The extraordinary real returns reflected both skepticism about the Federal Reserve’s ability or willingness to further reduce inflation or even to keep it from rising and the strong recovery from a deep recession. The high real rate was apparently balanced by an even higher expected return, especially to consumption in the event. The unemployment rate reached 10.8 percent at the end of 1982 and 10.3 percent in March 1983, when Wallich spoke. Experience in the 1960s and 1970s gave ample reason to believe that the Federal Reserve would again respond to the increased unemployment by inflating; hence consumer spending increased in anticipation of higher inflation.
Disinflation was incomplete. The economy was far from stable growth with low inflation. Interest rates reflected pervasive uncertainty and skepticism. The real trade-weighted exchange rate had increased 36 percent between third quarter 1982 and first quarter 1983, so the current account had moved to a large deficit.
The high rate of money growth reflected to an unknown extent the desired increase in real balances following the decline in inflation. Equations for the demand for money could not reliably separate the increase in money balances that the public wished to hold at the lower prevailing and anticipated rate of inflation and the desired reduction in money balances that presaged a return to higher inflation as reflected in the long-term rates.
The FOMC recognized that there was more to be done. It was more uncertain than usual about what to do and how to do it. Memories of the problems with an interest rate target and existing high real rates made an interest rate target unattractive. But the steep decline in monetary velocity and the continuing response to deregulation reduced support for a monetary target.
Wallich doubted that the monetarist predictions would be correct. Rapid money growth implied a strong economic advance followed later by higher inflation. “If I were to follow through the monetary implications of the recent monetary upsurge, I would have to say that the monetarist view is that half a year later the real sector would begin to move strongly. If that were the case, then we’d see something stirring now and certainly have very strong second and third quarters, which we don’t seem to expect, and a couple years later prices would begin to expand” (FOMC Minutes, March 28–29, 1983, 43).
His analysis was right, his forecast partly wrong. As in all previous periods since 1920, when real base money growth and real long-term interest rates gave opposite forecasts, the economy followed real base growth. This time was not an exception; it differed only in the unprecedented level of the real interest rate. Despite these rates, real GDP, pushed along by rapid money growth and the permanent tax cuts, rose at an average rate of 8.3 percent in the four quarters starting in second quarter 1983. At the time Wallich spoke, real first-quarter GNP growth was 3.5 percent followed by 9.3 percent in the second quarter. The deflator rose from 3.2 to 4.7 percent between first and fourth quarter 1983.
The Federal Reserve responded by raising the federal funds rate more promptly than in the past. The rate rose from 8.5 percent in February 1983 to 9.6 percent in February 1984, about as much as the deflator. But the funds rate continued to rise, reaching 11.6 percent in August 1984.
This rapid response hides the uncertainty that was a dominant feature of FOMC meetings. The members had a much clearer view of their responsibilities but a murky view of how to achieve them. It took several years to develop an operating procedure that improved their ability to maintain steady growth and low inflation.
Treasury officials during the early (1981–85) Reagan administration opposed exchange market intervention. Many at the Federal Reserve disliked this policy, but the Treasury controlled exchange market operations. Treasury Undersecretary for Monetary Affairs, Beryl Sprinkel, announced that the market would set the exchange rate. He limited intervention to a few periods of market disorder.
A major issue of the period resulted from the large deficits in both the budget and the current account. A popular and widely discussed explanation blamed the latter on the former, the so-called twin deficit problem. It was easy to demonstrate arithmetically that the current account deficit equaled the difference between private saving and investment plus the budget deficit. An increase in the budget deficit, with private saving and investment unchanged, increased the current account deficit. This was true ex post.
The problem with this explanation, as every economist should know, is that both deficits are endogenous variables that are dependent on responses to policy, interest rates, output, and other variables. The same forces that affect the two deficits also affect private saving and investment. For example, an increase in productivity growth increases output, raising tax collections and imports. The budget deficit declines and the current account deficit increases. Other things unchanged, the ex post relation will show that the current account deficit equals the difference between private saving and investment plus the budget deficit. But in this case investment rises, making it possible for the budget deficit to decline while the current account deficit rises. Data for the 1980s and 1990s show that the two deficits often moved in opposite direction. During the productivity surge in the late 1990s, the budget deficit became a surplus, but the current account deficit increased.
Meltzer (1993) studied changes in the trade-weighted real exchange rate using the analysis outlined in Friedman (1953). The principal factors affecting the exchange rate were the increase in military spending and the (lagged) increase in real money balances. Rearmament drew on domestic resources but also foreign resources, and imports substituted for domestic output devoted to rearmament. When the increases in defense spending slowed or reversed, changes in the real exchange rate reversed also.
