Members of FOMC again as in the 1950s used different measures to express their policy preferences. Some used M1; Wallich commented on the return of free reserves; he preferred gross borrowing, the measure used by the manager. The FOMC could agree only on a directive that named different measures but they did not know whether they were consistent. The vote was ten to one. Boykin (Dallas) favored tighter policy. Nancy Teeters had left the Board, and her successor was not confirmed. Martha Seger joined the Board in July.
17. At about this time Congressman Jack Kemp organized some Republican members to propose legislation to limit Federal Reserve independence, including returning the Secretary of the Treasury as an ex officio member of the Board. The legislation did not advance, but the threat of congressional action always alarmed the Board.
18. Federal Reserve response to Continental Illinois’s problems is discussed below.
Despite the decision to make no change, the federal funds rate rose in June and July. The average July rate, 11.23 percent, was 0.91 percentage points above May. Actual and anticipated inflation continued to fall slowly. The ten-year Treasury rate at 13.4 percent had not begun to reflect lower inflation. The S&P 500 fell sharply in July and several subsequent months as the economy began to grow more slowly and investors worried about widespread financial fragility. The staff forecast anticipated modest interest rate increases, but predicted a 15 percent decline in the exchange rate.
The July 1984 meeting had to decide on the projections for growth and inflation announced to Congress and the public. This time Board members’ median expected growth for 1985 fell to 3.75 percent from 6.75 in 1984; median predicted inflation was 3.75 rising to 5 percent in 1985. The medians for the presidents were similar. The precise meaning was unclear because the members made different assumptions about dollar depreciation. Expected dollar depreciation added up to one percentage point to the 1985 median inflation forecast. The members agreed to keep projected M 1 and M2 growth for 1985 about unchanged at 4 to 7 and 6 to 8.5 percent. This represents a slight reduction for the upper end of the M1 and M2 growth rates.19
Heightened uncertainty about growth and inflation, signs of slower growth, and increased banking problems made it difficult to reach agreement on proper action in August. Volcker took charge. After announcing ranges for the monetary aggregates, borrowing, and the funds rate, he told the Committee:
If we get a weaker employment number, weaker this number or that number—broadly lower than expectations, yes. If we have financial problems that are great enough, we would provide some liquidity. I don’t know how one judges that in advance. I think we have to play it by ear. (FOMC Minutes, August 21, 1984, 38)
19. Volcker’s testimony in the Senate blamed high interest rates on the budget deficit, not for the first time. He urged again that Congress reduce the budget deficit both as a way of achieving macroeconomic balance and to lower the size of the capital inflow. He warned about the inability to sustain the capital inflow and the risk of rising protection (statement, Paul Volcker to Senate Committee on Banking, Box 97645, Federal Reserve Bank of New York, July 25, 1984, 4).
Two years had passed since the end of the explicit disinflation policy. Measures of actual and anticipated inflation remained about 3 to 4 percent. Unemployment rates had declined from more than 10 to about 7.5 percent. But nominal interest rates had not returned to the pre-1979 ranges. In September, the federal funds rate averaged 11.3 percent, and the ten-year Treasury note yielded 12.5 percent, only one percentage point below its local peak. The ten-year rate remained almost two percentage points above the level in October 1982. Apparently, the public regarded the risk of inflation as very high. The FOMC had no plan for lowering these rates. Several of its members shared the concern that inflation would return.
Volcker stated their puzzle. “It’s a very strange period. Every indicator normally would be associated with lower interest rates” (FOMC Minutes, October 2, 1984, 6). Chart 9.10 above shows that the real interest rate remained in the 6 to 7 percent range. The real value of monetary base growth fell persistently during this period. In October, the staff began to reduce its forecast.
Though unremarked at FOMC, nominal wage growth had declined steadily from its peak in the early 1980s. By 1984, the nominal rate of change was about 4 to 5 percent at annual rates. Chart 9.11 shows these data at quarterly rates. This proved to be a harbinger of lower real and nominal interest rates. It is difficult to reconcile the expectations reflected in interest rates and nominal wage growth.
During the summer and early fall, St. Louis voted regularly to raise its discount rate to 9.5 percent. No other banks agreed with them. The Board recognized the large spread between the discount and the funds rate, but it rejected the requests because borrowing had declined and the economy showed evidence of slower growth. By mid-November evidence of slower growth prompted Chicago, St. Louis, Dallas, and San Francisco to ask for a reduction to 8.5 percent. New York, Philadelphia, and Kansas City joined them a bit later. The Board deferred the requests until November 21, when it approved 8.5 percent. The Board cited the slowing economy, the spread between the funds and discount rate, and slowing growth of M1 and M2.
By mid-December, three banks requested an additional reduction. The Board deferred the request but recognized that it might be appropriate after the FOMC meeting. On December 21, the Board voted for an 8 percent rate, citing the same factors as in November. Governor Gramley dissented because he thought that slower growth was temporary.
