounts to Massachusetts and New Hampshire. Of course, several of these restrictions proved difficult to monitor and enforce. Nothing prevented a person from having more than one account.

96. The origin and development of these accounts here follows the report to President Frank Morris of the Boston bank by his staff, dated March 29, 1973 (Board Records, March 29, 1973).