Emerging Capital Markets
Overview
A year ago few observers expected that by the end of 1995 capital flows to emerging markets would have surpassed their previous peak reached in 1993. The devaluation of the Mexican peso in December 1994 triggered a sharp revision in the expectations of international investors about the prospects in emerging markets, which was followed by a rebalancing of international portfolios. In the first quarter of 1995, portfolio flows to emerging markets declined dramatically, premiums on developing country bonds increased, equity prices in emerging markets fell sharply, and currency pressures were felt, at least temporarily, in a geographically dispersed group of economies ranging from Argentina and Brazil to Hong Kong and South Africa. The period of time over which exchange rates in particular countries or regions were under pressure ranged from a few days in Hong Kong to several months in Argentina. Some emerging market countries responded successfully to the changed investment climate with economic reforms and defensive policy measures. In addition, the concerted liquidity support, extended during the crisis in the beginning of 1995 by the international community, appears to have been successful, By mid-1995, stock prices in most Asian and East European emerging markets had recovered to their precrisis levels, while equity markets in Latin America took a longer time to regain lost ground, and issuance activity in the international equity markets and in the dollar-denominated bond markets by Latin American countries likewise recovered only in late 1995. Although the swift reactions of investors demonstrated that international capital markets have become more integrated and more able to punish countries with weak fundamentals, capital flows and asset-price movements in the weeks following the Mexican devaluation appear in many cases to have been an overreaction.
Concerns that the Mexican macroeconomic recovery might not be as strong as previously expected started a decline in equity prices, and to a lesser extent in bond prices, in several emerging markets in October and November 1995. The Mexican peso de-preciated some 22 percent between the end of September and November 11, 1995, and Mexican equity prices dropped sharply. The decline in asset prices observed in some of the Asian emerging markets during the same period was also related to the rise in the value of the dollar, which weakened the competitiveness of a number of Asian developing countries that peg their currencies to the dollar. However, this market reaction was short-lived. Most of the major market indicators in the emerging markets recovered by the end of 1995 and early 1996, including local equity price indices, new international bond issues, international syndicated loans, and currency values. The maturity structure and premiums of emerging market bonds also improved, although some Latin American countries still have to pay spreads slightly higher than precrisis levels. In the first quarter of 1996, capital flows to emerging markets were buoyant in all regions and in all asset classes. The generally good news regarding international capital flows was dampened by the realization that the health of the banking systems in some of the emerging market countries had become a sufficiently serious problem to impede macroeconomic performance.
Capital Flows
Despite the Mexican crisis, capital flows surged to a record $228 billion in the developing countries ($194 billion) and countries in transition ($34 billion) in 1995 (see Table 1). Overall, net inflows to the developing countries and countries in transition were bolstered by a strong performance by Asian emerging markets, liquid bank-loan markets, and strong demand for emerging market bonds (especially Latin American and East European issues) by Japanese and German investors. The pace of net capital flows into emerging markets was sustained in early 1996. The renewed interest in emerging market investments was associated with greater regional and asset-class differentiation than had been evident before the crisis. First, capita] markets in the economies in transition received a greater share of capital inflows: net flows to these countries increased more than 150 percent in 1995. making the economies in transition the third most important destination for foreign investment. By contrast, capital flows to Latin America in 1995 would have remained essentially unchanged from the 1994 level, which had already declined below the 1992-93 levels, had it not been for net inflows of IMF credits of about $13 billion. Second, while total net investment rose in 1995, net portfolio investment to the developing countries declined by 27 percent in 1995, as investors became more cautious about the higher-risk markets. This change in investor sentiment has meant that the Latin American countries, which are heavily dependent on portfolio flows, have experienced sharp adjustments and are still showing lingering effects of the crisis. Indeed, net portfolio flows into Latin America declined by 40 percent in 1995. Third, foreign direct investment increased in most regions, especially in Asia and in the countries in transition.
Major institutional investors are differentiating more between countries in two important ways. First, they are relying on a broader array of macroeconomic and financial indicators in assessing the sustainability of internal and external macroeconomic balances and investment opportunities. Second, they are paying considerably more attention to the safety and soundness of banking systems, the integrity of the financial data, their ability to gain legal recourse, and to the efficiency and safety of the financial infrastructure. Hence countries that are perceived to be pulling in place efficient financial market infrastructures, including a supervisory and regulatory regime that meets international standards, are likely to capture a greater and more durable share of global flows.
All in all, a large part of the investor community appears to be more risk-aware, more likely to do their homework in assessing the economic fundamentals, and more likely to respond to good data and equal treatment. However, the process of learning to assess on a continuous and timely basis the risk of various investments in emerging market countries has just begun, and there is a long way to go before the availability and quality of macroeconomic and corporate data and the supporting analysis compare favorably with that of the major international financial markets. In this regard, the IMF’s new standards for improving the quality and dissemination of economic and financial data were thought to be particularly helpful.5
International Bund Issues
After losing access to the international bond markets in the first quarter of 1995, many emerging market countries were welcomed back into international markets during the second and subsequent quarters (Chart 1). By the end of 1995, total bond issuance by all developing countries had surpassed the level reached in 1994. There was strong demand for Latin American issues from retail investors in Japan and Germany, who were attracted by the relatively high nominal returns on these issues. As a result, Latin American countries managed to increase international borrowing by nearly one third in 1995. Mexico, for example, returned to the sovereign bond market in July 1995 (both in dollars and in yen) and by the end of 1995. Mexican entities had launched new international issues worth $7,6 billion. Although borrowing terms have improved and maturities have lengthened since the worst of the crisis, terms are not yet back to precrisis conditions for all countries in the region. Latin American Brady issues also recovered their losses from the Mexican crisis by May 1995. but experienced a setback in October-November 1995. when concerns emerged about the strength of Mexico’s recovery. At the end of April 1996. spreads on Latin Brady bonds were still slightly above the precrisis levels.
Private Market Financing for Developing Countries
(In billions of U.S. dollars)
Source: IMF staff estimates based on Euromoney Bondware and Loanware.1 Preliminary data for May 1996.A noteworthy development in international bond markets was the shift in currency denomination of developing country issues from the dollar to the yen and deutsche mark. In 1995, 26 percent of new bond issues by developing countries were issued in yen, amounting to $15.3 billion, compared with 13 percent in 1994, while $6 billion or 10 percent were issued in deutsche mark, compared with 3 percent in) 1994. Yen-denominated bonds issued by developing countries—Euroyen issues or Samurai bonds issued in Japan—have become increasingly popular instruments for raising funds. One reason for the attractiveness of yen issues is that Japanese investors are willing to accept higher credit risk in return for yields that are much higher than the record low yields offered on domestic instruments, while the issuers are willing to accept the higher currency risk in order to be able to benefit from the immediate positive fiscal impact of lower nominal debt-servicing costs. In total, Argentina and Mexico issued the equivalent of $1.7 billion and $1.6 billion, respectively, of yen-denominated bonds in 1995. Issuers of yen- and deutsche mark-denominated bonds in 1996 included Argentina, Brazil (sovereign), and Colombia. Among European economies, Hungary and Turkey issued the equivalent of $2.6 billion and $1.3 billion in yen-denominated bonds, respectively. Romania issued its first-ever sovereign international bond in May 1996, also denominated in yen.
Retail investors in Japan and Germany have shifted currency risk to borrowers in 1995. after having earlier been disappointed by returns (in domestic currency terms) on foreign dollar-denominated or domestic fixed-income assets. Market participants suggested that retail investors—in contrast to institutional investors—might be underestimating the country risk inherent in these instruments. Likewise, the issuers, who could hedge the currency risk by swapping into currencies for which they own assets, often have not done so.6 Hence, there is concern that the attraction of low nominal interest rates on yen-denominated bonds may have led borrowers to underestimate the currency risk they have incurred.