Monetary policy continued to lack a consistent approach during the last six months of 1984. In the usual procyclical patterns, the FOMC permitted base and M1 growth to decline as the economy slowed. Twelve-month base growth fell to 6.8 percent, two percentage points less than in July. Falling nominal base growth did not change the inflation rate. Inflation remained about 4 percent as shown by the difference between nominal and real base growth in Chart 9.12. The SPF predicted that 4 percent inflation would continue.
The federal funds rate declined also. Its December 1984 average, 8.4 percent, was almost three percentage points below July. By November, the ten-year Treasury rate was below 12 percent. Real long-term rates began to decline slightly as inflation fell.
For the July Humphrey-Hawkins hearings, the Board forecast a rise in inflation to 5 percent in 1985; the presidents forecast a 5.25 percent median with an outlier at 6.5 percent. The staff forecast 4.75 percent. Members expressed concern about a rapid decline in the dollar exchange rate, a credit binge, wage increases, and always the budget deficit.20 Most of all, the members expressed uncertainty about the future and the difficulty of making forecasts. No one expressed sentiment for following a long-run policy path, although some of the remarks about more “automaticity” could be interpreted that way.
Volcker favored reducing slightly the 1985 M1 range from 4 to 8 percent to 4 to 7 percent. Boehne (Philadelphia) objected that would make an increase difficult in February 1985. Volcker responded:
Well what would happen next February? None of us knows. Let’s suppose we’re running above that [range] . . . ; that must mean we’ve worried about something—interest rates or something in the economy. In that kind of scenario we would go [to Congress] and say “Look you haven’t done anything about the budget. We can’t hold it by monetary policy alone without running undue risks. We’re going to have to raise the target.” (FOMC Minutes, July 16–17, 45)
Volcker told the FOMC that his concerns about the banking system kept policy from tightening. Martin showed that he was in the wrong job. He said that the FOMC had a bias toward tightening. He quoted people who called the FOMC “knee jerk inflation fighters” (ibid., 67). Volcker replied that it’s a good reputation to have, and Solomon said, “[T]hat’s what central bankers are supposed to have” (ibid., 68).
With Martin dissenting, the Committee voted to tighten policy slightly to slow economic growth. The exchanges at this meeting suggest the major change that had occurred since the 1960s and 1970s. Volcker’s anal
ysis of inflation and the deficit was much the same as William McChesney Martin’s. The difference was that he didn’t coordinate, emphasized the importance of controlling inflation, and was willing to act within the limits set by political concern about the banking system and the international debt.21
Lyle Gramley continued to express concern about inflation. He warned his colleagues not to repeat the two mistakes made by the Carter administration, where he had served. He described the mistakes as (1) underestimating the natural rate of employment as 5.5 percent, when it was higher, and (2) overestimating productivity growth. “So I think we ought to be very, very careful in looking at recent productivity statistics and not to get overly optimistic” (ibid., August 21, 1984, 26).
20. Volcker noted, “All FOMC members are expecting a big decline in the dollar” (FOMC Minutes, July 16–17, 1984, 30).
21. In August the Committee expressed concern about the large volume of collateralized borrowing by Continental Illinois. The loans pressed against the rates on collateral for Federal Reserve loans because the loans were secured by government securities.
By November, the Treasury had started to intervene in the exchange market as part of a consortium that wanted to depreciate the dollar. The Federal Reserve always sterilized intervention, usually including Treasury intervention, so the actions had no more than a brief effect. The dollar continued to appreciate.
By October, slowing activity and declining growth rates for M1 and M2 prompted an early move toward easier policy action. The federal funds rate declined on average in both months. Axilrod said that money demand had shifted downward, but he offered no explanation of the alleged shift. Several members began to emphasize M1 growth. Roberts (St. Louis) explained that M1 had not grown for five months. He blamed the use of borrowing and free reserve targets, and the decision to slow the decline in the funds rate. Morris (Boston) also blamed use of a borrowing target, citing the large increase in borrowing to support Continental Illinois.
Volcker then asked, “Do you want to target the federal funds rate? . . . I wouldn’t recommend it” (FOMC Minutes, November 7, 1984, 30). He defended FOMC actions, pointing out that higher-than-expected interest rates had occurred for several months without objection from the members. Others also objected to the control mechanism, but there were no agreement on what to do differently and no discussion of the reasons for their differences.
The discussion resumed in December. Axilrod prepared a paper on two problems with using a target for borrowing. One arose because the relation between borrowing, interest rates, and money changed at different levels of economic activity. At lower levels, he claimed that banks were less willing to borrow, so interest rates remained higher. The banks did not want to show large borrowing on their balance sheets, a version of the old “reluctance” theory from the 1920s. Second, higher interest rates reduced the quantity of money demanded. Axilrod also suggested using the monetary base as a target because the market focused on M1. He proposed more automaticity in their procedures.
This set off a discussion of the use of rules and discretion that was a rare event for the FOMC. The discussion also brought out criticisms of their procedures. Frank Morris described their procedure. The staff estimated required reserves, added an estimate of excess reserves, and subtracted expected borrowing to get nonborrowed reserves. The error in estimating borrowed reserves reduced total reserve growth because the desk aimed for nonborrowed reserves.