Chart 9.1 shows the rise and subsequent decline in the real exchange rate. Disinflation and rearmament brought the real exchange rate to a peak in first quarter 1985. In the second quarter, the exchange rate reversed course, declining more than 6.6 percent. The new Treasury secretary, James Baker, reached agreement with other principal countries to intervene, so intervention thereafter reinforced dollar depreciation by the market. By fourth quarter, the rate was back to the level two years earlier, in third quarter 1983. The dollar continued to decline for the next two years. The Louvre agreement early in 1987 attempted to bound the nominal rate, but after a worldwide stock market collapse in October 1987 governments abandoned the effort. Chart 9.1 shows that the exchange rate remained in a narrower range to the end of the decade.
Chart 9.2 shows annual consumer price inflation. The chapter detail discusses the adjustment to reduced inflation and the rise in 1983–84 that sustained concern for a time that high inflation would return. The chart shows that the inflation rate stabilized at 2 to 3 percent only in the 1990s.
Chart 9.3 compares a well-known inflation forecast made a year in advance to reported inflation. The forecast consistently overestimated inflation in this period, perhaps reflecting skepticism about the Federal Reserve’s willingness or ability to maintain low inflation.
In the 1970s, monetary policy usually reacted quickly and decisively to the unemployment rate. Fortunately, the stronger-than-anticipated recovery in 1983 lowered the unemployment rate by 2.8 percentage points. Thereafter, the rate continued to fall until the end of the decade. Chart 9.4 shows these data. The continued decline in the unemployment rate was one of the factors that eventually reassured the public and the markets that the Federal Reserve was less likely than in the 1970s to increase inflation.
Chart 9.5 shows the response of long- and short-term interest rates to policy and other variables. Long-term rates did not remain below 10 percent until 1985–86. The skepticism noted earlier is especially apparent in the movement of long-term rates in 1983–84. In July 1984, the ten year government bond rate again reached 13.8 percent. As Chart 9.5 shows, this was the beginning of the end of the exceptionally high rates, but two years passed before ten-year rates returned to single digits. The interest rate data confirm the uncertainty that the pubic shared with the Federal Reserve as the latter worked to establish t
hat monetary policy would not repeat past mistakes.
Productivity growth, the rate of change of output per hour, typically rises following a recession, then falls back to its trend rate. Chart 9.6 shows this pattern, a steep rise in 1983 and 1991. By the late 1980s the average rate outside of recoveries fluctuated around 1 percent. Difficulties in forecasting variable productivity growth are one reason forecasting output growth and inflation is not very accurate.
Chart 9.7 compares two forecasts to actual values of real GNP/GDP growth. The green book forecast is made by the Board’s staff, the SPF by private forecasters. The chart shows that the two are broadly similar and both differ much more from actual growth than from each other. The Federal Reserve at first persisted in adjusting its actions to its forecasts despite evidence of frequent, large, and persistent errors. Volcker then adjusted his actions to reports of actual data. The chart shows that errors in forecasts of GDP during the disinflation were exceptionally large, but sizeable errors were not unusual. Errors were exceptionally large in 1983, a period of great uncertainty.
A principal reason for these errors was reliance on a Phillips curve to separate output growth and inflation. The inflation forecasts in Chart 9.8 show again that the forecast error was often much larger than the difference between the two forecasts. The persistence of positive and negative errors in the 1970s or the 1980s is notable. The Federal Reserve and the SPF persistently underestimated the inflation rate when the inflation rate rose in the 1970s, then persistently overestimated inflation as it fell in the 1980s and again in the 1990s.
During the early 1980s foreigners continued to purchase and hold 15 to 20 percent of outstanding Treasury debt. Chart 9.9 shows that foreigners financed a sizeable part of the large Reagan era budget deficits. One of the likely reasons that researchers do not find much of an effect of budget deficits on domestic interest rates is that substantial foreign purchases substitute for increased domestic purchases. The chart shows that foreigners became larger holders in the 1990s and in the new century during years of budget surpluses in the late 1990s and the renewed deficits after 2001.
PERSONNEL
Two Board members left in the early 1980s. Governor Lyle Gramley resigned after little more than five years of a fourteen-year term. President Reagan appointed Wayne Angell to the position. Angell served from February 1986 to February 1994, then resigned to work in the financial services industry. In July 1984, Martha Seger, a Michigan economist, joined the Board, replacing Nancy Teeters, who resigned to accept an appointment as a corporate economist. Seger resigned in 1991 after more than 6.5 years as a governor. In 1993, Seger paid a civil fine for violating the one-year ban on lobbying after resigning. She arranged and attended a meeting with the Board in August 1991 as a director of Kroger and Co. During his eight years, President Reagan appointed all seven governors, the first president since Franklin Roosevelt to do that. Table 9.1 lists personnel changes from 1983 through 1985.
On June 18, 1983, President Reagan reappointed Paul Volcker to four more years as chairman. This was a surprise. Several of the administration heavyweights—James Baker and Donald Regan especially—found Volcker difficult to work with and insistent on independence. Although Volcker reduced interest rates, he did not agree to ease policy before the 1982 election, responding belatedly from the administration perspective. What would he do in 1984? They were not sure, and they wanted a more agreeable chairman who would at least listen with more understanding of their wishes. But Volcker had support from the bankers, financial markets, and President Reagan for his success in reducing inflation and also from budget director David Stockman. And he was helped by his opponents’ inability to agree on an alternative.
