International Capital Markets
Developments, Prospects, and Key Policy Issues
September 1997

Published by the International Monetary Fund
© 1997


A.    Foreign Exchange Markets: Capital Flows and the Rise of the Dollar

During the past twelve months, the key development in the major foreign exchange markets was the rise of the dollar against the deutsche mark and yen. The dollar's ascent occurred in relatively calm markets, and was interrupted only occasionally by increased market uncertainty about interest rates and U.S. equity prices. From its low of below ¥80 and DM1.35 in the spring of 1995, the dollar rose more than 50 percent against the Japanese yen and about 25 percent against the deutsche mark by May 1997 (Figures 1 and 2). In percentage terms, the movement in the dollar-yen rate was one of the largest dollar-yen movements to occur over a two-year period. As was recognized in the communique of the meeting of G-7 Finance Ministers in April 1997, the movements in the dollar away from the low reached in the spring of 1995 generally brought currencies into better balance. Roughly 10 percent of the yen's decline against the dollar was reversed in May and June 1997.

Figure 1

Figure 2

The dollar's strength since early 1996 was to a large extent due to differences in cyclical positions of the U.S. vis-à-vis Europe and Japan, as well as the related large inflows of foreign capital into the dollar markets, in particular into the U.S. domestic fixed-income markets, and relatively attractive dollar yields. Foreign net purchases of long-term U.S. government and corporate bonds reached a new record high in 1996, almost 70 percent larger than the historical high reached in 1995 (Table 1). During these two record setting years, foreign purchases of U.S. Treasury bonds exceeded the cumulative net purchases over the previous 10 years (1985-94). Much of the flow into dollar fixed-income markets came from Germany, France, Spain, Japan, and China, and large dollar flows also originated in the major international financial centers—Hong Kong1, Singapore, and the United Kingdom. By comparison, and despite the impressive performance of U.S. equity prices during the period under review, foreign investors accumulated less than $14 billion in U.S. equities in 1996.

Table 1. Net Foreign Purchases of U.S. Bonds, 1993-First Quarter 1997
(In millions of U.S. dollars)

Government
Bonds
Corporate
Bonds
 Total

1993 58,980 30,572 89,552
1994 100,481 37,992 138,473
1995 162,844 57,853 220,697
1996 293,685 77,978 371,663
Of which
 Europe 137,148 56,194 193,342
    Germany 19,297 3,514 22,811
    United Kingdom 76,323 43,702 120,025
    Spain 18,421 462 18,883
 Asia 112,597 9,806 122,403
    Japan 48,985 6,099 55,084
    People's Republic of China 17,209 257 17,466
    Hong Kong 15,281 1,737 17,018
1997
 1st quarter
77,048 20,826 97,874

Source: U.S. Department of Treasury, Treasury Bulletin, various issues.

The demand for dollar assets has been large not only in absolute terms, but also in relation to the U.S. current account deficit and to inflows into other mature markets. Net foreign purchases of U.S. bonds totaled about 5 percent of U.S. GDP in 1996, compared with a U.S. current account deficit of about 2 percent. On a bilateral basis, the magnitudes of capital flows into U.S. bond markets generally exceeded the current account surpluses of source countries (Germany and the United Kingdom had current account deficits). By comparison, foreign purchases of U.S. Treasury bonds were 10 times foreign purchases of either German Bunds or Japanese domestic bonds (government and private) (Table 2).

Table 2. Net Purchases of Domestic Bonds by Nonresidents, 1993-96
(In billions of U.S. dollars)
United
States
Japan
Germany


France United
Kingdom
Canada
Bunds Other

1993 89.55 -31.07 93.15 33.29 19.94 20.82 21.70
1994 138.47 -13.72 2.95 11.67 -36.15 1.85 10.83
1995 220.70 -8.46 35.38 25.35 2.84 5.26 21.78
1996 371.66 25.27 26.91 38.06 -31.29 15.75 13.58

Source: BZW Securities Limited; and U.S. Department of Treasury, Treasury Bulletin, various issues.

Official reserve accumulation accounted for more than two-fifths of total foreign net purchases of U.S. Treasury securities in 1996, half of which was the consequence of developing-county efforts to manage exchange rates in the presence of substantial and volatile capital flows. During 1995-96, the Japanese monetary authorities accounted for the single largest accumulation of foreign exchange reserves, as total reserves (minus gold) reached $217 billion at end-1996, an increase of more than $90 billion since end-1994.