Volcker objected that the FOMC had seen the result of these procedures at meetings from July to October without responding. Why change procedures?
John Balles favored introducing more rule-like behavior. He claimed that the more basic problem came from the use of borrowing or free reserves as a target. “Any given level of borrowing or any given level of free reserves is compatible with a wide range of interest rates, a wide range of economic outcomes, and a wide range of growth in the monetary aggregates.” He reminded them that the same problem arose in the 1950s, when free reserves were the target. He preferred nonborrowed reserves because it was more closely related to money.
Robert Black supported his proposal for a semi-automatic response to unforeseen changes in nonborrowed reserves. He urged a penalty discount rate also. Corrigan, Forrestal, and Gramley opposed Axilrod’s proposal.
Henry Wallich said “good judgment is always bound to be better than a mechanical rule. The question is whether good judgment is more likely than not” (ibid., 15). He thought judgment had been correct recently, so they should not change.
It is striking that this discussion took place seventy years after the Federal Reserve began and fifty years after abandoning the gold standard rule. No consensus had formed about correct procedures. No less striking was the lack of agreement about the use of rules and absence of any reference to the Kydland and Prescott (1977) paper showing the superiority of rules over single-period discretionary choices. The members may not have been aware of the paper, but the staff certainly was.
Volcker then turned the discussion to the use of an M1 target to the exclusion of everything else. He summarized: “I see a few hands go up weakly and then they go down again” (ibid., 20). In his view the main problems came from the difficulty of forecasting accurately. “You can’t project the economy quarter-to-quarter or half-year to half-year with a degree of sensitivity that is required ex post to make everybody happy. That’s another way of saying that you can’t fine-tune on the basis of economic projections” (ibid., 21).
This statement could have started a discussion of aiming at a target more than one quarter or half year ahead with less attention to short-term changes. But Volcker did not choose that path. Instead, he emphasized using information in “other indicators of what is happening” (ibid.). He mentioned the exchange rate and commodity prices, but he did not suggest how to interpret the information without a more comprehensive and accurate model than they had or were likely to develop.
After the discussion, the FOMC considered its policy for early 1985. The only sign of the preceding discussion was the increased number of references to the exchange rate. Governor Gramley emphasized that their concern should be the real exchange rate and that he did not know how monetary policy could change the real exchange rate.
The staff forecast growth of 2.9 percent in 1985. The Committee returned to business as usual, some wanting faster growth, some anticipating faster growth, and some favoring more attention to inflation. Preston Martin again introduced a political element by citing congressional response to a forecast of 2.9 percent growth when the unemployment rate remained above 7 percent. His concern was that there would be efforts to restrict independence.
The discussion, as so many others, turned on a narrow issue—a proposed borrowing target of $250 or $300 million. The issue was whether the funds rate would ease or remain about 8.5 percent. The vote was ten to two with Solomon favoring higher borrowing and Gramley favoring more attention to inflation. The FOMC did nothing to resolve their differences about long-term policy.
BANKING PROBLEMS
Persistence of high real interest rates and disinflation brought problems in the global financial system to attention. There were two distinct types of problems and solutions. Large-scale borrowing by developing countries during the 1970s became very expensive to service in the 1980s, when real interest rates rose. Beginning with Mexico in 1982, countries defaulted. Many of the debts were owed to large U.S. banks, especially New York banks, but European and Japanese banks faced similar problems. The Federal Reserve’s strategy was to use international loans, especially IMF loans, to pay the interest to keep the countries from defaulting. The countries’ debt increased but the new money went to the creditor banks. A later section discusses this problem.
Regulators treated domestic banking problems very differently. Mergers and closing of failed banks and thrifts or massive loans to restore solvency became standard practice. Among the failures was the Continental Illinois Bank, the largest Chicago bank. But it was no
t alone. The rise in oil prices had induced a real estate boom in Texas and Oklahoma. When oil prices fell and the boom collapsed, home prices followed. Many home mortgages went into default. If no local owners offered to buy failed or failing banks, out-of-state banks could buy them. Later, interstate banking became an accepted outcome even if the acquired bank remained solvent. This was a major change for the better in U.S. banking practice. It permitted much greater portfolio diversification and reduced the risk that difficult local economic conditions would result in bank failures or systemic financial fragility.
Volcker was on vacation in early August, when he received a visit from the chairman of Continental Illinois Bank, who told him the bank would require federal assistance to avoid failure (Volcker and Gyohten, 1992, 200). Continental had purchased many of its energy loans from Penn Square. When Penn Square Bank in Oklahoma failed, news of Continental’s problems spread through the financial markets. It faced a risk premium in the market for the large purchased deposits that it used to finance its activities. The bank could not replace the deposits that holders withdrew as they matured. By the end of the week, the Federal Reserve Bank of Chicago made $3.6 billion in loans to keep Continental Illinois operating. This was an unprecedented loan for the System. In July, the FDIC assumed $3.5 billion of debt.
A History of the Federal Reserve, Volume 2 Page 69