Maurits C. Boas Professor of International Economics, Harvard University, and Assistant Vice President and Economist, Federal Reserve Bank of Boston, respectively. This article is part of a paper presented at the forty-fifth economic conference of the Federal Reserve Bank of Boston, "The Evolution of Monetary Policy and the Federal Reserve System Over the Past Thirty Years: A Conference in Honor of Frank E. Morris," held October 11, 2000. The authors' acknowledgments appear at the end of the article.
U.S. monetary policy has a purely domestic mandate. The Federal Reserve's task is to promote "maximum employment, price stability and moderate, long-term interest rates" within the United States. (1) Or, as Arthur Burns put it in 1973, "American monetary policy is not made in Paris; it is made in Washington." (2) That said, this article will argue that global developments have played a significant role in setting the focus and practice of U.S. monetary policy in the years since Frank Morris became President of the Federal Reserve Bank of Boston. When Frank Morris joined the Fed in 1968, the Bretton Woods system--based as it was on the dollar's unsustainable link to gold--was on the verge of collapse. Even so, the U.S. dollar remained the only viable international transactions currency at that time, and the financial "world" encompassed a mere handful of nations edging the North Atlantic, plus, grudgingly, Japan. Today, of course, the major currencies are floating, the euro is increasingly used as a transactio ns currency, and investor horizons have widened to include emerging markets on every continent.
Within this changed setting, the U.S. economy has itself become considerably more open to international trade and investment flows. Thus, promoting U.S. price stability and maximum sustainable growth has increasingly required taking global developments into account. Usually, these developments have been taken as "givens," inputs to the data set on which policy decisions are based. From time to time, however, international developments--such as major exchange rate shifts--have elicited a Fed policy response aimed at influencing the course of these "external" events. The intent, of course, has always been an improved long-term outcome for the U.S. economy.
Beyond changing the setting in which U.S. policy decisions are made and the considerations on which they are based, international forces have also influenced the evolution of the U.S. financial system and, thus, the practice of U.S. monetary policy. Over the past forty years, foreign opportunities and foreign competition have helped drive financial innovation and regulatory change in this country. These developments, among others, eventually forced the Fed to de-emphasize monetary aggregates and to adopt the federal funds rate as its operating target instead. (3) These same forces also contributed to the demise of the Glass-Steagall (interindustry) and McFadden (interstate) restrictions on bank activities. (4) Shifts in central bank practice overseas may also have encouraged similar changes in this country.
This article examines the impact of global developments on the practice of U.S. monetary policy, broadly defined to include regulatory and lender-of-last-resort functions as well as open market, discount, and intervention activity, over the past forty years. The first section briefly reviews a few familiar facts establishing the increased openness of the U.S. economy. The second section explores episodes when external events beyond those included in the domestic outlook -- events like significant exchange rate shifts--appear to have influenced policy decisions. The authors would like to emphasize that the analysis relies in large part on an admittedly subjective examination of the Records of Policy Actions of the Federal Open Market Committee (FOMC) and of the Board of Governors found in the Board's annual reports. The section explores what sorts of events triggered a policy response. Have the frequency or causes of these episodes changed with the advent of floating rates and increased capital market integrat ion? Do the patterns suggest any lessons? Section three peers into the future, applying the lessons of the past, and section four provides a summary and conclusions.
I. The increased Openness of the U.S. Economy
Although the world is arguably no more open today than it was at the start of the twentieth century, the U.S. economy is certainly more open than it was at the beginning of the 1960s. As real exports plus imports have grown from less than 10 percent of GDP in the early 1960s to almost 30 percent currently (Figure 1), (5) net exports have come to exert a notable impact on GDP growth on a more frequent basis. In the early years (1960 to 1974), net exports added or subtracted 1 percentage point or more from GDP growth rather infrequently, as Figure 2 shows. But from 1975 to 1984 and again in the late 1990s, net exports affected GDP growth by 1 percentage point or more over 40 percent of the time. The periods of obvious impact coincided with the oil shocks, the huge dollar appreciation and LDC debt crisis of the 1980s, and the more recent Mexican and Asian financial crises and their economic consequences. The pattern is also highly countercyclical, naturally enough, with net exports having a positive impact on U. S. growth just before and during a U.S. recession and a negative impact during a U.S. recovery. Confirming the old saw about the contagious effects of U.S. colds leading to pneumonia overseas, this pattern suggests that the United States has frequently been the instigator of world downturns as well as the engine of world growth. But in recent years, the impact of the Asian crisis on the giant U.S. economy through the trade channel is clearly visible.
Less familiar is the fact that U.S. international trade in securities has grown even faster and now looms considerably larger, relative to GDP, than trade in goods and services (Figure 3). While nominal exports plus imports equaled 23 percent of GDP in 1999, gross U.S. international transactions in securities equaled 200 percent of GDP. These data suggest that private capital flows rather than trade flows have been driving the large exchange rate swings of recent years. Reflecting the promise of this country's "new economy," and concerns about needed reforms in Europe and Japan as well as in many emerging markets, these capital inflows surged notably in the late 1990s and now make up a significant share of the funds raised in the U.S. credit markets. Since 1995, foreigners have provided, on average, 35 percent of the total credit raised by the U.S. nonfinancial sector (Figure 4), up from an average 2 percent in the early 1960s. In addition, in the past three years foreigners have acquired about 10 percent of the increase in U.S. corporate equities outstanding. The stimulus provided by these capital inflows plus the wealth effect of the related increase in U.S. asset prices appears--rather unexpectedly--to have more than offset the negative impact of deteriorating net exports on the U.S. economy. (6)
As the U.S. economy has become more open, access to data on foreign economies has improved, and various new markets have emerged, the preparatory materials for FOMC meetings have come to cover a growing number of countries in greater detail. Still, the minutes frequently label international trends "a key area of uncertainty," and the cumulative spillover effects of synchronous downturns and recoveries remain hard to foresee. Over the forty years covered by this paper (1960 to 2000), FOMC interest has naturally turned from the deficits and gold outflows that drew its attention during the late 1960s to the likely impact of net exports on U.S. demand conditions and of the dollar's exchange rate shifts on price developments in this country. In recent years, for instance, some Committee members have attributed the surprisingly good behavior of U.S. inflation in part to the increased competition and the reduced capacity constraints facing U.S. producers, thanks to the strong dollar, slack conditions in our trading partners, and increased outsourcing from overseas. By comparison, the impact of international capital flows on U.S. asset prices and on investment and consumption activity in this country has received limited attention.
II. Beyond the Domestic Outlook: How Have International Events Influenced Fed Policy?
As described above, in pursuing their domestic goals of price stability and maximum sustainable growth, the members of the FOMC always consider the likely impact of net exports and the foreign exchange value of the dollar on U.S. demand conditions and inflation. But beyond that considerable influence, to what extent have Fed policymakers based policy decisions on "international" considerations--to affect the foreign exchange value of the dollar, for instance, or in reaction to external financial crises? The conventional answer is almost never. This section reexamines the issue and comes to a somewhat different conclusion.
To look for evidence of international influence, we examined the Records of Policy Actions for the Board of Governors and for the FOMC for the past forty years. (7) This exercise required considerable interpretation, of course, and other readers might come to some-what different conclusions in specific cases. However, the records usually provide a rationale for the decisions made and give some guidance on the weights assigned to the various, often conflicting, arguments put forth in determining the ultimate outcome. (8) For example, in February 1978, a majority of the FOMC members agreed that the weak dollar "militated against" easing while domestic sluggishness precluded tightening. Thus, the compromise outcome, clearly influenced in part by international considerations, was a decision to maintain existing monetary conditions. On occasion, moreover, and usually after some debate, the members of the FOMC have changed the wording of the highly formulaic directives to the New York Fed in order to highlight part icular concerns. For example, in May 1973, the directive instructed that FOMC policy be implemented "taking account of international and domestic financial market developments," reversing the usual order. A year later, the usual order ("domestic" first) was restored.
The results of this effort are displayed in Figure 5, which shows the number of times each year that international considerations other than those incorporated in the domestic outlook seemingly affected discount rate decisions or the domestic policy directive for open market operations. (9) The red bars indicate when international events were the primary reason for action (or lack thereof), while the grey bars show when international considerations helped tip the balance. The figure does not have a bar representing changes in Regulations D and M, (10) which the Fed made frequently in the late 1960s through the mid 1970s in an attempt to limit capital outflows or encourage capital inflows; such efforts had fallen out of favor by the late 1970s. A possible final instrument, U.S. intervention in the foreign exchange markets, is discussed separately below.
