A History of the Federal Reserve, Volume 2

by Allan H. Meltzer

  24. Volcker did not propose to let the federal funds rate change without restriction on the size of the change. He told the FOMC: “In this process the federal funds rate is going to be constrained but constrained over a substantially wider range than has been our practice. . . . We operate on a day-to-day basis with almost no range at all. . . . In practice, the kind of range we have is one-eighth of a point roughly” (FOMC Minutes, October 6, 1979, 26). The AxilrodSternlight memo used similar language.

  25. The System returned to contemporary reserve accounting in October 1982 just as they ended the policy of targeting a reserve aggregate. It no longer mattered. Lindsey speculates that pressure from Beryl Sprinkel, the first Undersecretary of Monetary Affairs in the Reagan administration contributed to the change (Lindsey, 2005, 5). Axilrod’s first presentation to the Board analyzed lagged reserve accounting and came out against it, but Martin adopted it (see Chapter 3 ). Axilrod continued to recommend contemporary reserve accounting without effect.

  At a special FOMC briefing on November 19, Axilrod and Sternlight explained how they controlled money growth. A straightforward way would have set a target for total reserve growth that would reduce inflation. If banks increased borrowing, the manager would supply fewer reserves by open market operations. To increase control, the Board would restore contemporary reserve accounting. A penalty discount rate would facilitate total reserve control. Changes in the multiplier relating reserves to the money stock would be offset by raising or lowering reserve (and base money) growth.

  That was not the procedure adopted by the staff. Axilrod explained that the staff estimated the demand for money, believed to be consistent with the System’s objectives, computed reserve growth to satisfy the demand for money, corrected for seasonal adjustment to get non–seasonally adjusted reserve growth as a four-week moving average, and assumed a value for member bank borrowing based on recent behavior to get unborrowed reserves.26 This procedure introduced possible error at several places, and it gave too much attention to very short-run changes in the demand for money and too little to control of sustained money growth necessary for controlling inflation. Also, it required adjustments for float, currency, and shifts in the composition of deposits subject to different required reserve ratios. The manager added that the desk did not try to control the weekly data; it aimed at a four-week average (FOMC Minutes, appendix, November 19, 1979).

  The Board’s staff improved models of borrowing, but the FOMC did not use any until 1981. One reason was that the Federal Reserve treated borrowing as a privilege, not a right of membership. The meaning of privilege changed from time to time and introduced non-market considerations into banks’ demand. Also, the System changed the rules to permit smaller banks to borrow more easily, especially for seasonal purposes. The difficulty in estimating the demand for borrowed reserves was well known at the time. Axilrod and Sternlight avoided econometric estimates of borrowing. See Hamdani and Peristiani (1991, 59–61), who explained why the borrowing relation was non-linear and more variable at high levels of the spread between the federal funds and discount rates, a problem during much of 1979–82. Research at the Board cast doubt on these findings.

  26. The staff was unable to estimate a reliable borrowing relation or a short-run demand for money. Axilrod noted that lagged reserve accounting made it “impossible” to hit the total reserve target. He argued, however, that it made little difference over three to six months (FOMC Minutes, appendix, November 19, 1979, 5–6). Control over longer periods would be less affected, but Axilrod’s statement ignores the market’s response to the short-run changes in reserve growth. Mark Willes (Minneapolis) commented on the procedure at the FOMC meeting the next day. “What we would end up with are reserves that are totally demand determined. . . . [W]e are trying to do just the opposite . . . to make reserves and, therefore, the aggregates supply determined. Yet the mechanisms we have set up would have just the opposite result” (FOMC Minutes, November 20, 1979, 20). Willes expressed skepticism as to whether the procedure could work except by chance. Important as his comment was, it did not elicit an answer and was ignored.

  No less important, the System misunderstood the effect of borrowing. An implication of the Axilrod-Sternlight framework was that increased member bank borrowing was contractive because interest rates initially rose. This is very reminiscent of the Riefler-Burgess framework from the 1920s, or control of free reserves in the 1950s, when the System exercised control by forcing banks to borrow and, it claimed, to contract. The reasoning was now somewhat different, but the implication was the same: an unchanged value of total reserves was more expansive if member bank borrowing was smaller. The staff did not offer an adequate explanation. As in Riefler-Burgess, they presumed that as borrowing increased, total reserves unchanged, interest rates would rise.27 One might think that having adopted a money stock target based on the monetary base or total reserves that members would have discarded this old error. That did not happen.

  If the demand for money declined in recessions, the staff projection would call for a decline in reserve growth, deepening the recession. Similarly, in expansions, money growth would rise with the demand for money. This procyclical policy characteristic of nominal interest rate targeting, as practiced by the System, remained.

  At the October 6 meeting, Volcker again emphasized the temporary nature of the procedural change and the importance of reducing inflation and of acting that day. “We can’t walk away today without a program that is strong in fact and perceived as strong in terms of dealing with the situation” (FOMC Minutes, October 6, 1979, 5). He did not equate “strong” with a shift to reserve targets; he said repeatedly that he could continue with traditional procedures, although the Board had accepted the change earlier.28 The main reasons for change were to give a psychological push to the idea that inflation would slow or end. Volcker put much emphasis on market psychology and its effect on anticipations.