The risks inherent in foreign-currency-denominated debt were most acutely demonstrated by the Mexican financial crisis, as foreign investors became concerned about the possibility of a default on Mexican tesobonos (U.S. dollar-indexed bonds) that were maturing during the first quarter of 1995. Such concerns are widely acknowledged as one of the more important elements of the crisis that demanded immediate attention. More generally, a currency depreciation causes the value of a country’s foreign-currency-denominated liabilities to increase and, therefore, poses an additional burden on the country’s external position. Moreover, when the stock of maturing foreign-currency-denominated bonds becomes large relative to the size of exports or foreign currency reserves, investors are likely to become concerned about liquidity. This affects the pricing of bond issues, and, in extreme circumstances, can also affect the country’s ability to access international capital markets.
Syndicated Bank Loans
Faced with the relatively high cost of funding in the international and domestic equity and bond markets in the wake of the Mexican crisis, developing countries, especially in Latin America, turned to the syndicated bank loan market as a source of funds. A high level of liquidity in international banking markets facilitated the relatively strong growth in syndicated lending to the developing and transition countries. Fierce competition among lenders resulted in favorable terms for borrowers. In addition, the major international banks in the syndicated loan market—having overcome their 1982 loan losses—increasingly see the emerging markets as offering a reasonable risk-return profile, as well as an opportunity to renew longer-term relationships. Overall, medium- and long-term syndicated loan commitments to all developing countries increased by about 35 percent in 1995 to about $75 billion. Asian countries continued to be the dominant borrowers, accounting for more than half of the total loans committed to developing countries, reflecting the increasing popularity in many regions of innovative approaches to lending for infrastructure projects. The developing countries in Europe and in the Western Hemisphere each accounted for about 13 percent of total loan commitments. China was the largest borrower, receiving about 14 percent of loan commitments. Other relatively large borrowers included Indonesia. Thailand. Turkey, and Mexico. These developments suggest that syndicated bank loans can be expected to make up a larger share of net capital flows to the developing countries and countries in transition in the future.
Equity Markets
In the period under review, international and domestic emerging equity markets did not fare as well as bond markets. During the first quarter of 1995, the total amount of new emerging market equity issues (that is, international issues) dropped to $0.6 billion from an average quarterly issuance of $4.5 billion in 1994 (see Chart 1). For the year as a whole, international equity issues by developing countries declined by nearly 40 percent, the developing country share in total international equity placements fell by 12 percentage points to 24 percent, and issues of depository receipts by all developing countries dropped by more than 60 percent. Latin American countries were again most severely affected and their issues fell by almost 80 percent in 1995 compared with 1994. Emerging market countries in Asia recovered more quickly than did countries in Latin America. Domestic equity markets were also adversely affected and equity prices weakened substantially in early 1995, although in most cases these losses were soon recovered (Chart 2). At least, two factors appeared to weigh on the emerging equity markets for most of 1995. First, the continuing record performance in the U.S. stock markets attracted the attention of equity investors at a time when emerging equity markets were struggling to recover from the Mexican crisis, and U.S. investors abandoned Latin American equity markets almost entirely, while reducing their net purchases of equity issued by Asian emerging market countries. Second, Japanese corporations and financial institutions withdrew almost completely from all international equity markets.
Share Price Indices for Selected Developing Countries
(IFC weekly investable price indices, December 16, 1994 = 100)
Source: International Finance Corporation, Emerging Markets Database.However, as of mid-1996 the worst appears to be over. Equity markets in all regions have seen gains in early 1996. and indicators for some countries suggest that equity prices may surpass the peaks reached in 1993-94. The postcrisis debut by Mexico in international equity markets in February 1996 is a clear indication of the more favorable environment for countries interested in raising funds via the issue of equity in international markets. Moreover, there has been a resurgence in equity portfolio flows from the United States, driven by a net inflow into U.S. global and international mutual funds. Private sector forecasts suggest that equity flows to both Asia and Latin America are likely to rise substantially in the future.
Developments in Selected Banking Systems in Emerging Market Countries
The crisis in emerging markets in 1995 illustrated vividly that countries with weak and inefficient banking systems are more vulnerable to contagion from financial turbulence elsewhere and are less able to manage the negative consequences of volatile capital flows and exchange rate pressures. Some of the problems experienced recently have included the widespread incidence of large-scale poor asset quality; outflows of deposits from the banking systems, or from smaller to larger banks within banking systems: banks’ vulnerability to loss of public confidence; inefficient and costly bank operations; lack of adequate controls of overseas expansions by banks: overexposure to movements in exchange rates: overregulation: unregulated, closely connected financial entities; excessive lending to closely connected commercial companies; and the availability and quality of supervisory staff. Countries with vulnerable banking systems—for example, those with poor asset quality, large loan concentrations in cyclically sensitive sectors, large foreign exchange and interest rate exposures (both on- and off-balance-sheet)—were forced to continue to deal with these banking problems in the second half of 1995 and well into 1996, at times incurring significant fiscal and quasi-fiscal costs. Furthermore, as already noted, the soundness of banking systems is increasingly viewed by international investors as one of the economic fundamentals, and as such it is likely to influence the destination of capital flows. The increased incidence of these problems argues for strengthened domestic prudential supervision and further improvements in financial infrastructures. Finally, improvements in the management of risk and in prudential supervision of banking can also, in the longer run, make an important contribution to the economic performance of developing countries. This section briefly reviews developments in banking systems in selected countries in the Western Hemisphere (Argentina, Brazil. Chile, Mexico, and Venezuela). Asia (India, Indonesia, and Thailand), and Europe (the Czech Republic. Hungary, the Baltic countries, and Turkey).6
The decisive measures taken in early 1995, supported by external funding provided by the IMF and the World Bank, allowed Argentina to weather a large outflow of bank deposits {18 percent of total deposits) without a casualty among the major banks. A high level of minimum reserve requirements proved helpful in providing a liquidity cushion during the deposit runoff and helped to set the stage for the long overdue restructuring of the banking sector. Two fiduciary funds created during the crisis helped the process of consolidation in 1995. In March 1996, there were 153 institutions in the Argentine financial system, down from 205 at the end of 1994. Two thirds of this consolidation entailed mergers and acquisitions of small private cooperative banks, but the number of wholesale banks and nonbank financial institutions also declined. Among the 33 public-sector-owned banks, 2 provincial banks were liquidated. 9 were privatized, and 9 were in the final stages of privatization.
Overall performance of the banking sector has improved, and the systemwide ratio of nonperforming loans net of provisions to total credit appears to have stabilized at around 10 percent since September 1995. In addition, lower operational and funding costs and the shift in August 1995 to a system of remunerated liquidity requirements allowed for a marked turnaround in banks’ net income, and the financial system—which experienced record losses in the first half of the year—recorded profits of more than $82 million ($136 million for private banks) in the second semester. The central bank has also sought to strengthen the system with the introduction of a limited, fully funded, obligatory deposit insurance scheme. It has also replaced the nonremunerated system of reserve requirements with a remunerated “liquidity reserve” system. However, according to market participants (banks and major multinational firms), the Argentine banking system remains relatively inefficient in delivering services to end users, and most wholesale finance related to liquidity and risk management is conducted through foreign financial institutions.