Clearly, as Figure 5 indicates, the episodes when the Board or the FOMC made policy decisions primarily for international reasons are rare. (11) However, possibly because Fed policymakers must often make decisions in the face of considerable uncertainty and on the basis of conflicting indicators, international developments appear to have influenced policy at least marginally a good deal more frequently than is generally recognized--even in the period of floating exchange rates.
As the bar graph suggests, international issues were generally more influential in periods of dollar weakness (for example, 1978-79 and 1985-87) or at times when currency or debt crises in emerging markets threatened the liquidity (or solvency) of U.S. financial institutions (for example, the mid 1980s and 1998). By contrast, it was relatively easy for central bankers with a mandate for maintaining price stability to ignore dollar appreciations that supported their fight against inflation. (Figure 6 shows the real U.S. trade-weighted foreign exchange value of the dollar in terms of the major currencies from 1973 on.) The most recent decade appears to be an anomaly, however. Other than the pronounced impact of the Asian crisis, international concerns were fairly muted in the early to mid 1990s, even during 1994 and 1995 when the dollar was approaching its post-Bretton Woods lows. Does the recent experience suggest that the United States has learned to live comfortably with a floating dollar? The following subs ections will review in more detail the periods when exchange rate shifts and international financial crises have had a perceptible impact on policy.
Exchange Rate Pressures
According to Figure 5, concerns about dollar outflows and the viability of the Bretton Woods arrangements were barely perceptible in 1960 but had become a "malignant preoccupation" (Eichengreen 2000) by 1967-68. These anxieties then diminished as the United States headed into the 1970 recession that prompted a temporary and misleading improvement in the U.S. current account. (12) It should also be noted, however, that Fed policymakers used a variety of tools not covered in Figure 5 as they tried to shore up the crisis-prone fixed exchange rate system. (13) The Fed's search for extra tools reflected its penchant throughout the 1960s and 1970s for pursuing several, often incompatible, objectives simultaneously. In addition to limiting the buildup of dollar assets in foreign official hands, these goals included easing Treasury financing operations, limiting financial disintermediation, (14) promoting the growth of bank credit, especially mortgage loans, discouraging inflation, and, clearly the top priority, enco uraging a more complete utilization of the nation's resources.
Among the supplementary tools used for international ends were capital controls, Operation Twist, and intervention in the foreign exchange markets. Begun in 1961, Operation Twist entailed concentrating open market purchases in long-term securities "when feasible." (15) In theory, international capital flows were more responsive to short-term than to long-term interest rates, and this effort was intended to keep short-term rates higher than they would otherwise have been. In addition, starting in 1965 the Fed was responsible for ensuring that banks complied with the government's restraints on foreign investment. When this voluntary credit restraint program promoted the development of the Eurodollar market, the Fed began to use its Regulations D (reserve requirements) and M (foreign bank operations) as capital controls, as discussed further below.
In the end, however, these multifaceted but limited efforts to save fixed exchange rates failed. As the United States pulled out of the 1970 recession, its current account resumed deteriorating, and, flooded with huge, unwelcome private capital flows from the United States, the German and Dutch authorities allowed their currencies to float. (16) The Swiss and the Austrians revalued. By August 1971, U.S. liabilities to foreign officials stood at $41.5 billion, almost double their value at the end of 1970, while U.S. reserve assets (mainly gold) had fallen to $12.1 billion. With no acceptable alternative, President Nixon ended the dollar's remaining links to gold. In the end, neither the United States nor the Europeans were willing to sacrifice their domestic economic goals for a chronically endangered system that both the deficit and the surplus countries had come to view as flawed.
International pressures reappeared in the Fed's deliberations when the currency revaluations negotiated at the Smithsonian Institution in December 1971 proved inadequate to correct ongoing U.S. payments deficits, and massive capital flows from the United States resumed. This renewed inflationary onslaught led the Europeans to initiate a joint float against the dollar in early 1973--a fateful step that ended the early Bretton Woods era and started the Europeans on the path to monetary union. (17)
Thereafter, the first oil shock triggered an inflow of petrodollars, a modest dollar appreciation, and a deep U.S. recession. These developments reduced the impact of currency concerns on Fed decision-making. (18) However, when the foreign exchange value of the dollar plunged to consecutive new lows in the late 1970s, the press of international issues reached new highs. Indeed, the 1-percentage-point rise in the discount rate that was announced on November 1, 1978, as part of a joint Treasury-Federal Reserve package to strengthen the dollar was the largest increase in forty-five years. The package included higher reserve requirements, the threat of forceful, coordinated foreign exchange intervention, plus an increase in the fed funds rate to accommodate the new discount rate. Under Bretton Woods, episodes of dollar weakness had merely threatened foreign complaints and, eventually, the collapse of an unworkable system. Under the new floating regime, a falling dollar meant "continuing domestic inflationary pres sures," which called for strong countermeasures.
By sharp contrast, despite a severe, double-dip recession, the strong dollar of the early 1980s elicited little concern. Beryl Sprinkel, who did not believe in foreign exchange market intervention, was Under Secretary of the Treasury, and dollar appreciation was helping the Volcker Fed to meet its overriding goal of subduing inflation. (19) Despite complaints from the nation's manufacturers, the Reagan Administration hailed the strong dollar as a vote of confidence in the vibrant U.S. economy. By late 1984, however, FOMC members began to see the dollar's unprecedented strength as a serious problem affecting manufacturers and farmers and their creditors and distorting investment decisions. Thus, limiting the dollar's rise became a policy goal. With this policy shift, the announcement of the Plaza Accord, and modest U.S. dollar sales, (20) the dollar quickly reversed its five-year appreciation in a matter of months--at which point limiting its plunge became a major concern, as emphasized by the G-10 announcemen t from the Louvre in February 1987. Renewed dollar strength in 1989 led to substantial dollar sales in the foreign exchange market. These dollar sales provoked expressions of concern among FOMC members, but, once again, the dollar's appreciation did not.
A further episode of notable dollar weakness occurred in 1994-95 when the dollar again approached its post-Bretton Woods lows on a trade-weighted basis and historic lows against the German mark and the Japanese yen. While this episode also led to modest intervention in the foreign exchange markets, Fed policymakers evinced limited concern about the dollar's behavior, in contrast to the more pronounced anxiety shown during previous periods of dollar depreciation. What had changed? According to the FOMC minutes, what had changed was the Fed's attitude toward foreign exchange intervention.
The legal authority and responsibility for U.S. intervention policy have never been clearly delineated. Nor have FOMC members always been comfortable with their role in this area. Both the Treasury and the Federal Reserve appear to have congressional authorization to intervene. The Federal Reserve Act allows Reserve banks to deal in specified assets, including "cable transfers," a phrase that referred to foreign exchange in 1913. In addition, section 10 of the Gold Reserve Act of 1934 gives the Secretary of the Treasury, with the approval of the President, the power to buy and sell foreign currencies "for the purpose of stabilizing the exchange value of the dollar." The Act also created a $2 billion Exchange Stabilization Fund (ESF) to permit the Secretary of the Treasury to carry out the provisions of the section. (21.)
By tradition, possibly because congressional intention seems a bit clearer in the case of the Treasury, that department has usually taken the lead in foreign exchange intervention policy. But the New York Fed actually conducts all intervention operations. While the Treasury can order the New York Fed to intervene on behalf of the Treasury, it cannot require the Fed to intervene for its own account. However, the two agencies almost always intervene jointly, since working at cross-purposes in foreign exchange matters would clearly be counterproductive. In addition, in the past and with the knowledge of the Congress, the Treasury Department has periodically increased the ability of the ESF to intervene by requesting that the Fed engage in swap transactions. Under these arrangements, which must be authorized by the FOMC, the Fed buys the Treasury's foreign currency spot for dollars, with offsetting forward contracts reversing the transaction. The purpose is to expand the ESF's ability to engage in purchases of fo reign currency.