  27. “I would say that we aim at a given volume of reserves and that the composition of that volume of reserves makes a difference. If they’re all nonborrowed, it’s more expansionary than if a higher percentage of them is borrowed” (Henry Wallich, FOMC Minutes, January 8–9, 1980, 7). The account manager agreed, using the same fallacious argument about bank reluctance to borrow that Riefler (1930) had used. By March, average borrowing exceeded $2.5 billion.

  28. “I am prepared, within the broad parameters, to go whichever way the consensus wants to go so long as the program is strong, and if we adopt the new approach so long as we are not locked into it indefinitely” (FOMC Minutes, October 6, 1979, 10).

  “Mr. Eastburn . . . There is a credibility problem if we launch this and stop and go with it. So I really think we are committed to this if we go forward.

  “Chairman Volcker: Well I don’t want to accept that” (ibid., 15).

  He was disappointed. He expected an increase in short-term interest rates, but he was disappointed by the big rise in long-term rates. “They went up much more than I would have expected” (Volcker, 2001, 3) “You have this little illusion. You were taking a tough measure, and the market ought to respect it. You’re the new Chairman of the Federal Reserve. It ought to have a salutary effect on expectations. Of course, it didn’t” (ibid.).

  The new program had three parts. First, the FOMC voted to adopt the new procedures and to bring M 1 growth to about 5 percent in the fourth quarter. The initial vote was twelve to five in favor, and eight to three among the voting members. Some of those who voted “no” wanted to raise the funds rate but not adopt the reserve target. All three then voted for the program so that it could be announced as approved unanimously. As part of the program, the FOMC permitted the federal funds rate to increase from about 11.5 percent to a range of 11.5 to 15.5, with an understanding that the FOMC would consult if the rate moved above 14.5 percent. Reflecting heightened concern, Nancy Teeters, who voted against a 0.5 percentage point increase in the discount rate in September, proposed that range. She favored a mone
tary target because it permitted the funds rate to fall.29 Partee agreed.

  Second, the Board met after the FOMC meeting. It approved a request from New York for a 12 percent discount rate, an increase of one percentage point. This was an unusual move intended to demonstrate that a strong program had started and to adjust the discount rate for the expected increase in market rates when the new program began. By the time the board voted on the New York request, Philadelphia, Cleveland, Richmond, Minneapolis, and San Francisco requested the 12 percent discount rate also. Within a few days, all reserve banks were at 12 percent.

  Third, the Board raised marginal reserve requirement ratios by 8 percentage points for increases in managed liabilities at banks with $100 million of such liabilities.30 This level eliminated all small banks, a move intended to disarm some congressional critics.

  If the FOMC hoped for a positive response, the markets disabused them quickly. On Monday and Tuesday the Dow Jones average fell about 4.5 percent. Gold prices fell at first but rose soon after. Reflecting the pervasive uncertainty and difficulty of interpreting what the announcement meant in practice, short-term rates fluctuated over a wide range. Long-term rates rose.

  29. Teeters (1995, 37) later wrote: “I was very concerned over the very aggressive move to fight inflation in October 1979 and the slowness in the willingness to back off as the recession deepened and dragged on into the middle of 1982.”

  30. Managed liabilities included: (1) time deposits in denominations of $100,000 or more with one year or less to maturity; (2) federal funds borrowings; (3) repurchase agreements on U.S. government and agency securities; and (4) euro-dollar borrowings from foreign banking offices (Board Minutes, October 6, 1979, 6).

  A week after the Federal Reserve’s announcement, Charles Schultze sent a memo to the president describing his speech about inflation. He stressed voluntary wage price guidelines and fiscal restraint. He said very little about monetary policy (memo, Schultze to the president, Schultze papers, October 13, 1979).

  In testimony to the Joint Economic Committee on October 17, Volcker repeated his main messages about the importance of ending inflation, the need for persistence, the central role of monetary control in an antiinflation program, and the importance of public support. He did not hide his belief that inflation control would be costly and would require an adjustment of expectations.

  The Federal Advisory Council (FAC) favored the new program unanimously but warned the Board that it must persist, despite the coming election and the “likelihood of a recession.” If strong words and actions are not followed by results, “then holders of dollar-denominated financial assets in the U.S. and abroad will conclude that the recent changes are no more significant than the statements and policy changes of prior years which did not reduce inflation. Where rhetoric sufficed several years ago, tangible proof is now required” (Addendum to Board Minutes, November 1–2, 1979, 6). The FAC called on the Board to improve communications by announcing “its targets for the growth of the monetary base as well as for other aggregates” (ibid., 7).

  Why the Change?

