The Federal Reserve had to choose between two unattractive alternatives. Either they made the Mexican loan or Mexico would default on its debt. The first put the Federal Reserve at risk, but it postponed and perhaps prevented a crisis. Failure to lend meant heavy losses by large money center banks as well as others. Grieder (1987, 684) puts Mexico’s outstanding debt at $80 billion and the share held by the nine largest U.S. banks equal to 44 percent of their capital. Even the prospect of default was likely to reduce or eliminate the money center banks’ ability to borrow on certificates of deposits. A scramble for liquidity seemed likely. The Federal Reserve estimated that as many as 1,000 banks had lent to Mexico.
138. Federal Reserve governors were cynical about the promise. Preston Martin said, “I think we’re defining austerity as halting construction of four office buildings” (FOMC Minutes, May 18, 1982, 12).
In late June, the Federal Reserve increased its Mexican loan to $700 million. Neither Volcker nor the Mexican government had a long-term plan to prevent default. Even after the election, it did not offer a plan. By lending, it permitted lenders to reduce or eliminate their loans. Most were shortterm. As they came due, some lenders withdrew. And the more astute or better-informed Mexicans sold pesos and deposited dollars in U.S. banks. Federal Reserve loans financed these transfers.
The Mexican election did not resolve Mexico’s problem. Volcker told the FOMC on August 24 that Mexico needed more assistance to avoid default. Working with others, the Federal Reserve and the Treasury arranged $3.5 billion in additional loans and a willingness of commercial banks to roll over their loans. Some small banks did not go along, pushing the problem on to the money center banks.
Governor Partee questioned the use of the swap line, reminding the members that “we are precisely in the situation the Committee was somewhat concerned we might be in when that swap was first approved some fifteen years ago” (FOMC Minutes, August 24, 1982, 5). Volcker agreed. Repayment would require “a very draconian adjustment program” by Mexico (ibid.). The present swap was “secured . . . by a Fund agreement that doesn’t exist yet” (ibid).
Members asked about other countries. Volcker told them that Argentina had the most serious problem after Mexico. It did not have a swap line, and most of its loans were to non-U.S. banks.
The Federal Reserve, and Volcker personally, took an active role in the discussions that followed. All of the problem countries defaulted, but the IMF prevented them from stopping interest payments to the banks. Regulatory rules required the U.S. banks to write down the loans if interest payments stopped. Instead, the banks rolled over the loans, and the IMF made additional loans to the countries. The money did not go to the countries; it went to the banks to pay interest due. The countries’ debt rose.
Volcker and the IMF chose to put the solvency of the banks ahead of the debtor countries’ growth. For the rest of the decade income in the debtor countries stagnated. Volcker and the IMF never developed a procedure for ending this arrangement by writing down the debt.139 In 1989, the U.S. Treasury began to resolve the problem by offering the Brady plan, which reduced the debt.
The Volcker policy also put the interest of financial markets and banks ahead of the interests of the developing countries. The latter reduced their exports from the rest of the world. The United States was a major supplier, so much of the burden fell on U.S. farmers and manufacturers. Writing down the debt and lending to sustain the debt market would have been a less costly solution. Once again, the Federal Reserve neglected Bagehot’s advice.
CONCLUSION
Volcker’s statements as well as data leave little doubt that monetary policy changed in the summer and fall of 1982, even if Volcker hesitated to say so explicitly. The System’s public statements emphasized the difficulties in interpreting money growth, given the changes in regulation and the expiration of All Savers Certificates. Internal discussion gave greater weight to continued recession, rising unemployment (to above 10 percent), risks to money center banks caused by exposure to a Mexican (and other) default, risks to Federal Reserve independence, and the international effects of historically high interest rates. For these reasons, Volcker and others gave priority beginning in June 1982 to lowering interest rates. Internally, Volcker was more forthright. He answered those who thought the policy change mistaken because it reduced credibility just before an election.
I don’t myself perceive that the risks of misinterpretation are as great as some people think. Obviously they are there. . . . [F]ollowing a mechanical operation because we think that’s vital to credibility and driving the economy into the ground isn’t exactly my version of how to maintain credibility over time. Credibility in some sense is there to be spent when we think it’s necessary to spend it and we can carry through a change in approach. . . . We are dealing with the real world and assessing where the risks are. (FOMC Minutes, October 5, 1982, 50; see also Cukierman and Meltzer, 1986)
Volcker then described the risks, including the political risk to the Federal Reserve, if the recession continued or worsened. He acted for a combination of reasons, including political and economic concerns, and domestic and international risks.
139. As an acting member of the President’s Council of Economic Advisers, I proposed a writedown of the debt in the 1989 Economic Report of the President. This met strong opposition from the Treasury Department, so only a part of the recommendation remained in the published report. After the 1988 election, the Treasury accepted the writedown.
