Abstract

Slowing economic growth and mounting concerns about corporate earnings and high corporate leverage initially caused a sharp decline in equity prices and widening credit spreads in late 2000. The fall in equity prices was especially pronounced among technology stocks, which experienced virtually simultaneous dramatic declines in all the major economies. Fixed-income markets deteriorated in response to concerns about credit risk. In the high-yield market, in particular, flows dried up and spreads peaked at the highest levels since the 1990–91 recession. Markets revived in early 2001 following significant easing in U.S. monetary policy, with a particularly pronounced rebound in the high-yield market. Nevertheless, on balance, equity prices were lower and credit spreads generally higher at end-May 2001 than a year earlier. Despite the sharp repricing in U.S. financial markets, the record U.S. current account deficit, and substantially more monetary policy easing in the United States than abroad, the dollar continued to strengthen, as international investors showed a sustained strong appetite for private U.S. assets. Deteriorating market conditions weighed on bank earnings but, except in Japan, no concerns arose about the stability of any major banking system.

Slowing economic growth and mounting concerns about corporate earnings and high corporate leverage initially caused a sharp decline in equity prices and widening credit spreads in late 2000. The fall in equity prices was especially pronounced among technology stocks, which experienced virtually simultaneous dramatic declines in all the major economies. Fixed-income markets deteriorated in response to concerns about credit risk. In the high-yield market, in particular, flows dried up and spreads peaked at the highest levels since the 1990–91 recession. Markets revived in early 2001 following significant easing in U.S. monetary policy, with a particularly pronounced rebound in the high-yield market. Nevertheless, on balance, equity prices were lower and credit spreads generally higher at end-May 2001 than a year earlier. Despite the sharp repricing in U.S. financial markets, the record U.S. current account deficit, and substantially more monetary policy easing in the United States than abroad, the dollar continued to strengthen, as international investors showed a sustained strong appetite for private U.S. assets. Deteriorating market conditions weighed on bank earnings but, except in Japan, no concerns arose about the stability of any major banking system.

Global Capital Flows and Developments in Foreign Exchange Markets

Global Capital Flows

With the globalization of finance, international capital flows have grown dramatically in the 1990s. Between 1990 and 1998, assets managed by mature market institutional investors more than doubled to over $30 trillion, about equal to world gross domestic product (GDP). Amid widespread capital account liberalization and increased reliance on securities markets, these investable funds became increasingly responsive to changing opportunities and risks in a widening set of regions and countries. Because global investment portfolios are large, proportionally small portfolio adjustments can be associated with large and volatile swings in capital flows. In 1997, for example, gross financing to emerging markets (which peaked in that year) and net foreign purchases of U.S. long-term securities were each equivalent to an adjustment of only about 1 percent in institutionally managed assets. Nevertheless, these adjustments sometimes had a significant impact on financial conditions in the recipient countries both when they flowed in and when they flowed out. This underscores the powerful impact that portfolio rebalancing by global investors can have on the volume, pricing, and direction of international capital flows and on conditions in both domestic and international markets. The period under review was no exception.

In 2000, the United States continued to absorb the lion’s share of global capital flows, attracting 64 percent of world net capital exports (as measured by the U.S. current account deficit relative to the sum of current accounts of surplus countries), compared with 60 percent in 1999, and an average of about 35 percent during 1992–97 (Figure 2.1). Net inflows to the United States exceeded $400 billion (Figure 2.2), including a record level of foreign portfolio investment that nearly could have financed the U.S. current account deficit on its own.1 As in previous years, overseas investors (particularly in Europe) bought large quantities of U.S. equities and corporate bonds and cut back net purchases of U.S. treasury securities, as the supply of U.S. treasuries shrank (Table 2.1).2 Gross foreign purchases of U.S. equities were particularly strong, rising to $3.6 trillion—a six-fold increase since 1996. Foreign direct investment (FDI) picked up to a record level of about $150 billion, about half of which represented mergers and acquisitions (M&A) activity. Net capital inflows from Europe continued apace in the first quarter of 2001, including continued strong purchases of U.S. corporate bonds and equities.

Figure 2.1.
Figure 2.1.

United States: Current Account Deficit as Share of Global Surpluses

(In percent)

Source: IMF, World Economic Outlook database.
Figure 2.2.
Figure 2.2.

Global Capital Flows1

(In billions of U.S. dollars)

Sources: IMF, World Economic Outlook database; IMF, International Financial Statistics: and ECS Monthly Bulletin.1The total net capital flows are the sum of direct investment, portfolio investment, and other investment flows. These figures do not include reserve assets. “Other investment” includes bank loans and deposits.2Total net private capital flows.
Table 2.1.

Net Foreign Purchases of U.S. Long-Term Securities

(In millions of U.S. dollars)

article image
Source: United States, Department of the Treasury, Treasury Bulletin (various issues).

Includes bonds issued by U.S. government corporations and federal agencies.

The sustained strong international appetite for private U.S. assets was maintained despite the deterioration in the U.S. economy and financial markets. The continued interest in U.S. investments no doubt strongly reflected investor beliefs about future prospects for the U.S. economy and financial markets. Notwithstanding uncertainties about the extent to which the U.S. slowdown reflected structural rather than cyclical factors, investors appeared to have been optimistic that the U.S. economy would rebound quickly from the current slowdown and reestablish strong growth relative to other developed countries. Accordingly, many investors seemed to believe that the correction in U.S. financial markets would be short-lived and that U.S. markets would outperform other markets on a risk-adjusted basis in the medium term. Underpinning both of these beliefs, investors evidently had considerable confidence that easier U.S. monetary policy would both dampen the cyclical slowdown and limit its repercussions for U.S. financial markets.