The Brazilian banking system, the largest in Latin America, has been under pressure as a result of the ongoing stabilization of the macroeconomy. The sharp decline in the inflation tax, combined with the high real interest rates, high reserve requirements, and a recession in mid-1995 led to a change in the banking environment to which some banks have had difficulties adapting. Since June 1994, the central hank has liquidated 19 private banks and has intervened in the activities of 4 private banks, including 2 large banks. In one of the banks, the central bank discovered irregularities that had gone undetected for several years, illustrating the difficulties faced by supervisors in obtaining a clear picture of the size of the underpeforming loans in the banking system. In the meantime, measures have been adopted to avoid a repetition. The central bank also placed five state banks under temporary administration, including the two largest. The financial situation of the state banks had been deteriorating for several years because of the failure of state governments to adequately service their debt to these banks. Although the banking system has been under pressure, it appears to contain a core segment, consisting of large and medium-sized private banks, that is essentially sound, well capitalized, and conservatively operated, which is demonstrated by its ability to respond quickly to serious pressures since the inception of the Real Plan. The ratio of loans in arrears and liquidation to total loans increased from 6.6 percent in December 1993 to 12.1 percent in December 1995. A new private deposit guarantee fund was created in November 1995. At the same time, the central bank introduced a program which is intended to promote consolidation among financial institutions, and it was also given greater powers to intervene in banks that are experiencing difficulties.
In contrast to other banking systems in Latin America, the Chilean banking sector prospered in 1995. Bank deposits and credit activity both expanded significantly in real terms during the year, with loan demand being particularly buoyant in the middle-market and consumer sectors. Banks’ profitability and asset quality remained high, return on equity increased strongly, and the ratio of past due loans to total loans dropped below 1 percent. The Chilean banking system was insulated from the regional crisis because of conservative lending practices, effective supervisory oversight, a stringent regulatory environment, and strong macroeconomic performance. The passage of a new law in 1995 has provided a major impetus to the resolution to Chile’s subordinated debt problem. Under the new law, banks may repay their remaining obligations to the central bank in equal yearly installments over the next 40 years, or they may agree to let the central bank sell all or part of the shares that constitute the central bank’s claim on banks’ profits into the national or international markets, thereby allowing for a faster resolution of the subordinated loan problem. At the end of 1995, a new banking law proposal was sent to Congress, The proposal, which is currently being discussed by parliament, will enable those Chilean banks that satisfy a minimum capital adequacy requirement to expand internationally through cross-border lending, the establishment of branches, subsidiaries, and joint ventures, and by moving into previously restricted areas such as factoring and derivatives. Discussions with banks and the Superintendency of Banking Supervision suggest that although it would improve portfolio diversification, the expansion of Chilean banks into these areas would pose serious challenges to the supervisory authorities. To avoid compromising the soundness of the banking system, the expansion of bank activities would have to he accompanied by an increase in responsibilities and expertise of Chilean bank supervisors, and the development of the capability to monitor, on a functionally and internationally consolidated basis, the risk exposures of bank operations.
A rapid expansion of bank lending from 1989 to 1993 following the liberalization of the financial system, the strengthening of public finances, and a large inflow of capital from abroad, in an environment of insufficient regulatory and supervisory oversight, led to an increase in the level of nonperforming loans in the Mexican banking system since 1993. However, improvements in oversight and in credit risk management on the part of the banking system kept the level of nonperforming assets within tolerable limits. But the unprecedented increases in interest rates in the aftermath of the devaluation of the peso, and the deteriorating macroeconomic environment, resulted in a rapid expansion of nonperforming loans across the entire banking system. In addition, several of the internationally active banks sustained large losses on their off-balance-sheet net foreign liabilities. As of March 1996, nonperforming loans for 12 recapitalized banks (with a share of more than 75 percent of total loans) were estimated at around 14 percent of total loans, measured according to internationally accepted accounting standards. In order to forestall a further deterioration of bank balance sheets and prevent a loss of depositor confidence, the authorities established a number of government-financed programs aimed at easing the burdens of both banks and their borrowers. Four institutions were subject to intervention during the year, and two more had been assisted as of June 1996. Despite the fragility of the banking system, aggregate net income decreased only slightly in 1995, and the aggregate capital ratio for nonintervened banks rose from 9.8 percent at the end of 1994 to 12.1 percent at the end of 1995 and 13.1 percent by the end of April 1996. Overall, government support to the banking system has come at a considerable cost, officially estimated at MexNSI48 billion (6.6 percent of 1996 GDP). Given the moderate level of public debt in Mexico, and the fact that these costs are spread over a large time span, the authorities consider that these programs will not affect significantly the strength of public finances. Although the banking system remains burdened by a significant volume of nonperforming loans, by the end of 1995 the banking system had been substantially stabilized.
The Venezuelan banking system continues to feel the effects of the severe crisis that began with the failure of Banco Latino in early 1994. Since that time, the government has provided an estimated $11 billion (15 percent of 1995 GDP) in direct financial assistance to 18 banks, which held more than half of all bank deposits. The cost of financial assistance could increase further in 1996, Systemwide nonperforming loans have declined since 1994 to an average of 7 percent of total loans in 1995. Controls on loan and deposit rates provided some support to the banking sector by maintaining spreads in excess of 10 percentage points, and a relatively sharp increase in lending rates allowed banks to maintain, and in some cases widen, these spreads after controls were lifted in April 1996. However, a weak economy and negative real deposit rates in 1994 and 1995 resulted in negative real growth in loans. Banks’ holdings of government securities have grown significantly in real terms, although they are earning relatively low real rates of interest. The government has announced plans to privatize four of the state-owned banks in 1996-97. With support from multilateral institutions, the authorities also plan to provide matching loans to banks that raise new shareholder capital, as well as provide financial assistance for bank mergers in order to promote consolidation.
In Asia, several banking systems are undergoing major transformations in the presence of volatile capital flows and exchange market pressures. The banking system in India remains burdened with nonperforming loans amounting to roughly 5 percent of GDP, only one third of which have been covered by reserves. The government has partially recapitalized banks and is requiring them to implement tougher asset classifications and provisioning and to adopt a more realistic income reporting policy in hopes that this will allow the banking system to resolve the bad loan problem out of current earnings—with little further financial assistance—in two to three years. It is likely that the ultimate resolution of the problem will extend beyond this horizon. One possible approach considered by the authorities was to recapitalize the banks and securitize their bad loan portfolios through a centralized nationwide Asset Reconstruction Fund. But this solution was not accepted on the grounds that a new institution would suffer from a lack of geographic reach and might give rise to moral hazard problems. Instead, a range of initiatives—taking into account the health of individual banks—was adopted. Any recapitalization of the banking system would have to be accompanied by greater supervisory efforts and further efforts to phase out the myriad regulations and to improve managerial autonomy, which both tend to impede the efficiency of operations in the commercial banks.
Significant improvements in the Indonesian banking system, including improved banking supervision and regulation, have brought most banks in compliance with Indonesia’s 8 percent minimum capital adequacy requirements. Nevertheless, deficiencies in the mechanisms for enforcing loan contracts and for liquidating banks have allowed several insolvent banks to remain in place.7 Banking authorities are pushing efforts to resolve these problems either through bank mergers with stronger institutions, or bank acquisitions by new investors, and Bank Indonesia has implemented a regime of moral suasion with the intention of vetting banks’ annual business plans, including credit expansion. In 1995 and again in early 1996, the authorities expressed concern about the growing exposure of the banking system to the booming real estate market, and growth in bank credit to this sector appears to be tinder better control. Prudential regulations have been tightened in recent months, including rules concerning lending to related parties and the participation of banks in the commercial paper market, and there has been an expansion in the supervision and regulation of nonbank finance companies, many of them having close ties to banks. Despite these measures, some banks appear to have remained vulnerable to macroeconomic imbalances and disturbances, including increases in domestic interest rates and reversals of capital inflows.