If we get this one wrong, we are going to have legislation next year without a doubt.140 We may get it anyway. . . . I think I know where the risks are. I’m not sure how it looks just in strict electoral terms, since the question has been raised, to sit here in some sense artificially doing nothing and then have to make a big move right after the election. . . . I’d prefer that this problem didn’t arise now. If business conditions looked a little better and interest rates were a little lower—and I wouldn’t care where the interest rates were if economic conditions looked a little better—and if we weren’t going to have to deal with a succession of sick foreign countries in this time period, if the dollar were not rising into the wild blue yonder right now, and if I thought that all these accumulating problems that we face could wait for a while, we’d have a much easier decision. . . . I don’t know what is going to happen in a number of directions over the next four weeks. (FOMC Minutes, October 5, 1982, 50–51).
The anti-inflation program became possible because President Reagan and, with the exception of credit controls, President Carter did not interfere. Leading members of Congress supported the policy, and those affected most—the homebuilders—reluctantly accepted the importance of reducing inflation. After three years, facing elections with rising unemployment rates and an incipient international crisis that threatened the United States’ financial system, political support wavered and waned. Legislation proposed restricting Federal Reserve independence or requiring it to target and reduce interest rates.
From the 1920s on, the System always responded to the threat of legislation. It was not helpless or without support. But Volcker did not wish to have a test of political strength. Inflation had fallen; the monthly CPI change for November and December 1982 was negative, and the twelve-month moving average fell below 4 percent, nearly ten percentage points below the peak. The December moving average was the lowest since 1973.
Volcker was cautious. He did not want to revive inflation or appear to make obvious policy changes before the November 1982 election. He lowered interest rates beginning in July, and watched the market and public reaction. The Federal Advisory Council told the Board in November that its policy was “appropriate”; it cited the current state of the economy, the level of unemployment, and the rate of inflation. “Moreover, we do not detect any significant loss of credibility on the part of the Fed in its fight on inflation” (Board Minutes, November 5, 1982, 5). However, the FAC warned that “stricter monetary targeting should be resumed in an early st
age of a recovering economy” (ibid.).
140. Volcker talked to Senator Byrd in August. The senator said he would not introduce legislation if interest rates continued to decline (Grieder, 1987, 514).
Although there was talk of inflation, the stock market boomed and longterm interest rates fell. That response encouraged continued ease and continued effort to forestall a banking crisis.
The market reaction has two possible explanations. One interpretation is that expectations of inflation had fallen in step with inflation. An alternative explained the positive market reaction as a response to the lower risk of a domestic and international financial crisis and a cyclical recovery in profits and output together with high, though reduced, inflation expectations.
Evidence supports the second interpretation. The SPF index put expected inflation near 6 percent at the time, a very high rate for peacetime. The ten-year Treasury bond rate remained above 10 percent until November 1985, three years later. In the interim, it remained between 10 and 14 percent, so real interest rates were 6 to 10 percent. These historically high real rates suggest that markets wanted to see sustained recovery without a return of high inflation. Some FOMC members shared the view that real rates would not decline until the recovery was complete without a return to inflation. By 1985, the economy had recovered. The unemployment rate had fallen to 7 percent, while inflation remained below 4 percent.
Goodfriend and King (2005, 982–83) report the wide gap between actual and published Phillips curve estimates of the output loss from disinflation. Okun (1978) summarized six Phillips curve estimates as showing that output would average 9 to 27 percent below capacity each year of the three-year disinflation. The total deviation of output would be 27 to 81 percent. The actual deviation was 20 percent, not small but smaller than even the most optimistic estimate based on published Phillips curves. The Reagan tax rate reduction limited the loss of output. The administration’s aggressive response to the air controllers strike helped also by reducing demands for wage growth.
The actual loss was almost certainly larger than required, and the recession probably lasted longer than necessary. The Federal Reserve’s control of money was imperfect and often absent. Its signals were hard to read. It did not make any of the institutional changes needed to improve monetary control until after it gave up monetary control. Better control procedures would not have removed all the random fluctuations in reserves and money, but lagged reserve requirements prevented the staff from controlling total reserves, as they reported to the FOMC several times. When reserve requirements had to be met, either the Federal Reserve supplied the reserves by open market operations or the banks borrowed the reserves. By keeping the discount rate often far below open market rates, the System favored borrowing. That left decisions about total reserves to the banks and reduced monetary control.
At least as important, institutional changes made it difficult to interpret money growth. Deregulation was long overdue, but its timing made growth of monetary aggregates difficult to interpret. The Federal Reserve was unfortunate in choosing to target money growth just at the time that the financial sector responded to deregulation. And it put too much emphasis on monthly or quarterly changes. As the charts show, growth of the real monetary base is a more reliable indicator during the period. The FOMC did not use it. Academic economists were probably too quick to conclude that the System could not control money. That conclusion puts too much emphasis on short-term changes.
A principal result was that the System made it much more difficult for the market to believe that the Federal Reserve was in control and would continue to reduce money growth until inflation slowed or ended. The System had abandoned anti-inflation policy on several previous occasions. Irregular base and money growth made its policy difficult to interpret. Table 8.11 shows the changes in real base growth and real ten-year Treasury yields dated at turning points in real base growth. Over a three-year period, base growth changed direction five times, often by relatively large amounts. On average, the change in direction occurred about every five months during the first two years. Anyone watching real base growth or real interest rates for evidence that the recession was about to end would have been uncertain until the prolonged increase in real base growth that began in September 1981 continued in 1982. See also Chart 8.5 above.