Underlying the net international capital outflows from Europe, euro-area investors sharply increased their net purchases of foreign portfolio assets—particularly equities, which rose by 85 percent. Meanwhile, foreign investors sold significant amounts of European shares received through cross-border M&A transactions.3 In addition, foreigners increased their purchases of euro-area debt securities as the differential between U.S. and euro-area bond yields narrowed. Both inward and outward FDI rose, as both European and foreign corporations diversified their operations internationally.

Japanese net capital outflows picked up in 2000, as both FDI and portfolio investment outflows rose. Foreign investors shifted their net purchases from equity to fixed-income markets and bought significant amounts of Japanese government bonds (JGBs). This may have reflected concerns by global fixed-income investors that they might underperform global benchmark indices. Many of these investors reportedly had held underweight positions in JGBs vis-à-vis international benchmark weightings while JGB prices rose.4 Both foreign purchases of Japanese equities and Japanese purchases of foreign securities seem to have picked up more recently. In March 2001, Japanese purchases of foreign bonds reached their highest level since June 2000.

Foreign direct investment continued to account for a substantial share of net capital flows from mature to emerging markets.5 Overall private flows declined, perhaps reflecting a reap praisal of the risk-reward trade-off in emerging market investments following successive crises in Asia, Russia, and other countries during past years. During 2000, the decline in flows also may have been influenced by concerns about the impact of the global slowdown on smaller countries. FDI inflows declined slightly but remained reasonably strong as foreign investors acquired emerging market telecommunications companies, power and other utilities companies, and banks. Net banking outflows continued to increase as petroleum exporters accumulated dollars in overseas bank accounts.

Although net capital flows provide useful insights about balance-of-payments financing and net funding requirements, they can considerably understate the volume and volatility of international portfolio rebalancing. Gross flows more closely reflect international transactions and are more relevant in terms of their impact on market prices and volatility. Since the mid-1980s, on a global basis they have risen sharply to about six times the level of net flows (Figure 2.3).6 The high level of gross flows relative to net flows suggests that countries and regions that have small net capital flows can nevertheless experience substantial gross inflows and outflows of capital. This also may help explain why such countries and regions may experience considerable volatility in asset prices despite relatively small net financing needs.

Figure 2.3.
Figure 2.3.

Gross Global Capital Flows Relative to Net Global Capital Flows1

(In percent)

Sources: IMF, World Economic Outlook database; and IMF, International Financial Statistics.1Ratio of the sum of absolute values of gross inflows and gross outflows to the sum of absolute values of current account balances.

Foreign Exchange Markets

During the year ending May 2001, the dollar continued to strengthen, the euro continued to weaken, and yen strengthening gave way to decline. The 1999 and 2000 International Capital Markets reports highlighted the possibility of misaligned currencies and the risk of significant adjustments in the major currency markets, in part due to growing external imbalances in some of the major industrialized countries, notably the United States and Japan. If anything, the possibility of misalignments increased during the period under review because financial market developments and capital flows—more than macroeconomic fundamentals and policies—may have influenced the values of the U.S. dollar and the euro. After appreciating a cumulative 9 percent in nominal effective terms over the two-year period to May 2000, the dollar appreciated by a further 8 percent during the year ending May 2001 amid strong net international demand for U.S. assets. This occurred despite the sharper slowdown in the United States than in other major countries, considerable monetary easing, broadly stable long-term interest rate differentials vis-à-vis other major countries, and another record current account deficit (Figure 2.4).

Figure 2.4.
Figure 2.4.

Selected Major Industrial Countries: Exchange Rates

(Weekly averages)

Source: IMF.Note: In each panel, the effective and bilateral exchange rates are scaled so that an upward movement implies an appreciation of the respective local currency.1Local currency units per U.S. dollar, except for the euro area and the United Kingdom, for which data are shown as U.S. dollars per local currency.21995 = 100; constructed using 1989–91 trade weights.3Prior to 1999, data refer to synthetic rate.

The euro depreciated by about 3 percent in nominal effective terms and by about 9 percent against the dollar (to $0.84), trading as low as $0.82 notwithstanding stronger growth in the euro area than abroad. Except for brief periods after three official foreign exchange interventions in the autumn of 2000 and a short-lived rebound in December, the euro fell steadily vis-à-vis the dollar. Relative firming of the euro area’s monetary policy stance compared with that of the United States seemed to have little effect on the exchange rate: the euro declined against the dollar as the European Central Bank (ECB) tightened in August and October, and again in the first four months of 2001 as the U.S. Federal Reserve eased key interest rates by 200 basis points. The euro’s depreciation against the dollar may instead have been associated with flows of portfolio capital, particularly equity capital, from the euro area to the United States.7

By contrast, the yen’s decline seemed to reflect Japan’s deteriorating macroeconomic performance and growing concerns about the health of the financial sector. During the year ending May 2001, the Japanese yen depreciated by 6 percent in nominal effective terms, by 9 percent against the dollar (to ¥119), and was roughly unchanged vis-à-vis the euro (at about ¥101). The yen held steady against the dollar during most of 2000, firming slightly after the Bank of Japan (BOJ) ended the zero interest rate policy in August. In the fourth quarter, amid accumulating evidence that Japan’s nascent recovery had stalled, the yen began to weaken sharply against the dollar. The decline continued into 2001 as falling stock prices fed concerns about the stability of Japan’s financial system and the health of its corporate sector and as the BOJ introduced a quantitative monetary policy framework. The yen’s weakening was briefly interrupted by capital reflows in the runup to the March fiscal year-end.