According to standard indicators, the banking system in Thailand appears robust. Thai banks are well capitalized, have low levels of nonperforming loans, and seem to comply with the prudential guidelines of the Basle Committee on Banking Supervision. Discussions with market participants and supervisors suggest that although the highly cartelized banking system, which is dominated by four large institutions, is likely to continue to control the domestic and external flow of funds, newly authorized banks and nonbank financial institutions are beginning to compete with the established banks in areas of activity requiring some degree of financial sophistication. This competition is gradually forcing banks to be more innovative, to generate more fee-based revenues, and to cut costs, all of which will improve their performance in the future. Lending has grown rapidly over the past five years and is highly concentrated in the few large industrial conglomerates and in real estate borrowers, and banks hold a substantial portfolio of medium- and long-term domestic assets funded by short-term domestic and foreign liabilities. Therefore, the ability of the banking system to withstand a more turbulent domestic or foreign environment—including reduced access to foreign markets, a sustained increase in domestic interest rates, and lower growth rates—cannot be taken for granted.
Turning to selected European countries in transition, banks in the Czech Republic and Hungary are continuing to struggle with the imposing demands of the transition to a market-oriented economic system and incentive-based internal operations. Although several rounds of bank restructuring and recapitalization have eliminated the bulk of inherited nonperforming assets, bank balance sheets have continued to deteriorate. In the Czech Republic, nonperforming loans amounted to 35 percent of the total loan portfolio at the end of December 1995. Most Czech banks, particularly the larger ones, are thought to have reserves that fully cover these bad loans. In Hungary, 19 percent of bank loans were nonperforming at the end of June 1995. The restructuring of these two banking systems is continuing, including the privatization of the remaining financial institutions that are still subject to state ownership.8
Against this background of fundamental change and concerns about the strength of some of the smaller banks, the authorities are monitoring closely the surge of capital flows into the region. The increase in foreign borrowing, particularly in the Czech Republic, where capital inflows exceeded 15 percent of GDP in 1995, has raised concerns about the net foreign liability positions of Czech enterprises and banks and their vulnerability to a depreciation of the koruna under the new wider exchange rate band.
Spurred on by their intentions to join the Organization for Economic Cooperation and Development (OECD) and the European Union, both countries have been making impressive progress in improving the legal and regulatory framework for the financial sector, and in introducing banking regulations that in most respects match those of the major industrial countries. There are, however, some doubts about their ability to implement and to enforce these regulations credibly.
All three of the Baltic countries have experienced serious banking difficulties since 1992—93. Common causes include a legacy of weak lending practices of state-owned banks, economic contractions in the immediate post-independence period, rapid expansions in bank lending after 1992 in a regime of low capital requirements, and insufficiently rigorous regulations and inspections. The Estonian banking system, where banks holding at least one third of the banking system’s assets have failed, has since put the crisis behind it, and rating agencies now view it as the best in the region. The crisis in the Latvian banking system dates back to 1995, when 15 banks were closed and the activities of another 2 were suspended, reducing the number of commercial banks in operation by 30 percent during one year. The crisis had been brewing for some lime and reflected major structural weaknesses in the banking sector. The failure of Bank Baltija (which held about 30 percent of total deposits), and the associated losses incurred by depositors, set the stage for a crisis of confidence in the banking system that resulted in a massive contraction of banking industry deposits. In Lithuania, ten banks were closed in 1995, including the two largest privately controlled banks (which held 23 percent of system deposits). The crisis led to a reduction of litas deposits by 15 percent and convertible currency deposits by 19 percent between December 1995 and February 1996. The three largest banks, all state-controlled, are believed to have high proportions of nonperforming assets, but bank liabilities are fully guaranteed by the government.
All three countries have responded to these banking problems by strengthening supervisory powers, prudential regulations, and capital-adequacy requirements. These measures are expected to lead to significant banking sector consolidation in the coming year or two. In April 1996, the local number of banks in the three Baltic countries had declined from a peak of 150 institutions to 80, of which 24 institutions did not have full deposit-taking licenses.
The sharp depreciation of the Turkish lira in the first months of 1994. and the associated surge in interest rates, created significant losses for many Turkish banks, which had large net foreign currency liabilities and sizable mandatory holdings of government paper at the onset of the crisis. These problems were exacerbated by a run on the banks in March-April 1994. and a deterioration in the quality of loan portfolios as the economy weakened and nonperforming loans increased, culminating in the closure of three banks in late April. The implementation of adjustment measures in the context of an IMF-supported program in April 1994 helped to reverse capital outflows. Together with the introduction of 100 percent deposit insurance in May 1994 and further increases in domestic interest rates, the adjustment measures laid the basis for a sharp recovery in bank deposits. With treasury paper paying high real interest rates, banks were able to record sizable profits on their securities portfolios during the second half of 1994 and into 1995. Asset quality also improved as the economy recovered from the 1994 recession. One small state-owned bank was privatized late in 1995, and a second is expected to be sold in 1996, but the four largest state banks continue to play a dominant role in the financial system. The banking system in Turkey will continue to be challenged by a difficult macroeconomic environment.
Global Bond Markets.
Between the end of 1994 and January 1996. the world’s major bond markets witnessed one of the greatest rallies in recent history. In this period, the total return on long-term U.S. Treasury bonds reached 30 percent and returns in most of the other major sovereign fixed-income markets were between 15 percent and 28 percent. The rally was supported by an environment of declining interest rates related to the optimism about the prospects for a U.S. budget deficit reduction, expectations of a soft landing in the United States, and continued weak growth throughout Europe and in Japan. Accordingly, there were si/able downward adjustments in the yield curves of major countries (Chart 3). and by early 1996 long-term interest rates in some of these countries reached their lowest levels in more than a decade (Chan 4). Furthermore, the high levels of volatility in bond markets that had emerged with the onset of the turbulence in carls 1994. and which were sustained by developments during the crisis in emerging markets in early 1995, started to decline to more normal levels during the summer and fall of 1995.
The factors that lifted global bond markets throughout 1995 began to subside in February 1996 as economic reports suggested the possibility of stronger than expected growth in the United Stales. Further evidence of signs of strength in the U.S. economy and disappointment over progress in achieving a reduction in the U.S. budget deficit then precipitated a pronounced shift in market sentiment. By the end of May 1996. the yield on ten-year U.S. Treasury bonds had risen by 128 basis points from its January low.