The top panel of Figure 7 displays U.S. intervention activity monthly since 1973, while the bottom panel shows U.S. official purchases and sales of dollars for foreign currencies relative to open market purchases and sales of domestic securities. As the second panel suggests, U.S. foreign exchange intervention has always occurred on a very modest scale compared with domestic open market operations. Rarely amounting to as much as 3 percent of domestic transactions, these data are certainly consistent with the Fed's statements that U.S. foreign exchange interventions are "routinely sterilized" by offsetting operations in domestic securities. Not only is U.S. intervention activity small-scale in comparison to domestic open market operations; it is also small-scale relative to foreign governments' intervention against the dollar. For instance, from October 1977 to the end of 1978, the U.S. authorities bought $10 billion to support the dollar while foreign authorities bought about $37 billion. Again, in the five w eeks after the Plaza Accord, the G-5 sold $9 billion, of which the United States sold $3.3 billion. (22) And, of course, all official intervention activity is dwarfed by the total volume of transactions in the foreign currency markets, which now exceeds $1.5 trillion each day. (23)
No wonder, then, that most studies of the efficacy of foreign exchange intervention conclude that this activity has no--or at most fleeting--impact on foreign exchange rates. (See Edison 1993; Humpage 1996.) Indeed, if foreign exchange interventions are routinely sterilized, they could only affect the exchange rate by serving as a signal that macro policy is about to change or that, by official judgment, market expectations are no longer linked to economic fundamentals (Dominguez and Frankel 1993 a, b). On the other hand, as Schwartz (2000) has pointed out in her review of the literature on foreign exchange intervention, models of exchange rate determination have never been able to explain or predict actual exchange rate behavior. Thus, it is not possible to compare the impact of intervention with the outcome prevailing in its absence. Moreover, these studies generally examine the impact of intervention holding macro policy constant. But, as a comparison of Figures 5 and 7 suggests, from 1973 to 1989 periods of heavy intervention appear to correspond with periods when international developments were exerting at least some marginal influence on Fed policy. (24) After 1989 the correspondence vanishes. But if exchange rate pressures do in fact influence policy outcomes from time to time, then the results of these intervention studies may be biased.
Why did the link between foreign exchange intervention and the influence of international issues on Fed policy appear to weaken in 1989? Or alternatively, why did Fed policymakers express only limited concern as the dollar sank to new lows in 1994--95, and the United States made the sizable interventions shown in Figure 7? The policy records suggest that this discrepancy between words and actions may reflect the fact that, starting in 1989, a significant minority of FOMC members began to oppose U.S. intervention activity. This opposition was not entirely new. When the FOMC first authorized System operations in foreign currencies in 1962, two governors dissented. (25) They argued that such a program required analysis by outside experts, public discussion, and legislative clarification. They also doubted the legality of Fed intervention (although the General Counsels of the FOMC and the Treasury and the Attorney General of the United States had all agreed that the System was authorized to conduct such operation s). Nevertheless, the majority endorsed the new program. And in the years thereafter, the FOMC repeatedly raised the ceiling on System holdings of foreign currencies for its own account and for the account of the ESF. And it repeatedly expanded its reciprocal currency (or swap) arrangements--both in terms of membership and of the dollar value of the commitments. (26) All of these cooperative arrangements provided resources for foreign exchange intervention.
Starting in 1989, however, while the United States was participating in coordinated intervention to limit the dollar's appreciation, a minority group of Fed governors (27) began to dissent from endorsing ever-greater holdings of foreign currencies. In March 1990, in particular, three FOMC members dissented from raising the ceiling for System holdings of foreign currency "warehoused" for the Treasury. They argued that recent interventions to weaken the dollar undermined the credibility of the Fed's commitment to price stability, were probably ineffective, and, in the case of the warehousing facility, were inappropriate without an indication of congressional intent. While the majority authorized the increase, many members expressed doubt that intervention could have lasting effect and suggested that future intervention should be limited. Because of these concerns, the System broke with the Treasury and did not participate in intervention sales of dollars made between March 5 and the end of 1990. (28) The follow ing year the Fed resumed intervening with the Treasury on a "moderate" scale, and any interventions since then have been joint.
As Figure 5 suggests, finally, the most recent period of significant dollar weakness engendered further doubt about intervention (29) and an unusually detached response to the dollar's fall. In mid 1994, with the dollar in a steep decline, two members of the FOMC took the unusual step of refusing to ratify System intervention activity that had already taken place. They argued that repeated failures to achieve intervention objectives would raise questions about the credibility of Fed policy more broadly. (30) Moreover, while FOMC deliberations sometimes referred to the inflationary impact of dollar depreciation, the rationale for the significant shift to tighter monetary policy that actually occurred at this time rarely cited the dollar. For example, in March 1995, the members noted that policy should not be directed to achieving a given level for the dollar but to implementing effective anti-inflation policy, taking account of all relevant factors.
Meanwhile, the fed funds rate doubled to 6 percent in fifteen months, while the discount rate went from 3 percent to 5.25 percent. In both cases, the last 50-basis-point rise took place in early 1995 as the dollar neared its low point. The domestic conditions that provoked these changes included an acceleration in real GDP growth to a peak of 4.4 percent, year-over-year, in the third quarter of 1994 (after which time GDP growth slowed) plus a decline in the unemployment rate from 6.6 percent to 5.5 percent. (31) Core inflation (consumer prices excluding food and energy) held steady near 3 percent. Given the policy changes that actually occurred, and the domestic context in which they occurred, it is hard to know whether the FOMC was quite as indifferent to the dollar's 1994-95 decline as Figure 5 suggests. Since then, of course, the dollar's strength has been an asset in the Fed's efforts to keep inflation low.
To supplement evidence of FOMC concerns about the exchange rate found in the Records of Policy Actions, the authors also looked at the impact of U.S. official intervention activity on FOMC decisions to change the fed funds rate, using logit estimations of the FOMC voting patterns and a model of the FOMC's reaction function developed by Stephen McNees (1986 and 1992) and Geoffrey Tootell (1997). In addition to the other explanatory variables used by Tootell, (32) we added U.S. official net purchases or sales of dollars (deflated by the GDP deflator) made in the weeks preceding the FOMC meeting. This variable represents a possible measure of official concern about the dollar exchange rate. In the reaction function equations, the coefficient on dollar purchases is positive, suggesting that intervention to support the dollar tends to be associated with an increase in the fed funds rate, and it is significant at the 10-percent level. Dollar sales, associated with a strong dollar, have no statistically significant impact. Otherwise, the introduction of the intervention variables has little impact on Tootell's original results.
In the logit estimations of FOMC voting decisions, dollar purchases are positively related to the probability of tightening, negatively related to the probability of loosening; both coefficients are highly significant. In other words, according to these results, the larger the official dollar purchases, the more likely the FOMC is to tighten and the less likely it is to loosen. In the case of dollar sales, which enter the equation as negative values, the coefficient is positive as expected and highly significant. That is, the larger the dollar sales, the less likely is an FOMC decision to tighten. Dollar sales do not have a statistically significant impact on the probability of a decision to loosen. The regression results and a figure showing a plot of the residuals from the baseline equation and real net intervention activity may be found in the Appendix.
All told, this review suggests that in periods of steep depreciation, the dollar has continued to exert at least marginal, although perhaps dwindling, influence on U.S. monetary policy-even in the current floating-rate era. The size of the U.S. current account deficit, now 4 percent of GDP and rising, could lead to a new test of this tentative conclusion at any time.
International Financial Crises
While less pervasive than exchange rate issues, international financial crises have also influenced Fed policy decisions from time to time. Moreover, the increased severity and scope of these crises have posed new challenges for the Fed in its roles as supervisor and occasional international lender of last resort.
During the 1960s, the notable financial crises involved the major industrialized countries, a club-like group whose governments usually responded to currency pressures in a fairly cooperative fashion--although the degree of cooperation clearly dwindled as the crises recurred. These years saw the development of the London gold pool, for instance, and of the reciprocal currency arrangements (or swap network) that allowed member central banks to draw foreign currencies as needed to bolster their foreign exchange reserves. (33) The FOMC generally expanded and authorized the use of these facilities without much debate, upon request by foreign governments and the U.S. Treasury. And more than once the FOMC actually cited currency weakness in another country as a reason for delaying a move to tighter policy. (34) While the citizens of Britain or France must surely have felt the impact of the devaluations and more restrictive macro policies triggered by crises involving the pound and the franc, central banks and treas uries generally contained these currency upheavals without anything resembling a serious economic downturn, and the U.S. banking system remained largely unaffected. (35)
That was not true in the aftermath of the LDC debt crisis of the 1980s, however. The oil shocks of 1974 and 1979-80 had led to considerable concern about the financing needs of the oil-importing developing countries. Thus, as OPEC members deposited the bulk of their surging oil revenues in U.S. banks or in the Eurodollar market, U.S. lenders took the lead in recycling the petrodollars to the developing countries. U.S. bank claims on all foreigners rose from less than 5 percent of total U.S. commercial bank assets in early 1974, at the end of the voluntary credit restraint program, to 20 percent of total assets at their peak in early 1983. As Figure 8 shows, this growth was particularly rapid between 1980 and 1982. While Fed Chairman Arthur Burns, Fed Governor Henry Wallich, and others periodically warned the banks against allowing excessive concentrations of LDC debt to build up, the banks faced significant market incentives (36) and official encouragement to make these loans. According to William Seidman, ec onomic counselor to President Ford, "the entire Ford Administration, including me, told the large banks that the process of recycling petrodollars to the less developed countries was beneficial, and perhaps a patriotic duty." (37) Moreover, contemporary studies generally concluded that this largely sovereign debt was relatively safe (Beek in Curry 1997; Terrell 1973).