Equity Markets

Last year’s International Capital Markets report highlighted the divergent performance of technology and nontechnology stocks in global markets.8 This divergence was particularly pronounced in the period leading up to March 2000, at which point valuations for technology stocks implied that investors expected future earnings to grow at an exceptionally rapid rate. During the period under review, the global economic slowdown, downward pressure on corporate earnings, and deteriorating investor sentiment resulted in plummeting equity valuations globally through early 2001, with a particularly dramatic sell-off in technology, media, and telecommunications (TMT) stocks (Figure 2.5). Notwithstanding the subsequent rebound in equity markets, compared with their early 2000 peaks, indices of TMT stocks have declined by 50 to 60 percent in the United States, Japan, and Europe. By contrast, indices of nontech stocks are little changed compared with their levels in February/March 2000 and are still about 30 percent above their levels in autumn 1998, prior to the recent run-up in global stock markets. Broad market indices that include companies from a variety of sectors have declined by about 10–20 percent; indices that have larger weights on technology stocks have registered larger declines.

Figure 2.5.
Figure 2.5.

Equity Indices: Technology Sector vs. Nontechnology Sector

(January 1, 1999 = 100)

Source: Primark Datastream.

Since mid-1998, TMT stocks, as well as the broader indices, have become highly correlated.9 For example, the correlation between U.S. and European share prices increased from 0.4 in the mid-1990s to 0.8 in 2000. Reflecting increased correlation, broad indices in the United States, Japan, and Europe fell from their peaks by surprisingly similar amounts (Table 2.2). This increase in correlation has coincided with a number of important structural trends. First, financial and economic globalization and the worldwide information technology boom appear to have increased the importance of global industry factors in determining equity prices.10 For instance, with the introduction of the euro and the elimination of foreign exchange risk in the euro area, international equity portfolios are increasingly managed from a sectoral rather than geographic perspective. In addition, an increasing number of companies are listed on more than one national exchange, and the exchanges themselves increasingly have overlapping trading hours. Second, portfolio managers tend to rely on similar portfolio- and risk-management rules and models. When equity volatility rises, as it has from time to time in the last year, a variety of such models and rules may signal portfolio managers to reduce overall equity exposure by selling shares in many national markets simultaneously.11

Table 2.2.

Equity Price Changes

(In percent)

article image
Source: Primark Datastream (Datastream proprietary indices).

During October 1998–February/March 2000 (peaks vary by country).

During February/March 2000–April 2001 (peaks vary by country).

Broad price movements in the major equity markets reflected this increased correlation, as the major equity markets generally peaked together in the second and third quarters of 2000. U.S. equity prices stabilized after the spring sell-off in technology stocks, and broad market indices recovered some losses as interest rates declined and earnings growth strengthened. Stock prices resumed their decline in September, led by the technology sector, amid profit warnings, ratings downgrades, and reduced short-term earnings forecasts (although long-run earnings forecasts remained at high levels—see Figure 2.6). In January 2001, U.S. equity prices recovered following monetary easing by the Federal Reserve, but the rally was cut short by corporate earnings disappointments in February and March. Tech stocks declined, then broader indices followed amid growing concerns about the U.S. economic outlook. Equity prices rebounded again in April after an inter-meeting cut in interest rates, but failed to recover the losses sustained since mid-2000. At end-May 2001, the S&P500 and Nasdaq indices stood about 12 percent and 40 percent, respectively, below the levels they had attained a year earlier.

Figure 2.6.
Figure 2.6.

S&P 500 Earnings Outlook

(In percent)

Source: IBES International.

By May 2001, broad indices of European share prices had declined by 15 percent from their September 2000 peaks.12 This drop occurred roughly in line with the U.S. markets, notwithstanding the relatively favorable economic outlook in Europe compared with the United States. Linkages of major European corporations to the U.S. economy may explain some of this co-movement: many large European companies are heavily exposed to U.S. demand conditions, particularly through their U.S.-based affiliates.

Japanese equity prices also fell substantially in response to deteriorating macroeconomic fundamentals, growing concern about the weak financial and corporate sectors, and technical factors. Starting in the second half of 2000, the outlook for corporate profits dimmed as the recovery stalled, adding to concerns about the highly leveraged corporate sector. The possibility that banks might unwind cross-shareholdings in the run-up to the introduction of mark-to-market accounting may also have weighed on share prices, along with sales by foreign investors who (despite their still-underweight positions) reduced their allocations to the Japanese equity market after raising them in the first half of 2000. Between mid-2000 and early March 2001, the relatively tech-heavy Nikkei index dropped by about 30 percent to around 12,000, while the broader Topix declined by about 25 percent. Concerns about the health of Japan’s banks, which have significant exposures to the equity market, grew as the Nikkei declined to post-bubble lows in March 2001.13 That month, the government announced an emergency package that included a fund to buy equity holdings from banks, the BOJ switched to a quantitative monetary policy framework, and a new prime minister assumed office with widespread popular support. Stocks rallied, and by end-May the Nikkei was up about 12 percent from its March lows.