Although bond markets elsewhere were not left untouched by this rise in U.S. interest rates, there clearly was a “decoupling” of global bond markets from the U.S. market, particularly when compared to the reaction to the correction in U.S. bond markets in 1994 (Table 2). In early 1996. among the major industrial countries. Canadian and U.K. ten-year interest rates were most highly correlated with U.S. ten-year interest rates, increasing by about half of the rise in U.S. rates. Meanwhile, German long rates increased by about 35 percent of the rise in U.S. rates: Japanese rates rose only marginally; and French and Italian rates actually fell quite sharply during this period—more than 100 basis points in the case of Italy. By contrast, during the market correction in early 1994, interest rates in all of the major industrial countries rose sharply, in most cases by more than U.S. long-term interest rates.9
Major Industrial Countries: Changes in Ten-Year Government Bond Yields
(In basis points)
Major Industrial Countries: Changes in Ten-Year Government Bond Yields
(In basis points)
| January 1, 1994- | January 1, 1996- | ||
|---|---|---|---|
| May 31, 1994 | May 31, 1996 | ||
| United States | 135 | 128 | |
| Japan | 62 | 7 | |
| Germany | 142 | 47 | |
| France | 173 | -17 | |
| Italy | 113 | -109 | |
| United Kingdom | 254 | 73 | |
| Canada | 211 | 63 | |
| Average | 155 | 27 | |
Major Industrial Countries: Changes in Ten-Year Government Bond Yields
(In basis points)
| January 1, 1994- | January 1, 1996- | ||
|---|---|---|---|
| May 31, 1994 | May 31, 1996 | ||
| United States | 135 | 128 | |
| Japan | 62 | 7 | |
| Germany | 142 | 47 | |
| France | 173 | -17 | |
| Italy | 113 | -109 | |
| United Kingdom | 254 | 73 | |
| Canada | 211 | 63 | |
| Average | 155 | 27 | |
There are, nevertheless, close parallels between the two recent swings in global bond yields from early 1992 through mid-1994 and from late 1994 through mid-1996 (see Chart 4). including a similar decline in long-term bond rates to below 6 percent, followed by a sudden and sharp adjustment in prices accompanied by a rise in both price volatility and turnover in related derivative markets. This raises the question why the adjustment in global bond markets was far more benign in the first half of 1996 than in the first half of 1994 (see Table 2). An important part of the answer is that the combination of factors that had driven events in early 1994 was not repeated, First, unlike 1994. the U.S. Federal Reserve did not raise interest rates in the first half of 1996. Second, in the earlier episode, concerns about government budget deficits and debt levels in Europe led to a rise in bond yields. Third, macroeconomic conditions remained weak in late 1995 and early 1996 in many countries in Europe and in Japan and are thought to be inconsistent with substantial increases in interest rates. These factors suggest that an important root cause of the decoupling of bond yields among the major industrial countries is that the economies of Canada, the United Kingdom, and the United States are at a different stage of the business cycle (with relatively stronger growth) than the economies in continental Europe and Japan.
Another important and related reason for the dramatic correction in 1994 was the unwinding of highly leveraged positions along the then very steep U.S. yield curve. The sleep yield curve in late 1993 and early 1994 made it profitable for hedge funds to finance purchases of long-term Treasury securities with short-term dollar borrowing. When the Federal Reserve tightened monetary policy in February 1994, the rush to unwind these positions hastened and steepened the correction in bond markets. During 1995, the yield curve in the United Stales had been too flat to make such trades profitable. In addition, there is, for now, considerably less leverage in the bond market at present than there was in early 1994 because macro-hedge funds and proprietary traders—which suffered serious losses from leveraged yield curve plays in 1994—appear to have since turned their focus toward less risky cross-market and cross-instrument arbitrage, and away from the more typical unhedged leveraged directional bets.10 Furthermore, risk-management systems of institutional investors have become more rigorous. The large losses in 1994 caused many institutions to impose and enforce strict limits on the size of market positions. This is not to say that instances of extreme market movements are a thing of the past, but it does suggest that for now the probability of a major bout of market turbulence is relatively low. Consistent with this background, market participants were not overly concerned that a modest increase in U.S. interest rates over the next six months, brought about by the U.S. Federal Reserve, might again produce serious market turbulence and financial stress among major market participants, even though such an increase is not expected. While a modest increase in U.S. interest rates would generally be expected to have a negative impact on flows to emerging markets, it was again not thought likely to precipitate the type of major rebalancing of international portfolios experienced in these markets in early 1995,
Issues relating to the Economic and Monetary Union (EMU) have again come to the fore in European bond markets. The steepness of the German yield curve is interpreted by some market participants and authorities in various countries as reflecting in part an “EMU premium” of 50-100 basis points in long-term rates. This may be one reason behind the convergence in interest rates in Europe since the first quarter of 1995, although in many cases spreads are wider than they were in early 1994 (Chart 5). From the peaks reached in 1995, spreads on ten-year bonds vis-à-vis Germany have narrowed in France from 101 basis points to 2 basis points at the end of May 1996: Belgian spreads have narrowed from 91 to 25 basis points: Spanish spreads fell from 528 to 267 basis points: Swedish spreads have narrowed from 444 to 203 basis points: and Italian spreads narrowed from 602 to 292 basis points. In contrast, U.K. spreads have widened from 102 basis points in early 1995 to 163 basis points at the end of May 1996.
Market participants and authorities in many countries attribute this convergence variously to improved prospects for EMU, a reversal of the “flight to quality” following the Mexican crisis, and country-specific factors, especially progress on fiscal consolidation in the cases of Italy and Sweden. However, even if much of it is attributable to the perception of improved prospects, it is clear that markets remain uncertain about the possibility of more than a few countries participating in EMU at the start. Interest-rate swap spreads can be used to gauge expectations of future interest rate differentials (Chart 6). For example, at the end of May 1996, spreads over deutsche mark five-year swaps were as follows: near zero for the French franc, the Netherlands guilder, and the Belgian franc: 225 basis points for the Italian lira; 210 basis points for the pound sterling: 290 basis points for the Spanish peseta: and 245 basis points for the Swedish krona. Turning to implied forward yield curves in January 1999, the spreads are roughly unchanged for the French and Belgian francs and the Netherlands guilder, and they narrowed by 60—85 basis points for the pound, the krona, and the peseta, and by 20 basis points for the lira. Looking out an additional two years to January 2001. these spreads narrow again, by about 20 basis points for the pound, the krona. the lira, and the peseta.
Selected European Countries: Spot and Implied Forward Yield Curves, May 31, 1996, January 4, 1999, January 3, 2000, and January 2, 2001
(In percent)
Source: Bloomberg Financial Markets.Note: Forward yield curves are calculated from the implied forward rates in the spot interest rate swap curves.1 Except for the United Kingdom (January 4, 2000).Japan and Germany: Exchange Rates
(Local currency per unit of U.S. dollar)
Source: Bloomberg Financial Markets.Regardless of how much of the recent developments in interest rates in Europe can be attributed to expectations of EMU, market participants confirmed that there was currently no evidence that large financial positions were being built up in anticipation of a smooth transition toward EMU. unlike the large buildup of convergence positions from 1990 to 1992. Hence, although there may be periods of increased volatility, it is not expected that the market impact of negative news regarding the prospects of EMU would be magnified by the unwinding of such convergence positions.
Major Currency Markets
In early 1995, a new episode of market volatility began to unfold in the major currency markets, particularly in the dollar-yen market.11 The dollar moved from a value of ¥ 101 in early January down to ¥79 in mid-April, and also weakened against the deutsche mark, declining from DM 1.56 to DM 1.35 over the same period (Chart 7). Currencies in the ERM of the European Monetary System (EMS) also experienced relatively sharp movements, reflecting growing skepticism about the sustainability of fiscal positions in many European countries, growing uncertainty about EMU, and political uncertainty in several countries (see Chart 14 in Annex I). The deutsche mark appreciated against the French franc, the Italian lira, and the pound sterling, and reached record highs against several other European currencies. Moreover, in March, the Portuguese escudo and the Spanish peseta were both devalued, the first realignment within the ERM since the widening of the bands in 1993.
RiskMetrics Daily Price Volatility for U.S. Dollar Spot Exchange Rates, January 19, 1995-May 31, 1996
(In percent)
Source: J.P. Morgan.A combination of macroeconomic factors was widely perceived to have contributed to the weakness of the dollar and the strength of the yen. These factors included spillover effects of the Mexican crisis, diminished expectations of further U.S. interest rate increases related to signs of a slowdown in U.S. economic growth, the persistence of the Japanese current account Surplus and the U.S. current account deficit, an intensification of Japan-U.S. trade tensions, and a shift away from dollar-denominated assets by Japanese investors.