Those studies notwithstanding, Mexico suspended debt service on an $80 billion foreign debt in August 1982; by late 1983, twenty-seven nations owing $239 billion had rescheduled. Soon after, references to financial strains and the need for monetary conditions to accommodate them began to crop up in the FOMC minutes. For example, in October 1982 the FOMC voted to reduce pressures in the private capital markets, pressures that reflected the "well-publicized problems of a few U.S. and foreign banks and the acute financial difficulties in Mexico and other developing countries." (38) These problems had caused a flight to quality, increased interest rate spreads, and a heightened demand for liquidity. (39) References to strained financial conditions and the impact that higher interest rates would have on LDC debtors and their U.S. creditors continued through 1983-84.
To be sure, the U.S. oil, agricultural, and real estate sectors also suffered serious financial problems in these years, but the LDC debt crisis clearly added significantly to the "financial fragility" of the decade. As Figure 9 shows, the nonperforming share of "all other" loans (the category that includes the LDC debt) and the charge-off rate on these loans grew rapidly in the early 1980s--even relative to the worsening experience with other types of assets. Moreover, while total real estate assets greatly exceeded loans to LDC borrowers, the impact of the real estate crisis was comparatively diffuse. By contrast, LDC assets were highly concentrated at the nation's eight largest money-center banks, where, according to the FDIC data, LDC debt amounted to 250 percent of their capital base. Thus, as was widely recognized at the time, several of these banks were technically insolvent. (See also Cline (1995) for a review of these developments.)
In the event, U.S. authorities did not immediately require the banks to set aside large reserves on their restructured and nonperforming LDC loans. The delay, coupled with public funding for the LDC debtors, allowed the banks to raise additional capital before they began, starting with Citicorp in 1987, to recognize the bulk of their losses on their LDC debt. By late 1989 the money center banks had posted reserves for almost 50 percent of their outstanding LDC loans. Finally, in 1989, the Brady Plan shifted the focus from debt rescheduling to debt relief. With the help of funds from the IMF and World Bank, the debtor nations used various instruments like debt--equity swaps to reduce their debt to manageable proportions. Private sector debt forgiveness amounted to about one-third of the face value of the $191 billion in outstanding loans, potential losses that could accrue to the lending banks' shareholders. (40)
In the wake of the first LDC debt crisis, U.S. banks remained relatively cautious about lending to the emerging markets; thus, most U.S. banks were not severely affected by the Mexican peso crisis of 1994-95 or by the early stages of the Asian crisis. Indeed, it was not until the "Asian" crisis had spilled over from Southeast Asia to North Asia to Russia and was starting to threaten Latin America and, finally, Long-Term Capital Management and other highly leveraged institutions and their creditors that U.S. financial markets began to seize up. Referring to the turmoil in U.S. and global money markets, in the fall of 1998 the FOMC lowered the fed funds rate by 75 basis points in three steps; the Board cut the discount rate twice. According to the Board's 1999 annual report, these moves were deemed desirable "to cushion the U.S. economy from the effects of disruptions in world financial markets and to ameliorate some of the resulting strains...."
International Standards, Disclosure, and Market-Based Discipline
The shock of the first LDC debt crisis gave fresh impetus to G-10 efforts to negotiate internationally accepted standards for capital adequacy and other supervisory and regulatory issues. The United States frequently drove these efforts because, obviously, the quality of foreign financial supervision was by then closely linked to the safety and soundness of U.S. banks, and, given the jurisdictional issues involved, improving this quality required a multilateral, market-driven approach. Although the first such multilateral agreement, the Basle Concordat, (41) had been concluded in 1975, progress since then had been slow. Indeed, as William White has pointed out, all too often an international financial crisis was required to move the negotiations forward; thus, real progress on the international capital adequacy standards only began after the Mexican crisis of 1982 (White 1996).
But the motivations behind these initiatives included competitive as well as prudential concerns. While it was essential that the U.S. banks repair the damage that recent crises had inflicted on their capital positions, U.S. regulators did not want to place U.S. banks at a competitive disadvantage vis-a-vis foreign banks enjoying less stringent capital requirements. At the time, foreign banks, particularly the Japanese, (42) were rapidly gaining market share--thanks in large part, it was thought, to their less costly capital requirements. (43) As Figure 10 shows, the foreign banks' share of total bank assets in the United States about tripled between the early 1970s and the late l980s. (44) Thus, when the U.S. regulators moved on their own to require U.S. banks to hold more capital after the Mexican crisis, they met a storm of criticism from the industry, which, in turn, encouraged renewed efforts to reach an international agreement. When international progress stalled, Fed and Bank of England officials forge d ahead with a bilateral agreement, incorporating the U.K. practice of using risk-weighted capital standards. Because other nations feared that their banks might lose access to the important U.S. and U.K. financial markets, this deal led to renewed negotiations and, in 1988, eventual agreement on the Basle Capital Accord.
The Capital Accord sets a capital requirement of 8 percent of risk-weighted assets for internationally active banks. Banks must meet half of the requirement with Tier 1 capital--primarily common stock and retained earnings. The balance may be met with Tier 2 capital, which includes loan loss reserves, subordinated debt, and revaluation reserves for fixed assets and equities--a category of particular interest to the Japanese. Reflecting a compromise between the U.S., U.K., and Japanese authorities, 45 percent of unrealized capital gains on equities may be counted as Tier 2 capital. (The Japanese had long held out for 70 percent.) Market discipline and international peer pressure among supervisors have encouraged widespread compliance with the Accord. Thus, by September 1993, all G-10 banks with significant international operations were meeting its requirements.
Experience has led to a series of revisions to the Accord. Because the original risk categories were coarse and arbitrary, treating all sovereign or all corporate debt the same, they encouraged lenders to favor the riskiest borrowers in each category. (45) Accordingly, the 1999 revisions reflect an increased concern with market and operational as well as credit risk. And, in a search for better measures of credit risk, they propose weighting bank assets by the credit ratings assigned by commercial rating agencies, like Moody's. With U.S. encouragement, the revised Accord also recognizes that banks' internal rating systems could in future provide the basis for setting capital requirements in some cases. (46)
Writing in 1996, William R. White attributed much of whatever success the international community had achieved in creating globally accepted standards to U.S. and, to a lesser extent, U.K. leadership. He suggested, however, that the growing importance of the emerging countries and the increased ability of the Europeans to speak with one voice might undermine U.S. leadership and complicate the process of making further gains by international agreement. In the event, the crises of 1997-98 served to reinforce the validity of the BIS approach with its emphasis on capital requirements, supervisory review, and disclosure and market-based discipline. As those crises also demonstrated, however, market discipline can be fickle as well as harsh.
The Fed as International Lender of Last Resort
Times when market discipline becomes too harsh and markets cease to operate may call for intervention by a lender of last resort (LLR). But international financial crises may have outgrown the Fed's ability to serve that function in the international arena. As these crises have turned global and become increasingly complex, thanks in part to the participation of growing numbers of private agents, the Fed has traded its occasional role as ILLR for the role of leading international facilitator.