Although the overall decline in equity prices has probably brought equity valuations more in line with company fundamentals, broad-market valuation indicators suggest that stock price indices in some countries and segments are still on the high side of historical experience. Notwithstanding the dramatic decline in prices, price-earnings ratios in the technology sector remain at high levels compared with past years, while in the United States price-earnings ratios for non-technology stocks also appear somewhat elevated. At the same time, declining long-term interest rates have lent support to equity market valuations. An assessment of expected real earnings growth implied by price-earnings ratios and long-term interest rates finds that such expectations are somewhat above their long-run averages (see Box 2.1).

Price-Earnings Ratios and Implied Real Earnings Growth in Major Stock Markets

This box examines the expected real earnings growth rates that are implied by the current levels of price-earnings ratios and long-term real interest rates in the United States, the European Union, and Japan, with a view to assessing whether stock prices currently reflect realistic expectations about earnings prospects.

Under the assumption that the dividend payout ratio is constant and equal to one, the current equity price, Po can be expressed as the discounted present value of future earnings expected at time t, El+ie(i1)

article image
,1

Pt=i=1Et+ie(1+ρt)i,(1)

where ρt denotes the expected real cost of equity capital. If future earnings are expected to grow at a constant real rate gte, equation (1) becomes

PtEt=1+gteρtgte.(2)

Based on equation (2), the implied real earnings growth rate, gte, can be estimated by setting the cost of capital equal to the sum of the real 10-year government bond yield, rp and the equity premium, e, assumed to equal 6 percent.2

The implied rate of earnings growth in the technology and non technology sectors in Europe, Japan, and the United States peaked in tandem with stock prices in the first quarter of 2000 (see the figures). Current price-earnings ratios are still somewhat higher than their long-run averages in most sectors and countries (see the table). Similarly, implied earnings growth rates are generally somewhat above the levels consistent with historical price-earnings ratios and current interest rates. These facts suggest that stock market valuations may still be at somewhat high levels compared with the past, particularly in view of the present low level of real interest rates.

uch02fig01

Implied Real Earnings Growth

(In percent)

Sources: Primark Datastream; and IMF staff calculations.
article image
Sources: Primark Datastream; and IMF staff calculations.

The price-earnings (PE) ratio is the historical average for January 1984–April 2001. The real implied earnings growth is calculated using the historical PE ratio and the April 2001 real interest rate.

uch02fig02

Selected Major Industrial Countries: Price-Earnings Ratios

Source: Primark Datastream.
1

Sec, for example, Gordon (1962) and IMF (1998b, 1999). The calculations are based on earnings instead of dividends owing to data limitations. In addition, dividends are affected by corporate financial policy and therefore may be less reliable indicators of future cash flows than earnings.

2

This assumption is consistent with the results of Mehra and Prescott (1985) and Campbell, Lo, and MacKinlay (1997). The assumed level of the equity premium does not affect intertemporal and crosscountry assessments as long as equity premia are constant over time and across countries. For instance, employing a larger (or smaller) constant equity premium would shift the implied earnings growth rates up (or down) by the same percentage amount. The analysis is based on proprietary price-earnings series compiled by Primark Datastream. Real bond yields were calculated by subtracting consumer price inflation from nominal yields. The German 10-year bund yield adjusted for consumer price inflation is used as a proxy for real bond yields in the European Union.

National and International Credit Markets

Credit Markets in the United States, Europe and Japan

During the year to May 2001, credit was repriced in the mature markets as the deteriorating global economic situation put increasing pressure on corporate earnings and heightened concerns about credit risk in the United States and, to a lesser extent, in Europe. Notwithstanding a decline in interest rates as monetary policy tightening gave way to actual and expected future easing (Figures 2.72.8), corporate borrowers came under increasing strain and default rates began to rise. Investor concerns about credit risk mounted, culminating in December in a repricing in the credit-sensitive high-yield market, where issuance dried up and spreads rose to recession levels (Figure 2.9). Credit markets revived in January 2001 after a surprise cut in interest rates by the Federal Reserve, although spreads remained at higher levels than before the turbulence.

Figure 2.7.
Figure 2.7.

Short- and Long-Term Interest Rates

(In percent)

Source: Bloomberg Financial Markets L.P.1For United States, Japan, and United Kingdom, three-month LIBOR; and for euro area, three-month EURIBOR.2Ten-year government bond yields.3Spread between yield on 10-year government bonds and three-month LIBOR or EURIBOR.4Spread between yield on 10-year government bonds and yield on 10-year U.S. government bond.
Figure 2.8.
Figure 2.8.

Monetary Policy Rates and Short-Term Rate Expectations

(In percent)

Source: Bloomberg Financial Markets L.P.
Figure 2.9.
Figure 2.9.

United States: Corporate Bond Market

Sources: U.S. Board of Governors of the Federal Reserve System; and Bloomberg Financial Markets L.P.1Spreads against yields on 30-year U.S. government bonds.

Starting in the second half of 2000, concerns about the credit risk of some U.S. and European companies began to grow following realizations that telecoms companies—which had relied heavily on the debt markets for financing—may have become overextended.14 As the year wore on, the deteriorating global economic outlook raised concerns about heavily indebted corporations in other sectors. These concerns were fed by episodes of financial stress at high-profile firms—as reflected, for example, in Standard & Poor’s downgrade of Xerox’s rating to sub-investment grade. Other financial indicators supported the impression of deteriorating credit quality: the default rate on high-yield bonds reached its highest level since 1991 and the delinquency rate on business loans at commercial banks continued to rise.15 U.S. corporate leverage rose, and reached all-time highs relative to net assets, partly reflecting sizable equity buybacks in recent years. Household leverage rose as well: debt service payments as a percent of disposable income increased slightly to 13.7 percent, the highest level since the mid-1980s.