Volcker did not request reappointment in 1987. Coyne (1998, 14) said he thought Volcker could have had reappointment if he had asked. “He always said he would never ask for a job.” Greenspan (2007, 98–99) reports that James Baker interviewed him for the chairmanship in March 1987, months before Volcker decided to leave. Speculation at the time suggested that Volcker often disagreed with the majority of the Board appointed by President Reagan. Volcker was accustomed to making decisions with advice from the senior staff, not the other governors. The governors could affect the decision made at the meeting. Between meetings, Volcker had authority to act.
Five bank presidents began five-year terms in 1983–85. E. Gerald Corrigan, a Volcker protégé, moved to New York from the Minneapolis bank, and Karen Horn was the first woman appointed as a bank president. Larry Roos and William Ford, two leading monetarists, retired.
POLICY ACTIONS 1983
The FOMC had given up on monetary targeting without a clear idea about whether the change was permanent. The members recognized that monetary velocity had fallen. They did not have a good explanation of the reasons, and they never mentioned the decline in inflation as a reason for holding more money balances per unit of income, thus lower velocity. The most common explanation was a sense that the new types of deposits now included in M1 had lower velocity, but the staff did not present evidence at the meetings. Nevertheless, they accommodated the increased demand for money.
What should replace the monetary target? The FOMC in 1983 could not agree and was often divided. It did not want to return to an interest rate target. Many expressed concern that an interest rate target would expose them to another surge of inflation. Others expressed concern that the market would interpret return to an interest target as the end of anti-inflation policy.1 A few members disagreed. They said that in practice the desk used a funds rate target; they urged that the desk make it explicit (FOMC Minutes, February 8–9, 1983, 27). In fact, members proposed different targets and indicators as in the 1950s. Most wanted “flexibility.” In practice this meant that they did not want a precise target and would respond to new information. Uncertainty and ambivalence often left the decision to Volcker.
The Humphrey-Hawkins legislation required the FOMC to announce a money growth target. When Volcker testified to the Senate Banking Committee on February 16, 1983, he announced a 4 to 8 percent range for the year to fourth quarter 1983. He did not tell Congress explicitly about flexibility or the uncertainty about the appropriate target. He told them that “some allowance should be made . . . for the uncertainties introduced by the existence of the new deposit accounts” (Monetary Policy Report to Congress, Federal Reserve Bank of New York, Box 97657, February 16, 1983, 27). The Federal Reserve, he said, would watch to see if “velocity behavior is resuming a more predictable pattern” (ibid., 28). He continued to insist, however, that “inflation requires appropriate restraint on growth of money and liquidity,” but he quickly added that the short-term relation between money and spending “may be loose” (Volcker papers, Federal Reserve of New York, Box 97649, May 16, 1983, 14).
1. Thornton (2004) summarizes the literature on the choice of target after October 1982. He argues that the Federal Reserve used an interest rate target after October 1982. Others choose later dates, 1987 or 1991. The FOMC chose a band for the federal funds rate in 1982–83 and after, but it had done that in 1979–82. The minutes make clear that the target was borrowed reserves most of the time until the late 1980s. In 1994 and thereafter, the FOMC announced a funds rate target and kept the actual funds rate very close to the target. This was not so in the 1980s, particularly in the first half of the decade.
Commitment to achieve the money targets was no greater than before. Many on the FOMC did not think they had a role. At the December 1982 meeting, President Black (Richmond) proposed a federal funds rate target. Volcker opposed. He distinguished “between having an explicit interest rate target and having . . . some limits of tolerance on what interest rate change one wants” (FOMC Minutes, December 20–21, 1982, 41).
What was the desk supposed to do? In practice, the Committee’s decision at the time was a return to the 1920s program of setting a target for member bank borrowing, very much like the 1950s choice of free reserves. David Lindsey (2005, 15) was responsible for estimatin
g borrowing. Wallich (1984, 11) claimed that after October 1982 the target was borrowed reserves. This acted as a loose target for the federal funds rate. The reason is that to the extent that the staff (Lindsey) had a useful analytic expression for borrowing, borrowing depended on the difference between the federal funds rate and the discount rate plus a random error. The error was often large, and at times the equation was not stable. Nevertheless, it offered a choice. The FOMC could set the funds rate and accept the level of borrowing or set borrowing and allow the funds rate to vary within a wide band.2
As was typical in the 1950s, the FOMC did not discuss the relative merits of different choices until much later. The minutes make clear repeatedly that Volcker did not want an interest rate target, mainly out of concern for the likely market interpretation that policy would repeat the errors of the 1970s. Unlike the 1950s, however, there was a thorough discussion of individual preferences at the February FOMC meeting.
A History of the Federal Reserve, Volume 2 Page 65