During the 1960s and 1970s, as already discussed, the Fed and the central banks of the other major industrial countries provided reciprocal ILLR facilities to members of the North Atlantic community plus Japan through the swap arrangements. Time and again, the FOMC authorized expansions of and drawings on these facilities--seemingly without much debate. Mexico was the only developing country to participate in these swap arrangements, which it used from time to time. For example, in 1976, after the peso devalued by 37 percent against the dollar, the FOMC's Foreign Currency Subcommittee approved a $600 million Treasury--Federal Reserve loan to the Bank of Mexico "to counter disorderly exchange-market conditions... pending the receipt of medium-term financing from the International Monetary Fund." Similarly, in August 1982 the Treasury and the Federal Reserve announced that they were participating in a multilateral package to provide the Bank of Mexico with a total of $1.85 billion in short-term financing. The f unds were to be made available in line with progress toward agreement between Mexico and the IMF on an adjustment program that would allow Mexico to qualify for drawings under the IMF's Extended Fund Facility. (47)
As these examples suggest, the swap program provided short-term LLR finance to its members in part because the IMF could not--and indeed was not designed to--serve in that capacity. The need for such finance was seemingly taken for granted. The creation of the General Arrangements to Borrow in 1962 also suggests that G-10 governments felt a need to supplement the resources available through the IMF. Like the swaps, this $6 billion facility was only available to the G-10. (48)
In contrast, by the time of the 1994 Mexican peso crisis, the atmosphere had changed. While a majority of FOMC members voted to make a $6 billion swap arrangement available to the Bank of Mexico and to increase from $5 billion to $20 billion the amount of foreign currency that the System was prepared to warehouse for the Exchange Stabilization Fund, two FOMC members dissented. The dissenters argued that developments in Mexico did not clearly threaten U.S. financial stability and that it was inappropriate for the Fed to participate in medium-term financing to facilitate debt restructuring. (49) They were also concerned about participating in what appeared to them to be a fiscal action without congressional authorization or more general public approval. The passage of the Mexican Debt Disclosure Act in April 1995 requiring the Treasury and the President to provide detailed reports on all guarantees to, and currency swaps with, the government of Mexico by the U.S. government, including the Federal Reserve, sugge sts the Congress's displeasure.
Accordingly, the U.S. response to the Asian crisis was very circumspect. This country did not participate in the multilateral package that was made available to Thailand, and it provided only backup, or second-line, support for the other Asian countries in crisis. U.S. funds were made available only on an if-needed basis. By contrast, the Fed did play an important role in facilitating the negotiations that led the major U.S. banks to roll over their Korean credits during that country's dollar liquidity crisis at the turn of 1997-98. (50) The New York Fed assumed a similar role at the time of the LTCM debacle nine months later.
In November 1998 the FOMC voted to allow the reciprocal currency arrangements with the Bank of Japan, the Bank for International Settlements, and the European governments to lapse. The only remaining swap arrangements are those with the Bank of Canada and the Bank of Mexico under the North American Framework Agreement (NAFA) set up in 1994.
For the members of the G-10 plus Mexico, the swap network represented a response to a felt need that the IMF could not fill. Its end reflects current exchange rate arrangements, the pending introduction of the euro and the creation of the European Central Bank, and extensive criticism from parts of academia and the press that domestic and international LLR facilities create moral hazard. It also reflects the fact that world politics are not yet as global as world financial crises. That said, engaging the private sector more fully in resolving financial crises is a highly desirable goal.
III. Reflections on the Future
The trends in the world economy identified in this article are likely to continue and indeed may even accelerate. In particular, the U.S. economy is now highly open to influence from abroad. Within the goods market, only the defense industries and a few agricultural products, such as sugar, are now seriously protected from international competition; and many services are increasingly subject to direct competition from abroad. Of course, some activities, such as construction, retail trade, schoolteaching, or medical care, will remain domestic, although even they are subject to influence and takeover through foreign direct investment. Similarly, financial markets are now wide open to international investors; with the securitization of mortgages even that very local form of loan is now internationally marketable.
A new element in the picture is the creation, in January 1999, of the euro, which in financial markets has replaced the German mark, French franc, Italian lira, and eight other European currencies, permitting the development of a Europe-wide capital market without currency risk. Japan has also shed its last restrictions on inflows and outflows of capital, although the poor performance of the Japanese economy and traditional conservatism of Japanese savers have postponed full exploitation of the new potentialities.
In general, floating exchange rates have served the United States well over the past two decades. An appreciating dollar dampened U.S. growth during the fiscal stimulus of 1981 to 1984, and again during the robust growth of 1995 to 2000; a depreciating dollar stimulated U.S. growth during the years 1991 to 1995, a period of recession and fiscal drag due to tax increases in 1990 and 1993. (The same cannot be said for Japan where, as McKinnon and Ohno (1997) have persuasively argued, exchange rate movements have contributed to destabilizing expectations and poor economic performance.)
A key question is whether past is prologue to the future. It is churlish to raise doubts about the future in a period in which the U.S. economy has performed so well, and to which such good performance Federal Reserve policy has been an important contributor, at least in the negative sense of not having aborted it prematurely. Nonetheless, a case can be made that exchange rate flexibility will not be so benign for the United States during the next two decades as it was during the last two. Indeed, many U.S. firms would welcome a weaker dollar now, both to improve their export competitiveness and to give them more room for domestic price increases in the face of rising wage rates and materials costs. Much action on the priceraising front would no doubt trigger a reaction from the FOMC, with the effect of strengthening the dollar but also weakening construction and other interestsensitive expenditures--as well as raising U.S. interest payments on the over $3 trillion in interest-sensitive securities and other c laims held by foreigners.
But suppose the U.S. economy slowed substantially without a significant depreciation of the dollar. Then it is likely that American businesses would begin to complain vociferously about foreign competition and would turn to the Congress for protectionist relief, as they did between 1983 and 1985. Or suppose that higher interest rates in Europe, combined with a slowdown in the United States, led to significant shifts in worldwide portfolios, away from dollars toward eurodenominated securities, so that the dollar depreciated not by a tolerable 10 percent but by a startling 25 percent or more. In addition to black headlines from financial journalists around the world, many asset holders would be thrown into confusion and would wonder when the rout would stop, since the potential for portfolio shifts would be huge and subject to shortrun herd dynamics. Extensive dollar depreciation, in turn, given the openness of the American economy, would permit a corresponding rise in prices of tradable goods and services and would put the FOMC in the dilemma of whether to raise U.S. interest rates in conditions of serious economic weakening.
The general point is that the United States is increasingly exposed to external events and to changes in portfolio preferences around the world. Asset holders (including Americans) face a much wider menu of choices than has historically been the case, as foreign securities markets improve. It is not too early to begin to reflect on how U.S. monetary policy might need to be recast to allow for these changes, with a view to mitigating their impact on the American economy.
Here are two thoughts: First, the Fed should think actively about, and begin to experiment with, undertaking open market operations in selected foreign securities, especially euro-denominated securities. Second, the Fed should examine switching the main focus of its attention for measuring price stability from the consumer price index (CPI) to the (finished goods) producer price index (PPI), encouraging the European Central Bank, the Bank of Japan, (51) and the Bank of England to do the same.
Since the PPI is composed predominantly of tradable goods, if all four major economies were to focus on the PPI, their targets would be similar, ultimately perhaps identical. (As with the CPI, allowance might be made for exceptional movements in particular prices, for example, oil products, giving rise to a "core PPI.") Over the past two decades, the U.S. PPI rose 1.5 percentage points less than the CPI, while over the past five years the difference has been about 1 percentage point. Thus, a PPI target could be numerically lower than the corresponding CPI target, and conceiving of "zero inflation" on this measure would be less problematic than it would be with the CPI.
Focusing on the PPI would provide the nominal anchor that many observers feel is necessary for monetary policy. In particular, it would satisfy the Maastricht Treaty's injunction to the European Central Bank to target "price stability." It would also achieve a degree of (long-term) coordination of monetary policy among the three major industrial regions, since the focus of monetary policy in each would be similar, if not identical.
With monetary policies focused on the PPI, the consumer price index could, if appropriate, rise by differing amounts in the major industrial economies, thus giving somewhat greater flexibility to differential movements in real wages, which would help to accommodate any low-frequency asymmetric shocks among the three regions. This would take some pressure off the need for exchange rate changes to deal with such shocks over time.
Foreigners hold nearly $6 trillion in marketable securities and other liquid assets in the United States, and, of course, Americans hold much more. A substantial portfolio shift (or, given the large U.S. current account deficit, even a substantial diminution of net capital inflows) would result in a depreciation of the dollar relative to the destination currencies, possibly a substantial depreciation. Given the extensive openness of the U.S. economy, that in turn would result in increases in the dollar prices of most goods and some services. (Ironically, U.S. antidumping laws strongly encourage foreigners in periods of dollar depreciation to raise their dollar prices after no more than 60 days.) (52) That in turn could induce the FOMC to tighten monetary conditions, perhaps stemming the outflow of interest-sensitive funds, but depressing the U.S. stock market. Thus, the impact on net capital flows is ambiguous and, especially in the short run, could stimulate further outflows. The question is how much damage to the real economy is tolerable, even if one is confident that some level of asset prices and exchange rates will lead portfolio allocations to be revised and perhaps reversed.