Mounting signs of deteriorating credit quality had the most pronounced impact on conditions in the high-yield market. High-yield investors showed heightened preference for liquidity, as spreads on medium- and smaller-sized high-yield issues rose particularly sharply, although treasury and investment-grade corporate bond markets remained liquid. Credit concerns also adversely affected lower-rated (but investment-grade) borrowers in the commercial paper (CP) market, who were reportedly shut out of the market and instead drew on bank credit lines and backup facilities or issued longer-term debt (Figure 2.10). Long-term credit spreads rose much more sharply than short-term spreads, however, perhaps due to concerns that credit quality might deteriorate further. By contrast with the high-yield market, pricing and issuance in the investment-grade bond markets were little affected, notwithstanding downward pressure on the profits of large U.S. companies.16

Figure 2.10.
Figure 2.10.

Selected Spreads

(In basis points)

Sources: Bloomberg Financial Markets L.P.; and Merrill Lynch.1Spread over 30-year U.S. treasury bond; weekly data.2Spread of a 30-year off-the-run treasury bond over a 30-year on-the-run treasury bond.3Spread of fixed-rate leg of 10-year interest rate swaps over 10-year government bond.4Spread between yields on three-month U.S. treasury repos and on three-month U.S. treasury bill.5Spread between yields on 90-day investment grade commercial paper and on three-month U.S. treasury bill.

After the U.S. Federal Reserve cut interest rates in January 2001, spreads declined and issuance resumed in the high-yield market. High-yield bond issues rebounded strongly in the first quarter—particularly in the U.S. domestic market—as further rate cuts were put in place, and investment-grade issuance set new records, although spreads remained at high levels.

Meanwhile, some highly leveraged telecoms companies reduced their debt burdens by selling assets, refinancing, and securitizing cash flows—a process that was still in train at mid-2001.17 Flows recovered and spreads narrowed in the CP market as well. With the benefit of hindsight, the: sharp widening in spreads for high-yield issues could be seen as reflecting a reappraisal of credit risk and reallocation of credit after a long-running boom, rather than the indiscriminate shedding of risk and cutting-off of credit flows that characterize a credit crunch. Instead, the tiering of credit spreads—that is, greater discrimination between spreads for investment-grade and sub-investment-grade borrowers—suggests increased credit market discipline.

The tiering in U.S. credit markets was broadly reflected in Europe’s nascent corporate bond market, which nevertheless continued to grow—probably because it mostly comprised investment-grade borrowers (Figure 2.11). The introduction of the euro in 1999 expanded the investor base for European corporate debt issuers and prompted a surge in euro-denominated corporate debt issuance, contributing to a further broadening and integration of Europe’s capital markets. To establish a reputation in the burgeoning market, some firms reportedly placed issues even in the absence of major financing needs. Issue sizes rose as investors sought liquidity and the credit spectrum broadened to lower-rated issues as investors increased their appetite for credit risk. Notwithstanding the recent boom, bonds still account for a small share of outstanding corporate liabilities in the euro area compared with loans. This is partly explained by the reluctance of some corporations to open their books to rating agencies and the continued support of house banks.

Figure 2.11.
Figure 2.11.

Nonfinancial Corporate Credit Spreads

(In basis points)

Source: Merrill Lynch.

Credit spreads in the Japanese corporate bond market have been highly compressed despite deteriorating economic fundamentals. For example, spreads on bonds issued by BBB-rated Japanese corporations are under 50 basis points compared with about 200 basis points for bonds issued by similarly rated U.S. corporations. Technical factors may explain this anomaly: Japanese institutional investors such as insurance companies, mutual funds, pension funds, and trust banks have been heavy buyers of Japanese corporate bonds amid shrinking corporate issuance and a rapidly growing supply of JGBs. Moreover, Japanese corporate credit spreads may not fully reflect intrinsic credit risk. In Japan, creditors are generally reluctant to pursue bankruptcy proceedings even for technically insolvent corporations. Finally, the comparatively narrow spreads also may be partly attributable to efforts by the authorities to promote corporate lending through loan guarantees.

International Securities and Syndicated Loan Markets

The reassessment of credit risks—particularly those associated with the telecoms sector—prompted shifts in issuance activity in the international securities markets (Table 2.3). In the first half of 2000, brisk issuance was supported by a huge volume of telecoms issues.18 In the second half of the year, as investors reevaluated credit risk in the light of concerns about telecoms firms’ rising leverage, credit spreads widened and telecoms issues fell sharply. Issuance of long-term securities by low-rated borrowers slowed particularly markedly and the pattern of financing shifted from longer-term instruments toward money market instruments (net issues of which almost doubled). This shift was reversed in the first quarter of 2001, as borrowers took advantage of improved conditions in the international bond market to lengthen maturities. Telecoms borrowers took part in the rebound, as their issuance of long-term international securities rose to a record of nearly $50 billion. Average deal sizes in the dollar and euro segments rose significantly in 2000, reflecting issuers’ increasing needs for capital, a broadening global investor base, and demand for secondary market liquidity. Following the 1999 surge of euro-denominated issues, issuers increasingly favored the U.S. dollar.

Table 2.3.