Despite these valid macroeconomic explanations, some market participants did not accept that all of the yen’s appreciation against the dollar was entirely consistent with existing fundamentals. Instead, short-term trading conditions and the impact of the use of “knock-out” options were frequently cited as contributing to the movements in the value of the yen.12
Market reports suggested that by late February 1995, the major dealers had accumulated on their books a substantial quantity of options with knock-out values somewhere between ¥90 and ¥80. These options were reported to have been purchased largely by Japanese corporations during the previous 12 months to partially hedge the yen value of expected dollar receivables against the possibility of a further modest appreciation of the yen. As the yen appreciated from ¥90 per dollar toward ¥80 in early March, dealers had an incentive—to the extent they could influence the market—to push the value of the yen up through the knock-out levels and thereby eliminate their obligations under the options contract. Market participants believe that under the circumstances prevailing in March 1995—a bunching of limit orders, spot prices approaching the knock-out levels, and momentum from technical analysis—trading by major dealers can influence the spot exchange rate for a short period around a specific knock-out value, and that this might have provided some pressure on the exchange rate to move to the next lower knock-out value, and so on.13 Despite uncertainty about their influence on the spot exchange rate, market participants generally believe that these instruments had some influence on currency options markets. Prices of options, expressed in terms of the implied spot price volatility associated with them, doubled within the four trading days from March 2 to March 7, 1995. It is possible that along with other macroeconomic factors, trading conditions contributed to the rise in the value of the yen against the dollar in March-April 1995.
Viewed from the perspective of international capital markets, the most notable feature of this period besides the dramatic currency movements themselves was the temporary increase in transactions volume and price volatility in a wide range of markets—foreign exchange, bond, equity, and derivatives. By some accounts, the average daily turnover in global currency markets increased by about 100 percent in March 1995 to about $2 trillion. Associated with this increased turnover was a sharp rise in both daily and weekly exchange rate volatility (Chart 8). For example, volatility in the dollar-yen rate tripled in March and rose further in April, and there were several days in March when volatility in the dollar-yen rate increased by 50 percent in a single day. Price variations also rose in virtually all of the cross-rates between the other major currencies, even in the relatively low-volatility cross-rate between the U.S. dollar and the Canadian dollar. Volatility also rose dramatically in bond and derivative markets, but less so in equity markets. For example, price volatility of yen-denominated bonds tripled between early March and mid-April 1995, and global monthly turnover in exchange-traded options and futures increased by one third, from 90 million contracts a month in January and February to 120 million contracts in March 1995.
Toward the end of April 1995, the value of the dollar stabilized against the yen at around ¥85 per dollar. The dollar staged a dramatic recovery in August and September, following an easing of interest rates in Japan, the liberalization of foreign currency investment limits for Japanese insurance companies, and coordinated exchange market intervention, and then experienced a more gradual recovery against the yen throughout the remainder of 1995. The dollar also stabilized against the deutsche mark during April-July 1995, experienced sharp gains and then losses in late summer, and began a more gradual recovery through the fall and winter months of 1995. In the initial period of the dollar’s recovery from April through September 1995, trading volumes and price volatilities declined somewhat in currency and derivative markets, but volatility appeared to have remained higher than it was before the period of dollar weakness early in the year. Market participants attributed some of the dollar’s recovery in April-September 1995 to several episodes of coordinated intervention by the major central banks, and especially by the Bank of Japan. In retrospect, however, two factors appear to have contributed importantly to the relative success of this coordinated intervention: the original decline in the dollar was not warranted by fundamentals, and the subsequent release of economic indicators and policy efforts implied that a higher value of the dollar against the yen and deutsche mark was appropriate.
In early 1996, the dollar’s recovery against the yen and the deutsche mark, and the accompanying drop in volatility, appears to have been sustained, although renewed concerns about the dollar-yen rate surfaced again in mid-February. While market participants believed that further easing of interest rates in Germany would hold the deutsche mark in check, they had several reasons to expect that the yen could strengthen: first, a high probability that near-zero short-term interest rates in Japan would soon rise; second, the possibility that such a rise in short-term interest rates would spark a sell-off in the U.S. government bond markets by hedge funds and proprietary traders covering their short-yen positions;14 and third, the long-anticipated repatriation of foreign investments by Japanese investors before the end of the fiscal year in
March 1996. In the event, volatility rose sharply in the yen-dollar market, but soon dissipated as heavy foreign exchange market intervention by the Bank of Japan was widely interpreted as a commitment to preventing a rise in the yen. Moreover, a greater than expected decline in the Japanese current account surplus in early 1996 was interpreted by market participants as a factor that could support a lower value of the yen. Within Europe, fiscal consolidation efforts in a number of countries—notably Italy and Sweden—appear to have reduced tensions within currency markets, notwithstanding continued uncertainty about the timing and country composition of EMU. Moreover, discussions with market participants in Europe and New York indicate that at this time, there is no evidence of a significant leveraged position taking against any of the major currencies.
Major Industrial Countries: Equity Price Indices
(Indices, January 1995 = 100; logarithmic scale)
Source: The WEFA Group.Note: Monthly data for the following indices are used: Dow Jones Industrial (United States); Nikkei 225 (Japan); Commerzbank (Germany); CAC General (France); Banca Commerciale (Italy); FT-SE 100 (United Kingdom); and Toronto SE Composite (Canada).Global Equity Markets
In 1995. equity markets in Canada, the United Kingdom, and the United States posted the largest gains, while stock prices in France, Germany, and Italy fluctuated throughout the year with no apparent trends (Chart 9 and Table 3). Although the Japanese stock market declined sharply in the first half of 1995, reflecting concerns about the record appreciation of the yen, investors returned to this market later in the year, as the yen depreciated, and pushed equity prices back to the level where they began the year. In 1996, concerns about the economic outlook and company earnings have reduced stock returns in Canada and the United Kingdom, but the U.S. equity markets continued to rise to new highs. In Europe, a more favorable outlook for interest rates and inflation in France, Germany, and Italy have allowed the equity markets in these countries to gain some positive momentum. The Japanese market has remained sluggish, in part because the heavy foreign buying of Japanese equities was offset by substantial selling by domestic institutions.
The equity market with the largest capitalization—the U.S. market—benefited to some extent in 1995 from the slowing of flows to emerging markets during the aftermath of the Mexican crisis, but solid corporate earnings and optimism about the ongoing restructuring of U.S. industry established a sound basis for the increase in equity values. Another important factor behind the surge in U.S. equities has been the steady flood of savings into equity mutual funds, rather than into bank liabilities. The U.S. mutual fund industry is well on the way of becoming as important a depository of the nation’s savings as U.S. commercial banks.15 This flood of savings into mutual funds has continued at record-setting levels into 1996, amounting to $600 billion since January 1995. The U.S. equity market has raised shareholder’s wealth (market capitalizations) on the three largest stock exchanges by roughly $2 trillion since the start of 1995. an increase of about 40 percent. Such gains have naturally attracted both attention and concern.
The chief concern in various financial centers is that the U.S. market might be poised for a sharp and sustained correction, which could spread to Other markets. Although the possibility of such a correction cannot be ruled out—there are some indicators suggesting equity price pressures might be high relative to fundamentals—the evidence suggesting significant risk of a massive correction is not overwhelming. Price-earnings ratios have increased from about 16 to 19 for the Standard and Poor’s 500 (S&P 500) index since the end of 1994, whereas between January 1987 and October 1987 it increased from about 16 to 22. But more important, the earnings yield on U.S. stocks has not diverged from the yield on long-term bonds, as it did in 1987 (Chart 10).
Major Industrial Countries: Bond and Equity Index Returns
(In percent)
For government bonds with maturities of seven-ten years of maturity.