It would be desirable for the Fed to have options other than simply tightening monetary conditions, that is, by intervening in foreign exchange markets with a view to influencing market expectations about exchange rates, which have been shown to be extremely fragile in recent years. One strategy would be to wait until a run occurs and rely on cooperation from foreign monetary authorities to provide adequate support in the emergency. Some cooperation is likely to be forthcoming, since other countries will not want to see their currencies appreciate rapidly and far, for reasons of international competitiveness.
An alternative, more foresighted strategy would be for the United States to build foreign exchange reserves or lines of credit during a period of relative financial calm. The swap lines with Europe and Japan that were allowed to expire in late 1998 could be renewed. And the Federal Reserve could begin, initially on a modest scale, to engage in open market purchases of foreign securities. (53) This would put some downward pressure on the dollar exchange rate, which would be welcome to many U.S. firms and would not put undue pressure on an economy whose growth seems to be slowing down. And it would build up U.S. holdings of foreign assets at a time when they are relatively inexpensive. Higher reserves would be welcome if significantly disturbing private switches out of dollar-denominated assets were to occur in the future. It would also represent a partial response to the diminishing supply of Treasury securities projected to be available for open-market purchase in the coming years. (54)
Of course, the Federal Reserve should not purchase foreign securities without cooperation with other central banks, to avoid intervention at cross-purposes. As is well known, in a world of n currencies there are only n -- 1 independent exchange rates; independent action by n central banks is not possible. The "nth country problem" has been generally solved by U.S. abstention from the foreign exchange market. Until September 2000 it seemed the European Central Bank would also adopt a position of abstention, leaving room for some U.S. activity. With European intervention, cooperation in a U.S. buildup of euro-denominated reserves becomes necessary.
IV. Summary and Conclusions
It is often suggested that U.S. monetary policy is largely or even entirely inward oriented, taking into account only the needs of the U.S. economy, with little or no reference to the rest of the world. We have tried to show that this view is largely incorrect, in at least three different respects.
The first, generally accepted respect is that greater engagement with the rest of the world in both trade and financial transactions leads the U.S. economy to be more directly affected by overseas developments than it was three or four decades ago. Exports have occasionally accounted for as much as 40 percent of annual U.S. economic growth; U.S. prices are influenced by world price developments, most obviously in the case of oil. Movements in exchange rates can also affect prices of imported goods, hence prices of domestic substitutes for imported goods. More intense foreign competition has reduced the sensitivity of domestic prices to the pressures of aggregate demand, and it has undoubtedly contributed to shifting downward the level of unemployment that is consistent with price stability. These changes in the structure of the U.S. economy in response to greater openness have of course affected the way the Federal Reserve responds to U.S. economic developments.
Second, a perusal of FOMC records reveals extensive references to international developments in discussions of the future direction of monetary policy. These international factors were not always or even often decisive in determining U.S. monetary policy; but they were frequently factored into the overall evaluation of the economic environment. And occasionally -- during some periods of pronounced dollar weakness, for instance, or during the emergence of the debt crisis in August 1982 and the international financial crisis of late 1998 -- the FOMC's desire to affect the course of these developments played a major role.
Third, external competitive pressures have facilitated substantial changes in the structure of the U.S. financial system. (55) Indeed, arguably, they were an important factor in breaking down the geographical and business barriers that had shaped the U.S. banking system since the 1930s. (56) This interplay between financial innovation and changes in the regulatory structure of the U.S. banking system has in turn affected how monetary policy works. As Frank Morris was one of the first to note, these innovations made the monetary aggregates increasingly poor guides for policy decisions. The ensuing search for a substitute has led to the current policy focus, both in the United States and abroad, on short-term interest rates and the central bank's ultimate goals --price stability and sustainable growth.
These diverse channels of international influence on U.S. monetary policy will no doubt continue and even intensify in the future. In the 1960s, "international" work could generally be left to one designated member of the Federal Reserve Board, with occasional attention from the Chairman. These days most members of the Board are drawn into international matters at least occasionally, as are many of the Reserve Bank presidents, and foreign developments demand much attention from the Chairman. Just as many domestically oriented agencies of the U.S. government, like the SEC or the FBI, are finding that they cannot meet their responsibilities without intensive work with their foreign counterparts, so too the Federal Reserve is likely to grow increasingly internationalized as it strives to stabilize the U.S. economy. Financial supervision and the provision of international lender-of-last-resort facilities are two areas where the need for cooperation is particularly keen. While the development of international stan dards represents considerable progress on the supervisory front, a resolution of the issues surrounding the lender of last resort remains more elusive.
Finally, as our review of FOMC decision-making suggests, on occasion, big exchange rate swings widely viewed as unrelated to macroeconomic fundamentals still plague even the largest economies. Thus, the world's major central banks are likely to want to devote ongoing -- or even increased -- attention to stabilizing their exchange rates. The fact that most major central banks are now focused on attaining similar low rates of inflation should help in this regard. Nevertheless, because exchange rate shifts sometimes reflect forces other than changes in the relative price of traded products, we believe that the Fed will need, among other things, to stand ready to engage more extensively in open market operations in foreign securities. Thus, it will also need to build its stock of such assets.
Oksana Nagayets and Matthew LaPenta provided excellent research assistance. The authors are also grateful to Geoffrey M. B. Tootell for sharing his reaction function model and accompanying data and to Peter Morrow for running the regressions. Many of Little's colleagues at the Boston Fed, particularly Jeffrey Fuhrer, Richard Kopcke, and Giovanni Olivei, made helpful suggestions.
(1.) Federal Reserve Act. The phrase "maximum employment" is generally interpreted to mean maximum sustainable growth and employment.
(2.) Cited by Wells (1994, p. 109) from Volcker and Gyohten (1992, pp. 103-104).
(3.) The shift from targeting monetary aggregates to relying on an exchange rate anchor or, more recently, an inflation target (with or without an intermediate interest rate target) is a global phenomenon.
(4.) These developments will not be covered in this version of the conference paper. See Cooper and Little (2001).
(5.) On a nominal basis, exports plus imports have increased from less than 10 percent to almost 25 percent of GDP. The difference reflects the fact that prices have increased less (or fallen more, in the case of computers and other high-tech equipment) for tradable than for non-tradable products.
(6.) In addition, White (1999) suggests that the stimulative terms-of-trade effects of an appreciation may work to offset the substitution effects, which tend to shift domestic demand to foreign products. He posits that combined terms-of-trade and substitution effects may partly explain the unexpectedly low inflation and strong consumer spending experienced in the United States in recent years.
(7.) Eventually, in the case of the FOMC, the Record of Policy Actions became the Minutes.
(8.) The text indicates, for instance, that "most members noted ..." while "some" or "a few" mentioned a different point.
(9.) The FOMC met sixteen times a year. on average, from 1960 to 1969 and then roughly monthly until 1981 when it began meeting eight times a year. To adjust for these differences in the frequency of meetings, in Figure 5, the number of FOMC decisions partly influenced by international considerations was divided by 2 between 1960 and 1971 and by 1.5 between 1972 and 1982. Because the Board of Governors changes the discount rate whenever it deems appropriate, the number of discount rate decisions influenced by international factors was not adjusted. Neither were the rare occasions when international factors were the primary reason for FOMC policy decisions.
(10.) Regulation D governs depository institution (originally member bank) reserve requirements. Regulation M governs the foreign branches of member banks.
(11.) The episodes we identified as times when international factors had a primary influence largely correspond with dates mentioned by Eichengreen (2000), Volcker and Gyohten (1992), and Wells (1994) as the rare instances when international pressures caused a change in Fed policy.
(12.) The official settlements balance and the liquidity balance, two other measures that attracted much attention in the 1960s, also improved in 1968 and 1969 and in 1970, respectively. See Eichengreen (2000) and Fieleke (1971) for a discussion of these measures.
(13.) The decade was studded by a series of currency crises involving the dollar to be sure, but also the British pound, the French franc, and the Canadian dollar, as well as recurring speculation concerning possible appreciation of the deutsche mark.
(14.) At that time Regulation Q (governing interest rates on deposits) frequently set ceilings on interest rates payable by member and FDIC-insured nonmember banks below rates available in the money market and at thrift institutions. The goal was to anchor interest rates, particularly for mortgage loans, and to prevent excessive competition for funds from leading to high-risk lending. As a result, during periods of rising interest rates, the banks generally had difficulty attracting or holding funds.