Private capital flows into U.S. markets totaling $425 billion were counterbalanced in 1996 by a substantial resurgence in U.S. investment abroad after the retrenchment from foreign markets in 1994; hence, official net capital inflows almost matched the U.S. current account deficit (Table 3). In 1995-96, the scale of these matching private flows were large relative to GDP. Increasing diversification of portfolios accounts for some of the two-way trade in financial assets, but the scale of such flows is also consistent with the view that the United States is playing the role of a global intermediary: it attracts international capital by providing relatively safe, liquid instruments (U.S. government and high-grade corporate debt securities) at relatively high returns and then reinvests them through international markets in less liquid vehicles for higher returns.2 A rationale for this role is that U.S. based institutional investors and global financial institutions are generally perceived as possessing advanced knowledge, expertise, and global reach in placing funds in the higher-yielding markets around the world.

Table 3. United States: International Transactions, 1990-96
1990 1991 1992 1993 1994 1995 1996
(In billions of U.S. dollars)

Current account balance -91.9 -5.7 -56.4 -90.8 -133.5 -129.1 -148.2
Foreign assets in the United States, net (increase/capital inflow (+)) 141.0 109.6 168.8 279.7 297.3 451.2 547.6
Foreign official assets, net 33.9 17.4 40.5 71.8 40.4 110.7 122.4
Other foreign assets, net 107.1 92.3 128.3 207.9 257.0 340.5 425.2
   Direct investment 47.9 22.0 17.9 49.0 45.7 67.5 77.0
   Domestic securities -0.9 54.0 67.2 104.5 91.2 195.9 289.4
   U.S. liabilities to
      unaffiliated
      foreigners
      reported by
      U.S.non-banking
      concerns
45.1 -3.1 13.6 10.5 -7.7 34.6 31.8
   U.S. liabilities reported
      by U.S. banks,
      not included
      elsewhere
-3.8 4.0 16.2 25.1 104.3 30.2 9.8
   U.S. currency flows 18.8 15.4 13.4 18.9 23.4 12.3 17.3
U.S. assets abroad, net (increase/capital outflow (-)) -74.0 -57.9 -68.8 -194.5 -160.5 -307.2 -352.4
U.S. official reserve
   assets plus
   government
   assets, net
0.1 8.7 2.2 -1.7 5.0 -10.3 6.0
U.S. private assets, net -74.2 -66.6 -71.0 -192.8 -165.5 -296.9 -358.4
   Direct investment -30.0 -31.4 -42.6 -77.9 -69.3 -86.7 -87.8
   Foreign securities -28.8 -45.7 -49.2 -146.3 -60.3 -100.1 -108.2
   U.S. claims on
      unaffiliated
      foreigners
      reported by U.S.
      non-banking concerns
-27.8 11.1 -0.4 0.8 -31.7 -35.0 -64.2
   U.S. claims reported
      by U.S. banks,
      not included
      elsewhere
12.4 -0.6 21.2 30.6 -4.2 -75.1 -98.2
Unrecorded outflows
   (statistical discrepancy)
24.9 -46.1 -43.6 5.6 -3.3 -14.9 -46.9
Memorandum items: (In percent of GDP)
Current account balance -1.6 -0.1 -0.9 -1.4 -1.9 -1.8 -2.0
Foreign assets in the
   United States, net
   (increase/capital
   inflow (+))
2.5 1.9 2.7 4.3 4.3 6.2 7.2
U.S. assets abroad,
   net (increase/capital
   outflow (-))
-1.3 -1.0 -1.1 -3.0 -2.3 -4.2 -4.7

Source: International Monetary Fund, World Economic Outlook database; and U.S. Department of Commerce, Survey of Current Business, various issues.

Large differentials between U.S. and German and between U.S. and Japanese interest rates were a key factor driving the large flows into U.S. markets and the dollar's strength. At the same time that interest rates remained low and even declined throughout Europe and Japan, monetary conditions began to tighten in the United States as market participants became concerned that economic activity continued to increase relative to estimates of U.S. capacity output. The Federal Reserve's 25 basis point increase in the federal funds rate in late March 1997 widened the differentials even further. In effect, the combination of asynchronous business cycles and divergent monetary conditions accounted for the relatively wide spreads (Figure 2). U.S. interest-rate differentials with Germany have been most significant at the shorter end of the maturity spectrum, and the spread on 3-month rates stood at 250 basis points at end-May 1997. Spreads between yen- and dollar-denominated bonds have been particularly large: long-term spreads have been in the range of 300-500 basis points since mid-1995, and averaged about 420 basis points in May 1997, and short-term spreads have been in the range of 450-525 basis points, and averaged about 510 basis points in May 1997.