Major Industrial Countries: Bond and Equity Index Returns
(In percent)
| 1995 | Fourth Quarter 1995 | January-May 31, 1996 | |||||
|---|---|---|---|---|---|---|---|
| Local | U.S. | Local | U.S. | Local | U.S. | ||
| currency | dollars | currency | dollars | currency | dollars | ||
| Broad Index1/ | |||||||
| United States | 23.73 | 23.73 | 6.38 | 6.38 | 4 62 | -4.62 | |
| Japan | 15.47 | 11.08 | -0.70 | -4.34 | 1.02 | -3.47 | |
| Germany | 18.91 | 28.15 | 5.82 | 5.06 | 0.35 | -5.61 | |
| France | 19.21 | 29.60 | 7.10 | 7.41 | 4.21 | -1.21 | |
| Italy | 19.83 | 22.40 | 7.65 | 9.21 | 11.48 | 14.65 | |
| United Kingdom | 18.69 | 17.53 | 6.53 | 4.16 | -0.80 | -0.72 | |
| Canada | 22.73 | 26.10 | 6.26 | 4.47 | 0.01 | -0.31 | |
| Equity index | |||||||
| United States (S&P500) | 34.11 | 34.11 | 5.39 | 5.39 | 9.33 | 9.33 | |
| Japan (Nikkei 225) | 0.74 | -3.09 | 10.91 | 6.85 | 10.51 | 5.70 | |
| Germany (Commerzbank) | 5.26 | 13.43 | 2.39 | 1.66 | 8.65 | 2.02 | |
| France (CAC 40) | -0.49 | 8.19 | 4.68 | 4.97 | 13.09 | 7.08 | |
| Italy (Banca Commerciale) | -6.78 | -4.78 | -4.50 | -3.12 | 13.49 | 16.25 | |
| United Kingdom (FT-SE 100) | 20.35 | 19.18 | 5.16 | 2.82 | 1.34 | 1.35 | |
| Canada (TSE-300) | 11.86 | 14 93 | 4.06 | 2.30 | 10.99 | 10.46 | |
For government bonds with maturities of seven-ten years of maturity.
Major Industrial Countries: Bond and Equity Index Returns
(In percent)
| 1995 | Fourth Quarter 1995 | January-May 31, 1996 | |||||
|---|---|---|---|---|---|---|---|
| Local | U.S. | Local | U.S. | Local | U.S. | ||
| currency | dollars | currency | dollars | currency | dollars | ||
| Broad Index1/ | |||||||
| United States | 23.73 | 23.73 | 6.38 | 6.38 | 4 62 | -4.62 | |
| Japan | 15.47 | 11.08 | -0.70 | -4.34 | 1.02 | -3.47 | |
| Germany | 18.91 | 28.15 | 5.82 | 5.06 | 0.35 | -5.61 | |
| France | 19.21 | 29.60 | 7.10 | 7.41 | 4.21 | -1.21 | |
| Italy | 19.83 | 22.40 | 7.65 | 9.21 | 11.48 | 14.65 | |
| United Kingdom | 18.69 | 17.53 | 6.53 | 4.16 | -0.80 | -0.72 | |
| Canada | 22.73 | 26.10 | 6.26 | 4.47 | 0.01 | -0.31 | |
| Equity index | |||||||
| United States (S&P500) | 34.11 | 34.11 | 5.39 | 5.39 | 9.33 | 9.33 | |
| Japan (Nikkei 225) | 0.74 | -3.09 | 10.91 | 6.85 | 10.51 | 5.70 | |
| Germany (Commerzbank) | 5.26 | 13.43 | 2.39 | 1.66 | 8.65 | 2.02 | |
| France (CAC 40) | -0.49 | 8.19 | 4.68 | 4.97 | 13.09 | 7.08 | |
| Italy (Banca Commerciale) | -6.78 | -4.78 | -4.50 | -3.12 | 13.49 | 16.25 | |
| United Kingdom (FT-SE 100) | 20.35 | 19.18 | 5.16 | 2.82 | 1.34 | 1.35 | |
| Canada (TSE-300) | 11.86 | 14 93 | 4.06 | 2.30 | 10.99 | 10.46 | |
For government bonds with maturities of seven-ten years of maturity.
Indeed, the sharp rise in equity prices can be viewed as partly reflecting a realignment toward parity between bond and equity yields in an environment of rising earnings. Further, although the implied price volatility embodied in futures call options suggests an increase in future price volatility in U.S. equity markets, the implied increase is well within the experience of other countries and within historical norms. Finally, it should be noted that market participants and supervisors are confident that the major financial systems are now much better prepared to absorb a substantial price adjustment than in earlier years. Significant progress has been made over the past decade both in terms of strengthening market infrastructure and in risk management by the major financial institutions with large equity positions. Hence, as in the bond markets, it is relatively unlikely that an adjustment in equity markets would initiate a period of major market turbulence.
Global Derivative Markets
Only a decade ago, global derivative markets were small and underdeveloped compared with the other international capital markets. Since then, the average annual growth rate of OTC and exchange-traded derivative markets has been 40 percent a year. The first survey by the Bank for International Settlements (BIS) of the OTC derivatives markets in 26 countries estimated the notional value of outstanding OTC foreign exchange, interest rate, equity, and commodity derivative contracts at $47.5 trillion (after adjusting for double counting) at the end of March 1995.16 All but about 2 percent of this total is accounted for by interest rate (56 percent) and currency derivatives (42 percent). In addition to OTC derivatives, the intermediaries captured by this survey were engaged in a further $17 trillion of exchange-traded derivatives.17 In aggregate, therefore, the global market currently is more than twice as large as world output, and in many cases exceeding the size of the underlying cash markets: for example, the stock of outstanding fixed-income securities in the OECD countries is about $24 trillion. However, compared with the total notional value of outstanding OTC derivatives, the replacement value of all OTC derivative positions—the outstanding credit exposure of the major banks—is estimated at $2.4 trillion, or only about 5 percent of the notional total. But although the replacement value is relatively small compared with the notional value of these derivative contracts, it is nevertheless more than three times as large as the capital of the world’s 75 largest banks.
The notional principal outstanding of exchange-traded derivatives rose by less than 4 percent in 1995, compared with an average annual growth rate of 40 percent during the past decade. There was also a shift in product types toward simpler “plain vanilla” structures with greater market liquidity and finer spreads, away from the more exotic option-related structures. For example, the structured note market in the United Stales virtually collapsed in 1995.18 While declining interest rates in several major industrial countries reduced the demand for interest rate hedging products, the enthusiasm in this market was also significantly dampened by the major losses experienced by some financial and nonfinancial firms. The greater use of plain vanilla products may be a sign of greater consideration of customer suitability by financial firms, and of more caution on the part of commercial companies about entering into contracts without fully understanding the risks. Although difficult to quantify, this trend away from exotic derivative structures toward greater use of plain vanilla structures in the OTC derivative contracts might have intensified the competition between the OTC and exchange-traded sectors of the derivatives market. Overall, in 1995, it became increasingly clear that derivative markets were showing signs of maturing.
United States: Earnings-Price Ratio of Stocks and Yield on Long-Term Government Bonds
(In percent a year)
Sources: Morgan Stanley Capital International and The WEFA Group.Developments in Selected Industrial Country Banking Systems
The recent performance of the banking systems in the Group of Seven countries has been mixed. On the one hand, banks in Canada, the United Kingdom, and the United States have high loan growth and minimal asset quality problems. On the other hand, banks in France, Italy, and Japan have struggled with weak loan growth and large portfolios of nonperforming assets.
The banking system in the United States is widely perceived to be in its best health in recent years (Chart 11). The major U.S. banks have been successful in reducing their real estate and other loan concentrations, in re-establishing a profitable and solid retail base, and in creating a “we 11-seasoned” on-balance-sheet asset portfolio. Asset quality is the best it has been in the past 11 years, although supervisors have noted a slight increase in late 1995 in defaults on consumer loans. Furthermore, discussions with regulators, rating agencies, and banks themselves suggest that the diversification in fee activities and loan portfolios has made major U.S. banks more robust to the macroeconomic shocks associated with business cycles. It has also been observed that U.S. banks and investment banks have made substantial progress in developing sophisticated risk management systems; indeed, U.S. institutions lead the industry in risk management technology.