(15.) For instance, in May 1967, the FOMC decided that "purchases of coupon issues, if and when feasible," could lighten the supplies of government securities in the maturities in which supplies were heaviest. They also noted that the substitution of purchases of coupon issues for purchases of bills "could be important for balance of payments reasons, as a means of reducing downward pressures on bill rates." Prior to 1961, System open market transactions were conducted only in short-term securities. With the change in policy, 30 percent of the securities purchased outright in 1961 had a maturity of more than one year.
(16.) The Europeans took action in May 1971; the Japanese floated in August. The Canadian dollar had been floating since June 1970.
(17.) Mileposts on the path to the European Monetary Union (EMU) included the three-stage Werner Plan for monetary unification (1970); the establishment of the joint float or "snake in the tunnel," an arrangement in which most EC members kept their currencies within a 4.5 percent band which undulated within a 9 percent band around the dollar (1973); the establishment of the European Monetary System (1979); the signing and ratification of the Treaty of Maastricht, which laid out the criteria for membership in the European Monetary Union (1992 and 1993); and the start of EMU and the euro, the single European currency, on January 1, 1999.
(18.) While the oil shocks clearly emanated from abroad, their impact became part of the domestic outlook. Moreover, the tensions involved reflected the domestic need to choose between offsetting either the demand or the price effects of an oil price shock, not a conflict between domestic and international priorities.
(19.) Rather, the renewed influence of international issues from 1982 to 1984 shown in Figure 5 reflected the impact of the LDC debt crisis on the large money center banks, as discussed further below.
(20.) These dollar sales involved the purchase of German marks and Japanese yen. By this time, James Baker had replaced Donald Regan as Secretary of the Treasury.
(21.) The $2 billion represented the bulk of the windfall accruing to the government when the nation revalued gold from $20.67 to $35 per ounce.
(22.) See Pauls (1990) and Solomon (1999, p. 15).
(23.) According to the Bank for International Settlements' (BIS) triennial Central Bank Survey of Foreign Exchange and Derivatives Market Activity last conducted in 1998. The number includes spot transactions, outright forwards, and forex swaps.
(24.) If so, whether or not intervention is always fully sterilized becomes a moot point.
(25.) The Treasury's Exchange Stabilization Fund had begun conducting foreign exchange operations, with the New York Fed acting as agent, in March 1961. These operations were part of a cooperative effort involving treasuries and central banks on both sides of the Atlantic to counter disorderly conditions in the foreign exchange markets (Board of Governors 1962, p. 54). The dissenting governors were Governors Mitchell and Robertson.
(26.) Membership in the swap network eventually included the central banks of Austria, Belgium, Canada, Denmark, England, France, Germany, Italy, Japan, Mexico, the Netherlands, Norway, Sweden, and Switzerland, and the Bank for International Settlements. At their peak in early 1995, these credit lines equaled $35.4 billion. As discussed below, most of these agreements have been allowed to lapse.
(27.) In 1989 and 1990 this group included Governors Angell, Johnson, and LaWare, and President Hoskins from Cleveland.
(28.) Board of Governors (1990, p. 33).
(29.) Expressed particularly by Governor Lindsey and Presidents Broaddus (Richmond) and Jordan (Cleveland).
(30.) Board of Governors (1994, p. 161).
(31.) These numbers exceeded the prevailing estimates of potential growth and the NAIRU.
(32.) Tootell's explanatory variables were Board staff forecasts of real GDP growth and core inflation in the six-month period immediately ahead and in the following six-month period, the forecast civilian unemployment rate one quarter ahead, and the lagged three-month moving average of Ml growth.
(33.) See Solomon (1982) and Perry and Tobin (2000), particularly the pieces by Bator, Cooper, and Eichengreen.
(34.) For example, in the fall of 1967 members of the FOMC cited the pressures on the pound as one of several reasons not to tighten U.S. monetary conditions.
(35.) Except for the impediments of the capital controls.
(36.) For example, the development of the commercial paper market in the 1970s and 1980s was luring the banks' best corporate borrowers.
(37.) Cited in Curry (1997, P. 206).
(38.) These specific problems occurred in the context of unusually high interest rates. Interest rates on 3-month CDs in the secondary market averaged 16 percent in 1981 and 12 percent in 1982.
(39.) Similarly, in May 1983 five members dissented from a shift to slightly more restraint, in part because of the "tenuous" situation in some developing countries.
(40.) Cline (1995, pp. 234-35). Debt forgiveness was calculated as the sum of the reduction in the face value of the original debt and the reduction in the present value of interest reductions, less the amount spent on buybacks. Not all of this debt forgiveness actually resulted in losses to the banks, however, because the value of much of this Brady debt appreciated on the secondary market that soon developed.
(41.) The Basle Concordat established the principle that no foreign bank should escape adequate supervision.
(42.) In 1981, one of the world's top ten largest banks, in terms of assets, was Japanese. In 1987, the seven largest banks were Japanese (Wagster 1996).
(43.) In 1987, unrealized capital gains seemingly comprised the bulk of Japanese banks' capital. Equity plus reserves with no debt provisions (similar to core or Tier 1 capital) amounted to about 2 percent of Japanese bank assets. The ratios for U.S. and U.K. banks were 4.9 percent and 5.4 percent, respectively (Wagster 1996).
(44.) In the case of consumer and industrial loans, where foreign competition was particularly intense, the foreign share was approaching 20 percent. Counting loans booked offshore, Japanese banks were thought to have captured about 12 percent of the U.S. banking market (Wagster 1996).
(45.) Indeed, critics like Jeffrey Sachs have pointed out that the Basle capital standards may have contributed to the Asian crisis because the risk weights encouraged short-term interbank lending. Interbank claims with a residual maturity of one year or less carried a risk-weight of 20 percent, regardless of where the borrowing bank was incorporated. Longer-term claims on banks incorporated in an OECD member have a 20 percent weight. Longer-term claims on other banks are weighted at 100 percent.
(46.) The Basle Committee on Banking Supervision has sought to apply capital standards to market, interest rate, and operational risk as well as to credit risk. For this purpose, and again under pressure from the United States, where risk measurement and management are relatively well advanced, the Committee has agreed to accept a bank's internal model for assessing "market value-at-risk." In a related initiative, the Fisher Committee, a subcommittee of the BIS Eurocurrency Standing Committee, has also urged greater disclosure, as opposed to regulation, of various types of risk--particularly for those risks associated with off-balance-sheet items such as derivatives.
(47.) Mexico prepaid all of its U.S. swap drawings.
(48.) Plus Mexico, in the case of the swaps.
(49.) Actually, all of the Fed's swap transactions with Mexico were short-term. By contrast, the bulk of the Treasury's swaps were medium-term.
(50.) The congressional constraints on ESF lending that followed the Mexican peso crisis limited Treasury--Fed participation in the Thai loan package but had expired when the Indonesian and Korean packages were put together.
(51.) The Bank of Japan has recently begun announcing an inflation forecast as a means for guiding inflation expectations.
(52.) Failure to do so subjects them to dumping charges and leads to the imposition of anti-dumping duties. Once imposed, such duties are difficult to remove.
(53.) In June 1999 the Bank of England began to accept euro-denominated bonds issued by European governments and international agencies as collateral for repurchase agreements. The Bank started to conduct open market operations via repos in early 1997.
(54.) This strategy might also deliver some advantages in terms of increased risk-sharing across nations, which analysts suggest could increase welfare by noticeable amounts. See Athanasoulis and van Wincoop (2000).
(55.) As noted at the beginning of this article, this subject is dealt with in detail in the conference paper from which this article has been taken (Cooper and Little 2001).
(56.) The Board of Governors (as opposed to the FOMC) spends much time on regulatory issues, which have been dominated, except in periods of financial crisis, by the changing competitive structure of financial services.
Athanasoulis, Stefano G. and Eric van Wincoop. 2000. "Growth Uncertainty and Risk-Sharing." Journal of Monetary Economics 45(3), pp. 477-505.
Bank for International Settlements. 1999. Central Bank Survey of Foreign Exchange and Derivatives Market Activity 1998. Basle, May.
Beek, David. C. 1977. "Commercial Bank Lending to the Developing Countries." Federal Reserve Bank of New York Quarterly Review, Summer, pp. 1-8.
Board of Governors of the Federal Reserve System. 1960 to 1999. Annual Report. Washington, DC.
Cline, William R. 1995. International Debt Reexamined. Washington, DC: Institute for International Economics.