  A group that divided sharply in September over a 0.5 percentage point increase in the discount rate to 11 percent voted unanimously less than three weeks later for a 12 percent discount rate, a federal funds rate band up to 15.5 percent and other restrictive measures. They would soon approve still higher short-term interest rates. Also, the FOMC had discussed for years whether to control reserves or interest rates. They now voted unanimously to control reserves. And at least some recognized why their earlier efforts at reserve control failed.31

  A sense of current or impending crisis often causes officials to adopt changes that they earlier rejected or even scorned. That brought floating exchange rates in 1971 or 1973. Now it brought “practical monetarism.” Although the FOMC did not adopt the procedures that monetarists advocated, they now accepted the importance of controlling inflation by controlling money, permitting much wider fluctuations in market rates, and distinguishing between real and nominal interest rates.

  31. Robert Mayo commented on previous experience with reserve control. “I think the RPD [reserves against private deposits] experiment . . . failed we were too timid on the federal funds ranges that we associated with it, and it killed itself” (FOMC Minutes, October 6, 1979, 17).

  Volcker was clear about the impending crisis. At the meetings on October 5 and 6, he referred repeatedly to the precipitate depreciation of the exchange rate and possible flight from the dollar and the rise in commodity prices. He didn’t claim there was a crisis; he expressed concern that one would come if inflation continued (FOMC Minutes, October 6, 1979, 12).

  Those who dissented at the September 18 meeting agreed that the commodity markets and exchange rate reacted negatively to their dissent. They too believed a crisis might occur. They supported a policy change that they would have opposed strongly three weeks earlier. Some, including Volcker, recognized that the FOMC was unlikely to vote for interest rates high enough to reduce inflation. By choosing a reserve target, it could blame the market for the level interest rates reached.32

  It was also an opportunity to make a major change. The president did not object openly. Key legislators favored monetary control (ibid., 8, 9). European policymakers expressed alarm at the fall in the dollar and urged decisive action (ibid., 17). Perhaps most important of all, the domestic public expressed concern about inflation. Data from Gallup polls starting in 1970, when annual inflation reached 6 percent, show only 14 percent named inflation or “the high cost of living” as one of the country’s most important problems. The percentage rose and fell with reported inflation. It did not remain persistently above 50 and as high as 70 percent until 1980–81.33 Volcker persuaded his colleagues to seize the moment.

  The Major Change

  The October 6 meeting did not dwell on the most important change. Perhaps without recognizing it, the System implicitly changed the weights on unemployment and inflation. It now regarded control of inflation as its principal current responsibility. That had happened before. The FOMC recognized that its previous efforts failed because it did not persist when the unemployment rate rose. The change in operating procedures intended to signal the change in the System’s commitment to put greatest weight on inflation and expectations of inflation. Orphanides (2005, 1021) presents some evidence of this change. The change in objective was much more important and more durable than the change in procedures.

  32. Charles Schultze described the procedural change as “a political cover. They’re not monetarists, but it allowed them to do what they could never have done. . . . They could never have done what had to be done if it looked as if they were the ones raising interest rates . . . But with fixed monetary targets they could just say, ‘Who us?’” Volcker disliked suggestions that it was a public relations move to avoid blame for the rise in interest rates. “I never thought it was that. . . . It was a very common thing to say that we just did it to obfuscate” (Mehrling, 2007, 178; Hargrove and Morley, 1984, 486). Schultze also described the administration’s guidepost policy. “We preached and promoted and jumped up and down, but with little effect” (ibid., 488).

  33. I am indebted to Karlyn Bowman of the American Enterprise Institute for retrieving the polling data.

  The FOMC may not have recognized the change, but Volcker certainly did. In response to a question from the press, he rejected the Phillips curve tradeoff as a useful tool. Even more than in his colloquy on Face the Nation, cited earlier, he emphasized the centrality of ending inflation.

  Question: How high an unemployment rate are you prepared to accept in order to break inflation?

  Chairman [Volcker]: That kind of puts me in a position of I accept or unaccept or whatever. You know my basic philosophy is over time we have no choice but to deal with the inflationary situation because over time inflation and the unemployment go together. . . . Isn’t that the lesson of the 1970s? We sat around [for] years thinking we could play off a choice between one
or the other . . . It had some reality when everybody thought prices were going to be stable . . . So in a very fundamental sense, I don’t think we have the choice. . . . The growth situation and the employment situation will be better in an atmosphere of monetary stability than they have been in recent years. (Volcker papers, Federal Reserve Bank of New York, speech at the National Press Club, Box 97657, January 2, 1980, 6)

  The Federal Reserve now claimed that a policy of maintaining low inflation would increase employment in the long-run. Instead of trading off higher inflation to get lower unemployment, policy would lower both. Twelve years after Friedman’s (1968b) insistence on the effect of expectations, the Federal Reserve not only accepted that it could not permanently reduce unemployment by increasing inflation, but it now claimed that low inflation increased employment. Other leading central banks did the same. The way was open for inflation targets and other ways of recognizing that the principal, but not only, responsibility of a central bank was to maintain the value of money.

  The changes in the Federal Reserve’s perception of its responsibility eventually produced good results. In the following twenty-five years, the United States experienced two very long expansions followed by two relatively mild recessions. The variability of output growth declined. The United Kingdom also had a very positive response to persistent low inflation. The new anti-inflation policy remained in place until 2004.34