Large global macro hedge funds viewed the relatively wide yen-dollar interest-rate spread as a potentially lucrative trading opportunity, because they presumed that the Bank of Japan did not want the yen to strengthen in 1996-97 and preferred not to raise interest rates in light of the continuing cyclical weakness as well as concerns over the loan books of Japanese banks and the banks' relatively large exposure to interest rates. If expectations about the yen-dollar rate and yen interest rates proved correct, then borrowing cheaply in yen, selling yen for dollars, and lending the proceeds to the U.S. Treasury would generate a net profit equal to the sizable interest-rate differential. While Japanese banks reduced total cross-border positions by $20 billion in 1996, they increased lending to nonbank entities located just in the Cayman Islands by almost $19 billion—a home for some major hedge funds. On the other side of the ledger, entities in the Cayman Islands accumulated $20 billion of U.S. long-term bonds in 1996.3 These yen-carry trades were even more profitable than anticipated because the yen depreciated in 1996 and the first four months of 1997.

A second factor boosting flows into U.S. markets was a diversification out of yen and deutsche mark denominated instruments, in part reflecting a precautionary motive to avoid risks associated with uncertainty about the likelihood, timing, and country composition of EMU4 and about the resolution of financial system problems in Japan. This revealed preference for reducing yen and deutsche mark exposures was also associated with the appreciation of the currencies of some of the higher-yielding EU and "dollar-bloc" countries, such as the United Kingdom, and Australia, Canada, and New Zealand. A possible third interrelated factor was that Europe and Japan were net suppliers of international liquidity. In both locations, real money growth exceeded output growth, and domestic demand for funds fell short of domestic supplies.

In European currency markets, renewed optimism about the prospects for EMU and progress in reducing inflation and fiscal deficits led to the appreciation of several EU currencies against the deutsche mark (Figure 3). The Finnish markka joined (October), and the Italian lira reentered (November), the exchange rate mechanism (ERM) of the European Monetary System in 1996. While both currencies strengthened upon entry, they have since lost these gains. The strongest currencies have been the pound sterling and the Irish punt, both of which have been supported by robust economic activity and expectations of rising interest rates. The Irish punt is the most appreciated currency in the ERM grid, having risen about 10 percent since mid-1996 above its central rate against the deutsche mark, and the pound sterling was the only major currency to appreciate against the dollar over the past 18 months.

Figure 3

Despite the large swings in the major currencies, month-to-month volatility in foreign exchange markets fell substantially in 1996, particularly for second-tier European currencies (Figure 4).5 A tangible effect of the drop in volatility has been a sharp drop in turnover in currency spot markets, and both of these developments were associated with a scaling back of European foreign exchange trading and dealing operations.

Figure 4


1Hong Kong, China, is hereinafter referred to in this report and background material as "Hong Kong."

2In the 1960s, C. Kindleberger (1965) argued the United States played the role of international banker, selling liquid, short-term obligations to nonresidents and buying longer-term claims against them. Triffin (1966) claimed U.S. short-term dollar liabilities to nonresidents were too large relative to the U.S. gold reserve, a view that led to the establishment of the SDR. It is also consistent with the view that capital markets have become more globally integrated. According to M. Feldstein and C. Horioka (1980), in an integrated global capital market, domestic savings and domestic investment will be uncorrelated. For a survey see M. Goldstein and M. Mussa (1993).

3These data are from the Bank for International Settlements and the U.S. Department of Treasury, Treasury Bulletin.

4The diversification away from Europe and toward the dollar can be seen as reflecting uncertainty about the initial strength or weakness of the euro once EMU begins. There were also occasions when the deutsche mark strengthened against the dollar and other European currencies, usually on the release of economic and financial data or on policy developments that called into question whether EMU would go ahead on time, in which case the deutsche mark could be seen as a "safe haven" within Europe.

5Lower volatility has led to, and perhaps been supported by, increased activity in currency options (binary and range options), which allow investors to fine-tune exposures to the level, direction of change, and volatility of underlying asset prices.


Table of contentsNext