A History of the Federal Reserve, Volume 2

by Allan H. Meltzer

  On July 27, Connally presented his recommendation to the president calling for a wage and price freeze, closing the gold window, a tax on imports, and tax reduction. Nixon wrote in his memoirs: “Even I was not prepared for the actions he proposed” (quoted in Matusow, 1998, 113). The president accepted the full program but decided to wait for Congress to return on September 7 before closing the gold window and possibly postpone the price and wage freeze until January 1972, as Burns had proposed.

  Events took control. Belgium and the Netherlands asked that their swap lines be paid and closed. This was a major step away from the 1968 agreement to avoid a gold drain. The trade balance worsened. The OECD and BIS members other than France “agreed that a wider band would be better and they had also agreed that occasional floats for brief periods to change rates would be all right” (Maisel diary, July 15, 1971, 69).88 The discussion suggested that adjustment had more support and opened the possibility of agreement at the September IMF meeting.89

  McCracken did not know about the president’s decision, but he recognized that some change had to be made. On August 9, he described the deteriorating trade position, growing overseas investment by domestic corporations, and increased reliance on foreign plants to service the domestic market. After rejecting proposals to insist either upon fewer restrictions against U.S. exports in Europe and Japan or deflation, McCracken joined those who favored either a “border tax” on imports and a credit for exports, devaluation against other currencies, or a floating dollar (McCracken to the president, Nixon papers, Box 42, August 9, 1971).

  The Joint Economic Committee made a major contribution to ending gold convertibility. Before leaving for summer recess, the committee, led by Congressman Henry Reuss, issued a report that: (1) declared that the dollar was overvalued; (2) declared that the IMF should recommend exchange rate changes for countries in fundamental disequilibrium; (3) cited the United States as a country in fundamental disequilibrium; and (4) urged the U.S. governor of the IMF (Connally) to ensure that the IMF take on this responsibility (James, 1996, 217–18).

  87. Maisel described the Treasury as opposed to devaluation but that he did not agree with the official position. He explained the Treasury’s position by saying that “you might be worse off after the move” (Maisel diary, June 8, 1971, 50). Volcker, as noted, held a different view.

  88. Burns wrote a letter to the president at about this time saying that the Japanese ambassador told him that Japan would agree to revalue as part of a multilateral adjustment but not unilaterally (Burns to the president, Burns papers, August 5, 1971).

  89. Maisel (diary, July 15, 1971, 69) notes that Burns and Connally did not get along and did not meet regularly. They had little direct communication.

  If markets needed to be convinced that the dollar would be devalued, the Joint Economic Committee report provided that evidence. Flows from the dollar to foreign currencies rose during the week of August 9. On August 11, OMB director George Shultz met with the president. The president explained Connally’s program. Shultz was against price and wage controls, but the president had decided. The president said that only the three of them would know what he planned to do ( Shultz, 2003). Shultz, a skilled negotiator, advised the president that he would need to impose the temporary 10 percent tax on imports that Connally recommended to supplement closing the gold window. That would force negotiations on exchange rates and other issues.

  Table 5.4 suggests the size of the flows into foreign central bank reserves in 1970–71. West Germany held the largest share of West European reserves, and Japan held the largest share of Asian reserves. The market expected devaluation of the dollar to come principally through revaluation of the mark and the yen.

  The remaining issue was how to make the announcement. The president again considered doing it in steps, but Connally urged a one-time announcement. If they just closed the gold window, it would look as if they were forced to do it and had to take time to decide what else to do. The president liked the idea of a bold program; it reminded him of his decision to go to China. They would go to Camp David to work out the details. “Connally suggested that at Camp David all should be encouraged to participate in the discussion, without their letting on that the decisions had already been made” (recording of conversation among President Nixon, Secretary Connally, and Budget Director Shultz, Nixon Executive Office Building Tapes, conversation 273-20, August 12, 1971, 5:30–7 p.m.).

  The capital outflow forced action in August that would have been delayed until after September 7. “We were on the brink of a market panic that willy-nilly would force us off gold. If we were going to take the initiative of suspending the convertibility of dollars for gold and present it as the first step of a considered and constructive reform package, the decision would not wait” (Volcker and Gyohten, 1992, 76).90 Volcker notified Connally, who talked to the president. The decision to go to Camp David that afternoon came quickly.

  CONCLUSION: WHY FIXED EXCHANGE RATES ENDED

  The Bretton Woods system of fixed but adjustable exchange rates broke down because no major country or group of countries was willing to subvert domestic policy to improve international policy.91 Exchange rate stability was a public good; no country was willing to pay much to supply it. The United States chose to maintain high employment even if its policy required rising inflation, as it did after 1965. When problems arose, it used capital controls to hide the problem temporarily. The Johnson administration developed many clever stopgaps, but it would not adopt a long-term solution. And it mistakenly kept the target unemployment rate at 4 percent. They would not accept that meeting their domestic concerns was incompatible with their international standard or that the 4 percent unemployment rate was too low for stable inflation.