Outstanding Amounts and Net Issues of International Debt Securities by Currency of Issue1

(In billions of U.S. dollars)

article image
Source: BIS, International Banking and Financial Market Developments (various issues).

Excludes money mantel instalments.

Prior to 1999, the underlying data refer to the European Currency unit (ECU); from 1999 onward, the underlying data refer to the euro.

Activity in the international syndicated loan market surged by more than 40 percent in 2000, boosted by strong M&A activity and heavy loan demand by telecoms firms (Table 2.4). Announced M&A-related transactions rose by 22 percent to $214 billion and syndicated credits to telecoms tripled to $256 billion. The surge in borrowing by European telecoms during the second half of the year mainly took the form of bridge loans as firms delayed tapping the securities markets. Syndicated lending slowed sharply in the first quarter of 2001, as borrowing by telecoms fell and banks tightened conditions for supplying backup and standby facilities. In recent years, deals have become larger and have increasingly taken the form of bridge loans and standby facilities; average maturities have fallen from six years in 1992 to about three years in 2000.

Table 2.4.

Announced International Syndicated Credit Facilities by Nationality of Borrowers

(In billions of U.S. dollars)

article image
Source: BIS, International Banking and Financial Market Developments (various issues).

Derivatives Markets

Outstanding notional amounts in global over-the-counter (OTC) derivatives markets continued to grow, while outstandings on organized exchanges declined (Tables 2.52.7). In the 18 months to December 2000, notional principal in global OTC derivatives markets grew by about 17 percent to $95 trillion, while notional principal on organized exchanges declined by about 8 percent to $14 trillion.19 (However, notional principal rebounded strongly in the first quarter of 2001.) Gross market values rose in most segments of the OTC derivatives markets, accompanying the increase in notional principal. Growth in OTC derivatives activity was mainly attributable to the continued rapid expansion of interest rate contracts, which reflected growing corporate bond markets, increased reliance on swaps for hedging, increased interest rate uncertainty around the end of 2000 (as reflected in option implied volatilities), and (in Japan) increased yield volatility and convexity risk.20 Foreign exchange contracts increased moderately, although contracts involving the euro rebounded by 28 percent in 2000 along with rising issuance of euro-denominated securities.

Table 2.5.

Exchange-Traded Derivatives: Notional Principal Amounts Outstanding and Annual Turnover

article image
Source: BIS, International Banking and Financial Market Developments (various issues).
Table 2.6.

Global Over-the-Counter Derivatives Markets: Notional Amounts and Gross Market Values of Outstanding Contracts1

(In billions of U.S. dollars)

article image
Source: BIS (2001a, 2001b).

All figures are adjusted for double-counting. Notional amounts outstanding have been adjusted by halving positions vis-à-vis other reporting dealers. Gross market values are calculated as the sum of the total gross positive market value of contracts and the absolute value of the gross negative market value of contracts with nonreporting counterparties.

Single-currency contracts only.

Adjustments for double-counting are estimated.

Estimated positions of nonreporting institutions.

Gross market values adjusted for legally enforceable bilateral netting agreements.

Table 2.7.

Global Over-the-Counter Derivatives Markets: Notional Amounts and Gross Market Values of Outstanding Contracts by Counterparty, Remaining Maturity, and Currency Composition1

(In billions of U.S. dollars)

article image
Sources: BIS (2001a, 2001b).

AII figures are adjusted for double-counting. Notional amounts outstanding have been adjusted by halving positions vis-à-vis other reporting dealers. Gross market values are calculated as the sum of the total gross positive market value of contracts and the absolute value of the gross negative market value of contracts with nonreporting counterparties.

Residual maturity.

As both currency sides of each foreign exchange transaction are counted, the currency breakdown sums to twice the aggregate. Data for euro before end-June 1999 refer to legacy currencies of the euro.

Single-currency contracts only.

Data before end-June 1999 refer to legacy currencies of the euro.

Adjustments for double-counting are estimated.

Estimated positions of nonreporting institutions.

In Europe, the introduction of the euro and a shift of organized exchange trading of long-term interest rate derivatives from London to Frankfurt have brought increased reliance on the German government securities market for hedging interest rate risk. The 1998 merger of Deutsche Termin Börse and the Swiss Options and Financial Futures Exchange into Eurex preceded the creation of Europe’s most actively traded interest rate contracts—the euro-bund and euro-bobl futures—and made Eurex the world’s largest derivatives exchange. Trading of the futures contracts has increased to the point where open interest often exceeds the amount of outstanding underlying bonds, which, along with shortcomings in the repurchase agreement (repo) market, provides conditions conducive to squeezes. For example, prior to the expiration of the March 2001 bobl contract, open interest was almost twice as large as the deliverable basket of underlying securities, which led to a significant squeeze (see Box 4.5 in Chapter IV).21

Recent fluctuations in a gross leverage indicator for the top 25 U.S. commercial banks are consistent with anecdotal evidence about leveraging activities of market participants in the derivatives markets (Figure 2.12).22 For example, the indicator tracks the buildup of leverage in early 1998 and the reduction in leverage after the near-collapse of the major hedge fund Long-Term Capital Management (LTCM) in October 1998.23 According to the indicator, leverage rose again in the fourth quarter of 2000.

Figure 2.12.
Figure 2.12.