Cooper, Richard N. 1999. "Exchange Rate Choices." In lane Sneddon Little and Giovanni P. Olivei, eds., Rethinking the International Monetary System, pp. 99-123. Federal Reserve Bank of Boston Conference Series No. 43.
_____. 2000. "Foreign Economic Policy in the 1960s: An Enduring Legacy." In George L. Perry and James Tobin, eds., Economic Events, Ideas and Policies: The l960s and After. Washington, DC: Brookings Institution Press.
Cooper, Richard N. and Jane Sneddon Little. 2001. "U.S. Monetary Policy in an Integrating World: 1960 to 2000." In Richard W. Kopcke and Lynn Elaine Browne, eds., The Evolution of Monetary Policy and the Federal Reserve System Over the Past Thirty Years, pp. 77-121. Federal Reserve Bank of Boston Conference Series No.45.
Curry, Timothy J. 1997. "The LDC Debt Crisis." History of the Eighties-Lessons for the Future. Vol. 1. Federal Deposit Insurance Corporation.
Dominguez, Kathryn and Jeffrey Frankel. 1993a. "Does Foreign-Exchange Intervention Matter? Disentangling the Portfolio Effect." The American Economic Review 83 (5), December, pp. 1356-69.
_____. 1993b. Does Foreign Exchange Intervention Work? Washington, DC: Institute for International Economics.
Edison, Hali J. 1993. "The Effectiveness of Central Bank Intervention: A Survey of the Literature after 1982." Special Papers in International Economics No. 18, Princeton University.
Eichengreen, Barry. 2000. "From Benign Neglect to Malignant Preoccupation: U.S. Balance of Payments Policy in the 1960s." In George L. Perry and James Tobin, eds., Economic Events, Ideas and Policies: The 1960s and After. Washington, DC: Brookings Institution Press.
Fieleke, Norman S. 1971. "Accounting for the Balance of Payments." New England Economic Review, May/June, pp. 3-15.
Henderson, Dale and Stephanie Sampson. 1983. "Intervention in Foreign Exchange Markets: A Summary of Ten Staff Studies." Federal Reserve Bulletin, November, pp. 830-36.
Humpage, Owen F. 1996. "U.S. Intervention: Assessing the Probability of Success." Working Paper No. 9608, Federal Reserve Bank of Cleveland. September.
_____. 1998. "The Federal Reserve as an Informed Foreign-Exchange Trader." Working Paper No. 9815, Federal Reserve Bank of Cleveland. September.
Lewis, Karen K. 1995. "Are Foreign Exchange Intervention and Monetary Policy Related, and Does It Really Matter?" Journal of Business 68 (2), pp. 185-213.
McKinnon, Ronald I. and Kenichi Ohno. 1997. Dollar and Yen: Resolving Economic Conflict Between the United States and Japan. Cambridge, MA: The MIT Press.
McNees, Stephen K. 1986. "Modeling the Fed: A Forward-Looking Monetary Policy Reaction Function." New England Economic Review, November/December, pp. 3-8.
_____. 1992. "A Forward-Looking Monetary Policy Reaction Function: Continuity and Change." New England Economic Review, November/December, pp. 3-13.
Miller, William G. 1978. "Statement to Congress." Presented before the Subcommittee on Financial Institutions of the Committee on Banking, Housing and Urban Affairs, U.S. Senate. June 25.
Pauls, Dianne B. 1990. "U.S. Exchange Rate Policy: Bretton Woods to Present." Federal Reserve Bulletin, November, pp. 891-908.
Perry, George L. and James Tobin, eds. 2000. The Economic Events, Ideas, and Policies: The 1960s and After. Washington, DC: Brookings Institution Press.
Savage, Donald T. 1993. "Interstate Banking: A Status Report." Federal Reserve Bulletin, December, pp. 1075-89.
Saxton, Jim. 1997. "Establishing Federal Reserve Inflation Goals." A Joint Economic Committee Study, U.S. Congress. April.
Schwartz, Anna J. 2000. "The Rise and Fall of Foreign Exchange Market Intervention." National Bureau of Economic Research Working Paper No. 7751, June.
Solomon, Robert. 1982. The International Monetary System, 1945-1981. New York: Harper & Row.
_____. 1999. Money on the Move: The Revolution in International Finance since 2980. Princeton, NJ: Princeton University Press,
Terrell, Henry S. 1973. "The Foreign Term-Lending Activities of U.S. Banks." International Finance Discussion Paper No. 23, Division of International Finance, Board of Governors of the Federal Reserve System. February 2.
Tootell, Geoffrey M. B. 1997. "How Farsighted Is the FOMC?" New England Economic Review, January/February, pp.49-65.
Volcker, Paul A. and Toyoo Gyohten. 1992, Changing Fortunes: The World's Money and the Threat to American Leadership. New York: Times Books.
Wagster, John D. 1996. "Impact of the 1988 Basle Accord on International Banks." The Journal of Finance, Vol. LI (September), pp. 1321-46.
Wells, Wyatt C. 1994. Economist in an Uncertain World: Arthur Burns and the Federal Reserve, 1970-78. New York: Columbia University Press.
White, William R. 1996. "International Agreements in the Area of Banking and Finance; Accomplishments and Outstanding Issues." Bank for International Settlements, Working Paper No. 38, October.
_____. 1999. "Evolving International Financial Markets: Some Implications for Central Banks." Bank for International Settlements, Working Paper No. 66, April.
Appendix Table 1 FOMC Voting and Reaction Functions Logit Estimations of FOMC Voting (1) Original (2) Original (3) Original Equation Equation Equation with Dollar with Dollar Purchases Sales Dependent Variable: Vote for Tightening C -.02 -.26 -.05 (-.04) (-.69) (-.13) QH1 .07 .07 .09 (2.24) (2.48) (3.08) QH2 .11 .13 .05 (2.68) (3.02) (1.18) PH1 .12 .18 .16 (1.57) (2.39) (2.16) PH2 .11 -.0008 .11 (1.31) (-.009) (1.34) URF -.46 -.41 -.46 (-8.40) (-7.54) (-8.43) M .04 .04 .05 (3.70) (3.47) (3.99) RBUY .03 (5.27) RSELL .05 (5.44) Dependent Variable: Vote for Loosening C 1.56 1.76 1.52 (3.82) (4.27) (3.71) QH1 -.31 -.33 -.31 (-10.81) (-11.22) (-10.88) QH2 -.04 -.04 -.05 (-0.93) (-.84) (-1.12) PH1 .13 .07 .16 (1.55) (.80) (1.85) PH2 -.41 -.35 -.45 (-4.35) (-3.57) (-.63) URF -.08 .09 -.07 (-1.50) (-1.66) (-1.33) M -.05 -.06 -.05 (-4.22) (-4.55) (-4.02) RBUY -.03 (-3.17) RSELL -.006 (-1.42) Observations 2406 2406 2406 Log-Likelihood -2002.7 -1977.8 -1979.4 Reaction Functions (4) Original (5) Original (6) Original Equation Equation Equation with Dollar with Dollar Purchases Sales Dependent Variable: Dependent Vote for Tightening Variable: Change in Federal Funds Rate C -.10 .08 -.04 (-.62) (-.41) (-.20) QH1 .01 .01 .009 (.87) (.87) (.73) QH2 .05 .06 .05 (2.49) (2.70) (2.59) PH1 .002 .02 .006 (.06) (.38) (.14) PH2 .04 .02 .03 (.94) (.37) (.73) URF -.5 -.06 -.06 (-2.31) (-2.14) (-2.38) M .009 .009 .01 (1.58) (1.63) (1.63) RBUY .005 (1.87) RSELL -.000004 (.002) Dependent Variable: Vote for Loosening C QH1 QH2 PH1 PH2 URF M RBUY RSELL Observations 237 211 211 Log-Likelihood -116.8 -108.1 -109.9 Definitions of Variables QH1 Forecast of real GDP growth over the next 6 months QH2 Forecast of real GDP growth over the 6 months starting 6 months from now PH1 Forecast of inflation over the next 6 months PH2 Forecast of inflation over the 6 months starting 6 months from now URF 1-quarter-ahead civilian unemployment forecast M Lagged 3-month moving average of Ml growth RBUY U.S. official purchases of dollars (deflated by GDP deflator) in intermeeting period just past (millions of dollars) RSELL U.S. official sales of dollars, that is, negative purchases (deflated by GDP deflator in intermeeting period just past (millions of dollars). The Green Book was used for all the forecasts.