  Policymakers recited a standard mantra that recognized three problems: liquidity, adjustment, and confidence. Despite the rising stock of dollar reserves abroad, they gave greatest effort to creating more liquidity, and almost none to reaching agreement on an adjustment mechanism that could have sustained the fixed exchange rate system. The United States predicated its policy on the belief that other countries would match any devaluation of the dollar. This seems highly plausible for some countries, but implausible for the principal surplus countries, especially Germany. So they solved a problem that was unimportant by agreeing on special drawing rights or SDRs and ignored the more critical adjustment problem.

  90. Some versions of these events, including Haldeman’s diary, cite a British demand for $3 billion in gold as the triggering event. Solomon (1982, 185) corrects the amount, which was actually $750 million; also, the request was for an exchange rate guarantee, not for gold (Volcker and Gyohten, 1992, 77). Volcker explained that the British request came after the decision to go to Camp David had been made. He called Charles Coombs to permit him to make one last plea for retaining the $35 gold price. Coombs took a call from the New York bank saying that the British had asked for $3 billion in gold. This message was garbled. “The momentum toward the decision was by that time, in my judgment, unstoppable” (ibid., 77). When I asked George Shultz about the amount, he said, “The size of the demand was not the point because there would obviously be a run” (Shultz, 2003).

  91. The papers in Bordo and Eichengreen (1993) suggest inter alia the following reasons: (1) change in relative productivity growth (225–26); (2) higher saving rates abroad (261); (3) the permanence of systemic shocks (266); (4) increasingly efficient capital markets (509); (5) a desire by some Europeans to discipline the United States by preventing it from devaluing to pay off its liabilities; and (6) subordination of international to domestic policy (617) The last of these (Feldstein, 1993) seems to me the most fundamental in that the other problems would have been more easily resolved if countries had been willing to accept temporary changes in the level of employment caused by exchange rate adjustment or devaluation of the dollar. More generally Obstfeld and Rogoff (1995) show that most fixed exchange rate systems lack durability.

  The Uni
ted States had the greatest responsibility because the system had become a dollar standard and most of the main problems came from an excess supply of dollars. This was particularly true after 1968 when gold sales effectively ended. But the surplus countries, West Germany, Japan, and for a time France and Italy, were also unwilling to take actions that would preserve the system. These countries would not willingly revalue their currencies enough to reduce their trade surpluses, or reduce restrictions on agricultural and other imports from the United States, or inflate. Although greater exchange rate flexibility became more acceptable by 1971, countries were far from agreement on the particular form it would take.

  Time ran out before agreement could be reached and before there was reason to believe that a new agreement was possible. Working against agreement was the loss of confidence in the ability (or willingness) of the United States to redeem its liabilities. An odd result was that after forcing an end to the fixed rate system, countries and their citizens accumulated vastly more dollars after the dollar became legally inconvertible. The reason again was full employment policy.

  Loss of confidence reflected more than the growing stock of dollar liabilities and the reduced U.S. gold stock. Under fixed exchange rates, unless countries follow the rules or respect the public benefit, exchange rate stability is lost. A common lesson of Bretton Woods, the interwar gold standard, the Tripartite Agreement, and other efforts at cooperative solutions to fix exchange rates or bands is that enforcement of the rules is weak at best. Countries act in their perceived self-interest; typically they overweight short-term costs and underweight any long-term benefits.

  In the early 1960s, when the United States’ gold stock was large relative to its external liabilities, it was not difficult to believe that the U.S.’s problems were temporary. The strong expansion increased productivity growth and kept inflation low until 1965. President Kennedy’s concern strengthened this belief. By the late 1960s, it became increasingly clear that the trade balance had declined, the flow of dollars had increased, the Johnson administration would not slow the economy to reduce inflation, and the Federal Reserve had abnegated much of its independence by accepting policy coordination. The new administration after 1969 was unlikely to shift policy from domestic to international concerns. And the Federal Reserve would not even raise its discount rates during a period of exchange market turmoil. Many in the administration and some at the Federal Reserve preferred to end the fixed exchange rate system. As markets learned about willingness to end the exchange rate system, confidence that it would survive withered.

  One contentious issue about the breakdown is the role of money growth. Bordo (1993) and Darby, Lothian, et al. (1983) concluded that excessive money growth in the United States was the principal source of domestic inflation and its transmission by capital flows to the rest of the world. Bordo (1993, table 1.31) shows that the loss of gold did not deter the Federal Reserve from expanding domestic credit (see also appendix to chapter 4). Cooper (1993, 106) challenged that conclusion, claiming that the difference between growth of money and output in 1970 was no greater than in several previous recessions.

  I believe Cooper misses the point for several reasons. First, productivity growth declined in the late 1960s, but money growth did not. The persistent excess increased. And it was the persistent excess, not the temporary change, that the Europeans and Japanese had to absorb persistently. Second, the system was much weaker and the U.S. gold stock much smaller in 1970–71 than in the earlier years that Cooper cited—1954, 1958, and 1967. And the U.S. current account surplus had become a sizeable deficit. It was reasonable to believe through 1964 that the increased trade balance and real exchange rate appreciation would end the problem. By 1970–71, this belief had withered. The steady decline in the current account surplus after 1964 and the rise in U.S. unit labor costs suggested that the current account deficit would continue.92 Third, and most important, by 1970–71 the rest of the world had ample reason to believe that exchange rate stability ranked far below higher employment as an administration objective, so inflationary policies would continue. Fourth, the U.S. budget deficit increased on average. Federal Reserve policy would not raise interest rates enough to slow monetary expansion. A prudent observer would expect capital outflow to continue.