United States: Banks’ Total Gross and On-Balance-Sheet Leverage Ratios

Sources: United States, Office of the Comptroller of the Currency; and IMF staff calculations.1Total gross leverage equals assets plus notional amounts outstanding as a percentage of regulatory capital. Regulatory capital refers to Tier 1 plus Tier 2 capital.

Although small compared to the more mature OTC derivatives segments, the market for credit derivatives is growing rapidly. A recent survey by the British Bankers Association estimates that in 2000 the global credit derivatives market grew by 50 percent to $893 billion, representing a fivefold increase since 1997. Recent credit strains in the telecoms sector reportedly boosted the demand for instruments to hedge credit risk and contributed to this growth.24 The market remains geographically concentrated—nearly half of all transactions originate in London—but products are becoming more diverse as the share of the most liquid, actively traded “plain vanilla” instruments, such as credit default products and total return swaps, declines. In addition, participation in the market is broadening, improving its depth but also raising questions about whether new participants fully understand the attendant risks. Insurance companies are now the largest net sellers of credit protection, accounting for 23 percent of sales and 7 percent of purchases. Supervisors, regulators, and some market participants welcome the increased potential to distribute credit risk among a larger set of institutions, but nonetheless remain concerned about the lack of transparency and legal and other operational risks. Advances in documentation, such as the 1999 International Swaps and Derivatives Association (ISDA) standard credit derivatives confirmation, may alleviate operational risks. However, a variety of other issues, such as a standard definition of the event of default, still needs to be resolved.

The hedge fund industry—which includes important participants in the OTC derivatives markets—appears to have experienced some noteworthy changes during the past two years. Partly as a result of these changes, activities in OTC derivatives markets and some of the underlying markets have become more highly concentrated. Moreover, OTC derivatives and underlying markets are widely seen as less liquid than they were in the mid- to late-1990s.25 This applies to advanced country markets as well as emerging debt and foreign exchange markets. More specifically, five main recent developments are relevant.26 First, hedge funds recorded, on average, modest positive returns on capital under management during the past year, outperforming most of the major market indices. Second, the number of hedge funds (and, to a lesser extent, total capital under management) has rebounded from the contraction during late 1998 and 1999. Growth in the industry is strong, with an acceleration in growth in Europe and, to a lesser extent, Asia. The limited information available on hedge fund activities, including press reports, suggests that the main sources of inflows to hedge funds have been pension funds, insurance companies, and major banking institutions. Investment allocations by these investors still appear to represent a very small percentage of portfolio assets. Third, the average size of hedge funds probably decreased. This was mainly due to the closure of several very large hedge funds in 2000. Closures seem to be based on two considerations: a reassessment of the risk-adjusted expected returns on large directional positions on asset prices; and the perception that increased scrutiny of hedge fund investments would adversely affect potential returns. Fourth, as expected by market observers, market discipline of hedge funds appears to have increased since the near-collapse of LTCM. However, disclosure of investment strategies and positions by hedge funds, even to investors, remains very limited. Fifth, overall leverage within the industry has probably fallen.

Developments in Major Banking Systems

Despite the global slowdown and deteriorating credit quality among corporate borrowers, banks in the United States and the major countries in Europe generally performed solidly in 2000 and appeared to remain in good financial condition as credit expanded.27 Some large banks, especially in Europe, had significant telecoms exposures but, as of May 2001, regulators and the major credit rating agencies saw the banks as adequately managing these exposures. Banking consolidation continued, partly in response to competitive pressures in wholesale finance and asset management. In Japan, banks continued to struggle with low profitability and substantial bad debts. Falling equity prices reinforced concerns about bank capitalization and prompted the authorities to announce measures to address bad debt problems and to reduce bank equity exposures.

U.S. banks appeared to be more robust than prior to previous downturns and seemed well enough capitalized to weather the ongoing deterioration in credit quality. In 2000, as U.S. economic growth slowed, the fraction of non performing commercial and industrial loans rose by 50 percent to 1.7 percent (well below the 4.3 percent recorded in 1991, at the end of the last recession). In response, banks increased provisions and sharply tightened lending standards.28 Lending continued to expand and the growth of consumer lending picked up. U.S. commercial banks’ return on equity (ROE) declined from 15.6 percent to 13.7 percent as growth in non-interest income slowed sharply. Rating agencies and supervisors view the U.S. banking system as financially strong. In 2000, for example, commercial banks had an average risk-based capital ratio of 12 percent and supervisors classified 98 percent of the banks as “well capitalized.”

European banks generally reported higher or stable profits and remained well capitalized. Banks have acquired significant telecoms exposures by financing roughly $400 billion in telecoms-related M&A activity and $125 billion in purchases of “third generation” telecoms licenses. As of end-April 2001, total syndicated loan commitments (drawn and undrawn) of all international banks to European telecoms operators and equipment manufacturers totaled about $180 billion. Official concerns about the size of telecoms exposures were reflected in public statements by regulatory authorities including the U.K. Financial Services Authority (FSA) and the French Banking Commission. Nevertheless, as of mid-2001 rating agencies, supervisors, and bank analysts generally believed that European banks were managing these exposures effectively and did not see them as a major risk for two reasons. First, the bulk of exposures are to highly rated national telecoms companies that have solid fundamentals. Second, banks have been able to reduce and diversify these exposures by securitizing and selling telecoms loans and by encouraging telecoms firms to substitute market funding for loans.