Although some members at times recognized that transitory changes in reserves, money growth, output growth, and measured inflation distorted current period data, few proposed to center attention on underlying permanent changes. Volcker and many others recognized at times that their goal was a lower trend rate of inflation, but they made no effort to emphasize the permanent trend or to ignore transitory changes. Their most important change was to put little weight on the unemployment rate, a decision that did not have unanimous support.
Despite these impediments, the effort succeeded to a considerable extent. Inflation did not end, but it fell to levels that had not been sustained for a decade. The FOMC and Paul Volcker deserve our praise for this achievement and for sustaining the disinflationary policy until inflation fell. The minutes show that Volcker had support from a few governors and presidents throughout, but he also had opposition much of the time. Success depended on his persistent concern to reduce inflation, and his firm belief that failure would make stability more difficult to achieve.
The 1979–82 disinflation demonstrates the importance of credibility in the implementation of monetary policy. As Goodfriend and King (2005) emphasize, the public did not fully believe that the Federal Reserve would keep its commitment to end inflation.141 Too many previous attempts were followed by the inflation rate heading to a new peak. The Federal Reserve added to the problem by failing to distinguish between a large one-time increase in the price level, resulting from the 1979–80 oil price increase, and the maintained rate of inflation. Reported rates of inflation would have declined even if the Federal Reserve maintained its earlier policy stance. The Federal Reserve succeeded in reducing the underlying maintained rate of inflation, however.
Continued high real interest rates suggest that more had to be done to convince the public that inflation would not return. Volcker’s lasting influence was (1) to increase the weight on inflation and lower the weight on unemployment, in the Federal Reserve’s objective function and (2) restore System credibility for controlling inflation. Central bankers in other countries made similar changes at about the same time. Central banks apparently learned that disinflation was costly to society and to them because they were blamed for the surge in unemployment rates and loss of output during disinflation.
The Federal Reserve did not yet have a consistent, transparent means of implementing policy. Procyclicality and excessive attention to near-term changes remained. There was ample reason, therefore, for skepticism about whether the reduction in inflation was permanent.
141. Hardouvelis and Barnhart (1989) used a Kalman filter to measure changes in credibility. They found that the public restored credibility slowly. Apparently, credibility depended on sustained reduction in inflation, not on announcements or the decision to abandon reserve control.
NINE
Restoring Stability, 1983–86
Mr. Wallich. . . . We have evidence of a very strong monetary expansion. On the other side is a high real interest rate. Which of the two really drives the economy?
—FOMC Minutes, March 28–29, 1983, 43
Chairman Volcker. I don’t think we know what GNP is going to be in this quarter.
—FOMC Minutes, November 4–5, 1985, 21
Inflation was greatly reduced but not eliminated by the end of 1982. The GNP deflator for fourth quarter rose at a 3.6 percent annual rate, down from a peak of 12.11 percent two years earlier. Real growth followed money growth, not the real interest rate. A strong recovery had begun.
The years of high inflation were over, but inflation and disinflation left problems behind. Four major problems continued in the 1980s. First, real interest rates remained far above historic
norms. Foreign governments and banks that borrowed during the years of negative real interest rates could not earn enough to pay the borrowing cost or retire the debt. Beginning in 1982, borrowers began to default, causing large potential losses at major banks at home and abroad. IMF and Federal Reserve policy protected the banks at the expense of growth in the debtor countries and to a lesser extent at home. Second, domestic banks also faced large losses on loans to real estate and energy companies. Adjustment to reduced oil prices, reduced home building, and historically high real rates on mortgages left many banks facing loan losses. Third, the dollar appreciated in the early 1980s, stimulating imports and reducing exports. Reducing the dollar exchange rate later became a goal of the Treasury, assisted by the Federal Reserve. This involved agreement with other leading countries, especially West Germany and Japan. This problem occupied the Federal Reserve in the 1980s. Fourth, the FOMC wanted to prevent a return of high inflation. It was uncertain about how to do that. Policy was judgmental, based mainly on Volcker’s judgments.
It is not possible to date precisely the end of the inflation. By 1986, long-term interest rates, exchange rates, and changes in wages and other prices no longer incorporated high expected inflation. This evidence of the adjustment of expectations is a good measure of the end of inflation. As Governor Wallich noted above, money growth and real interest rates in 1983 gave very different signals about the future. Base growth had increased from a 5.3 annual rate at the end of 1981 to 8.3 at the end of 1982. By the spring and summer of 1983 twelve-month average base growth was above 10 percent. But yields on ten-year Treasury notes reached 10.6 percent as he spoke and rose further as the year progressed. The SPF expected inflation for the next four quarters fell to 5 percent in first quarter 1983, so the anticipated real yield was about 5.5 percent. It rose with the rise in nominal yields and a further decline in anticipated inflation.
A History of the Federal Reserve, Volume 2 Page 64