  There were few voices in the Nixon administration that would fight for the fixed exchange rate system. Only Burns openly opposed a suspension of gold payments. At the Federal Reserve Arthur Burns, Alfred Hayes, and Charles Coombs wanted to maintain convertibility and defend the exchange rate, but they offered no effective means of doing so while expanding the economy. The latter was, of course, critical for President Nixon because reelection was his main objective.

  The design of the Bretton Woods system had fundamental flaws. It reflected the prewar experience, especially the British problem of reconciling employment and exchange rate stability in the 1920s. Countries could borrow from the International Monetary Fund to adjust to a temporary problem; they could devalue if they faced a fundamental disequilibrium. The agreement gave insufficient attention to distinguishing temporary from permanent or fundamental disequilibrium. Also, the designers wanted to make adjustment symmetric for deficit and surplus countries, but they neglected the political pressures against revaluation in surplus countries. Surplus countries viewed revaluation as a penalty for success instead of as a reward for better policies. With the passage of time, the system became more rigid. Britain delayed devaluation several years before 1967. France delayed devaluation in 1968. Japan never revalued; it supported the yen by buying dollars for a few weeks even after August 15, 1971.

  92. Eichengreen (1996, 129) also cites the importance of the persistent or permanent current account deficit in the system’s breakdown.

  By 1970–71, it must have been clear to outside observers that the United States did not intend to follow policies consistent with a fixed exchange rate system. Abroad, the German and Japanese current account surpluses seemed persistent. Sooner or later, these currencies were likely to appreciate. The United States trade account declined steadily after 1965. By 1971, the account was in deficit. The run from the dollar began. Germany added $3.1 billion to its reserves in the first six months of 1970. German money growth rose from 6.4 percent in 1970 to 12 percent in 1971. German consumer prices rose 1.8 percent in 1969 but 5.3 percent in 1971. Despite rigid exchange controls, Japan could not escape U.S. monetary expansion. Japan’s reserves nearly tripled in the first nine months of 1971, rising $8.6 billion. The Japanese (M1) money stock rose 25 percent in 1971.

  The U.S. capital outflow dwarfed previous (but not later) experience. For the full year 1971, the deficit on capital account was almost $30 billion, after $12 billion in 1970. Of this amount $40 billion became dollar reserves of other countries. Japan and Germany accumulated $11 billion each, and Britain acquired $10 billion.

  The 1960s witnessed renewed efforts to solve international monetary or economic problems by coordinating policy actions. At a fundamental level, the problem was to find a way to maintain a public good. In practice, this usually meant that surplus and deficit countries were supposed to agree on different mixes of monetary and fiscal policy actions. Discussions produced few concrete steps. Foreign countries accepted the expansions implied by the flows of U.S. dollars, but they did not systematically reduce government spending or raise taxes to slow their expansions and lower domestic interest rates. And, as discussed earlier, the United States focused mainly on domestic objectives. The dialogue about coordination, once started, was hard to stop. It continued into the 1970s and 1980s. Countries that faced a balance of payments deficit usually favored coordinated action. Countries in surplus usually opposed.

  By 1971, Germany accepted that appreciation of the mark was the best way to prevent domestic inflation. The German government tried to get agreement from other European countries for a joint float against the dollar, but France and Italy opposed. They disliked inflation but were unwilling to accept the cost of adjustment.
Once again coordination failed. Germany especially had less aversion to exchange rate adjustment than to continued inflation.

  One of the lessons of 1965–71, known as the policy trilemma, is that independent monetary policy, fixed exchange rates, and capital mobility are incompatible. As long as low unemployment was the principal policy objective on both sides of the Atlantic, the choices for the Europeans were inflation, greater exchange rate flexibility, or rigid capital controls. The experience of 1965–71 taught many about the choices they had to make; President Nixon’s decisions at Camp David forced them to take the first steps to greater exchange rate flexibility.

  SIX

  Under Controls: Camp David and Beyond

  The frequently heard argument that “needed” fiscal and monetary stimulation will be possible if there is an “adequate incomes policy” is proof enough of the most pernicious aspect of controls. It suggests why controls or any reasonably formal incomes policy are likely to lead to more inflation and not to less.

  —Shultz, 1975, 4

  On August 15, 1971, President Nixon announced the New Economic Policy to the nation and the world. Consumer price inflation was at a 3.6 percent average rate for the eight months of 1971 before controls began. In the eight months after controls ended in April 1974, the average inflation rate was 12.2 percent. Inflation rose, as Schultze suggested it would. Of course, the reason inflation rose may be unrelated to the control program.1 Part of the increase resulted from the oil price increase. This raised the price level. Since the increase occurred over time, the rate of price change increased also. But that increase was temporary, unlike the price level change. Once oil prices reached a new level, the rate of price change declined.