French banks performed well in 2000, as several large banks reaped considerable profits from wholesale finance and asset management. Average ROE for the five largest private banks rose from about 13 percent to 15 percent and was higher for banks with strong wholesale franchises. ROE was also boosted by low provisions, which reflected very low non performing loan (NPL) ratios—a legacy of more cautious lending following France’s real estate crisis in the mid-1990s. At the same time, the French Banking Commission became concerned that banks might be underprovisioning to increase profits. The major French banks were well capitalized, with capital ratios stable in the 11 to 13 percent range. Consolidation among French banks paused, although one large French bank was acquired by a major foreign bank in a friendly takeover.

Top German banks’ average ROE in 2000 was broadly unchanged at about 8 percent after adjusting for one-time capital gains from sales of cross-shareholdings. However, the best-performing banks recorded adjusted ROEs of about 12 percent. Rating agencies and German supervisors regard the German banking system—with few exceptions—as sound. The top four banks had capital ratios of 10–13 percent and NPL ratios of 2–3 percent. Faced with strong competition in retail banking from the many smaller savings banks and cooperative banks, large banks continued to refocus on wholesale finance and asset management. This shift in focus was facilitated by tax law changes that reduced the tax burden of unwinding cross-shareholdings among financial institutions and helped to move the German banking system toward the bancassurance model. One top-four bank merged with the largest German insurance company, and another bank strengthened its ties with an insurance company.

ROE for the six major private banks in Italy rose by over 2 percentage points to about 15 percent in 2000, reflecting strong retail, asset management, and corporate lending franchises. Some individual banks achieved especially strong performances by successfully focusing on developing strong franchises and reducing non-performing loans, including through securitization. Between 1998 and 2000, the six major banks cut NPL ratios nearly in half to 4 percent and raised their capital ratios modestly to around 10 percent.

Spanish banks were strongly profitable in 2000, as ROE for Spanish credit institutions rose from 19 percent to 22 percent. Spanish banks also had robust asset quality. The NPL ratio stood at 1.1 percent at the end of 2000; in addition, provisions covered 166 percent of impaired assets. They were also well capitalized. The overall capital ratio stood at 11.1 percent, of which 9.1 percent corresponded to Tier 1 capital.

Average ROE for the five largest U.K. banks held constant at about 19 percent in 2000, reflecting highly profitable retail banking activities. In addition, loan credit quality remained solid as economic growth picked up, permitting a low level of provisioning. The top banks’ average NPL ratio declined moderately to about 2 percent. U.K. banks remained well capitalized, with the average capital ratio for the five largest banks stable in 2000 at just above 12 percent.

In both the United States and Europe, larger internationally active banks have sought to diversify into higher-margin, fee-generating activities in an effort to raise ROE. At the same time, competition in wholesale finance is eroding margins. Competitive pressures may also have driven banking consolidation as banks have sought to capture economies of scale and scope. The opportunities presented by the rapid expansion of euro-area corporate securities markets have accelerated the shift by euro-area banks into wholesale finance. European corporations still rely much more heavily on loans than on securities for financing needs, however, and euro-area banks that underwrite market financing often also continue to function as the client’s “Hausbank”—for example, by maintaining a long-term lending relationship and providing backup financing. Meanwhile, European banks have relied more heavily on credit-risk transfer vehicles (such as asset-backed securitization and collateralized loan obligations) to manage their credit and balance-sheet risks. The shift from lending to market finance may also have influenced the balance of risks in the U.S. and European financial systems. For example, while the distribution of financial risk may be diversified, it may also be less transparent.

Japanese bank performance remained weak in 2000.29 Based on fiscal year 2000 interim results, ROE for Japanese city banks was broadly unchanged at about 3 percent, reflecting high provisioning and narrow interest margins.30 Profits at regional banks were also adversely affected by provisioning. Major Japanese banks reported an average capital ratio of 12.3 percent in September 2000, but more than half of Tier 1 capital consisted of public capital, deferred taxes, and preferred equity instruments. In addition, some private analysts estimated that uncovered exposure to future losses remained large relative to capital—for major banks, equivalent to roughly half of aggregate Tier 1 capital, and for regional banks, over 100 percent of Tier 1 capital. In March 2001, as the stock market slide heightened market concerns about the stability of the Japanese banking system, the Japanese authorities announced a package of measures intended to stabilize and revive the banking sector. This package included measures to accelerate the disposal of bad loans and to reduce bank equity market exposure through government purchases of bank equity holdings. Market participants viewed the measures as falling short of addressing the banking system’s fundamental weaknesses.

The consolidated cross-border exposures of mature market banks to emerging market countries is large but fairly well-diversified across countries and regions (Table 2.8).31 Banking systems in two countries—Germany and Spain—have significant exposures to emerging market countries currently under stress (Turkey and Argentina, respectively). In each case, however, these exposure concentrations represent small fractions of outstanding loans. The risks to Spanish and German banks may also be mitigated by the fact that about 10 percent of exposures booked in Argentina and Turkey are actually to counterparties headquartered outside these countries. Finally, much of German banks’ Turkish exposure reportedly carries export credit guarantees (shifting the credit risk to the export credit agency and ultimately to the budget).

Table 2.8.

Mature-Market Bank Exposures to Emerging Markets, End-December 2000

(In millions of U.S. dollars)

article image
Sources: BIS (2001a); and Bankscope database.

Austria, Belgium, Finland, France, Germany, Ireland, Italy, the Netherlands, Portugal, and Spain; omits countries for which data are not available.

Sum of developed countries, developing countries, and offshore financial centers.

Emerging markets and developing countries in Europe.

1999 figures.