In contrast, market participants generally believe that even with a moderate economic recovery in Japan, it would take some time to completely resolve Japanese banks’ nonperforming loans problems. Nevertheless, the international banking community is relatively unconcerned about the possibility of a market disruption due to the failure of a major Japanese bank, and it no longer requires a “Japan premium” on its exposure to Japanese banks. The Ministry of Finance reports that in March 1996 there were ¥34.7 trillion of nonperforming loans or about 5 percent of total loans outstanding for deposit-taking institutions, of which ¥12.5 trillion were covered by special provisions, ¥12.0 trillion were covered by collateral, and ¥1.9 trillion of the exposure to housing loan companies will be written off but will still remain on the books; this left ¥8.3 trillion {equal to the banking system’s unusually high 1995 operating profit level) to be disposed of in the future.19 However, market participants cite various estimates of nonperforming loans by international rating agencies and private financial institutions. These estimates tend to be based on broader definitions and coverage of suspected nonperforming loans, and are much higher than the Ministry of Finance’s estimates. In particular, recapitalized loans (additional loans to pay for interest) and loans that have been restructured at above the official discount rate are not classified as nonperforming loans. Some observers believe that the book value of collateral could be above present and expected future market prices so that the amount to be disposed of might be higher even if the amount of nonperforming loans remained the same. The Ministry of Finance has also declared that it will observe the situation carefully and that it will make its best effort to collect the loans. Some market participants also warn about the possibility of a further increase in non-performing loans held by nonbank financial institutions. These nonbank loans do not pose a systemic risk, but in practice the losses might be absorbed by parent financial institutions.
United States: Net Income of Commercial Banks
(In percent of total assets)
Source: Federal Deposit Insurance Corporation.Several failures of smaller Japanese institutions from December 1994 to the summer of 1995, followed by the problems at Daiwa Bank in October, appear to have strengthened the authorities’ determination to tackle the problems in the financial sector. The action was also prompted, in part, by the emergence of the “Japan premium” in the interbank market. After years of limited disclosure about the magnitude of problem loans, detailed estimates were released in November 1995 by the Ministry of Finance, and in December plans were announced for liquidating the jusen (housing loan companies), winding up other failed institutions, and improving bank supervision—including adopting a “prompt corrective action” approach by supervisors. These measures are generally regarded as a step in the right direction, reducing the risk in a continuing resolution of the nonperforming loans problem.
In mid-1995, ¥6.4 trillion of immediate losses in the jusen and ¥6.7 trillion, or about I percent of total loans held by deposit-taking institutions, of other doubtful jusen assets became the focal point of the fragility of the Japanese financial system, particularly since the jusen losses amounted to about one third of all losses in the banking system. Whether the banks that “founded” the jusen and owned them as subsidiaries or whether the agricultural cooperatives that lent to the jusen should pay for the losses became a political issue once a decision to close the jusen was made. The government announced on December 19 (as a Cabinet decision) that most of the immediate losses (¥5.2 trillion of ¥6.4 trillion) would be shouldered by the banks, while the contribution from the agricultural cooperatives was limited to ¥530 billion. The government would inject ¥680 billion from the FY 1996 budget to fill the gap in the “immediate losses.” In addition, the government would make a ¥5 billion contribution to the jusen account of the Deposit Insurance Corporation. The Jusen Resolution Corporation will be set up to assume the remaining assets of ¥6.7 trillion to write off bad debts in the future. The banks and agricultural cooperatives will be asked to contribute to the Jusen Resolution Corporation. In addition to the jusen resolution, other financial reform bills were prepared, including deposit insurance reform and the introduction of prompt corrective action.
Despite the relatively small amounts involved, the ¥685 billion in public funds became the center of controversy in the budget debate, and widespread political opposition to the use of public money had thrown the original jusen plan into doubt. The budget—including the ¥685 billion for the jusen resolution—was enacted in early May. but the disbursement of the jusen allocation was frozen until related legislation is passed. On June 18, 1995, the financial reform bills passed the Diet. In the meantime, additional burden sharing by the private sector is being sought by the government to reduce this politically unpopular public contribution. Market participants agree, however, that if confidence and transparency in the financial system are to be maintained during the resolution process, it is important that the financial reform plan is implemented as planned initially and that the burden-sharing rule becomes transparent.
The banking system in France remains under pressure from weak loan growth and the decline in property markets. After some delay, banks have in the past year moved to shore up their reserves against property-related loan losses, including extraordinary measures by two banks to rid themselves of real estate exposure—Banque Indosuez sold a large portion of its portfolio to its parent company, while Compagnie Financière de Paribas reported a net loss in 1995 due to aggressive write-downs and provisions against property loans. Credit Lyonnais, after two years of record losses and government support packages, earned a very small profit in 1995 for the first time in five years. However, much progress remains to be achieved in re-structuring the bank; the bank has warned that 1996 will be a difficult year, and the government has ruled out a third rescue plan. Meanwhile, new problems emerged at Credit Foncier, a specialized financial institution offering subsidized household mortgages and whose franchise was eliminated by the cancellation of the government’s housing loan subsidy program. The bank reported a net loss for 1995 that wiped out its equity capital, leading the government to announce that it would take all necessary measures to ensure that the bank would meet all payments of principal and interest and that the bank would be recapitalized. A credit line of F 20 billion from the Caisse des Depots et Consignations was arranged in the fall of 1995; a permanent solution is expected with the announcement of a restructuring plan in July 1996.
The Italian banking system is similarly suffering from the after-effects of the recent recession. Banks in the southern part of the country have had an especially difficult time recovering from the recession, in part because of cutbacks in government subsidies and the bank’s lending activities, which were not always linked to economic considerations. Banco di Napoli, a partially government-owned commercial bank, had its capital wiped owing to losses incurred in 1995 by providing against loan losses and restructuring costs. In November 1995, the Bank of Italy arranged a Lit 2.5 trillion loan from a group of 11 banks and the Cassa Depositi e Prestiti; and in March 1996 a decree authorized the Treasury to inject up to Lit 2,000 billion in capital into the Banco di Napoli (I percent of banking system capital plus reserves) and to privatize it in 1997. The difficulties at Banco di Napoli echo earlier problems at Banco di Sicilia, also a public law bank, which during 1992-95 had received capital injections from the Treasury totaling Lit 350 billion, and at Sicil-cassa, the second largest bank in Sicily. The banking system in southern Italy has been hard hit by a downturn in the economy and the restructuring necessitated by the liberalization of the industry. Liberalization has led to a rapid increase in the number of branches and shrinking intermediation margins, but it has been hampered by labor-market rigidities that impede cost cutting.
The Special Data Dissemination Standard provides a set of minimum standards or best practices to which member countries could voluntarily adhere for the provision to the public of comprehensive, timely, and accessible economic and financial statistics. The IMF will maintain a Dissemination Standard Bulletin Board on a World Wide Web site on which will be posted, for each country that subscribes to the standard, information regarding the data that are provided and their periodicity and means of dissemination.
Moreover, even if yen- or deutsche mark-denominated bonds can be swapped into the U.S. dollar, a depreciation of the local currency against the U.S. dollar would increase the debt burden in terms of local currency.
More detailed discussions of banking developments in most of these countries can be found in Annex II.
Banking problems first arose in 1991 in the wake of macroeconomic tightening and were initially concentrated in the large stateowned banks, which had high levels of problem loans. A World Bank financed program has been under way since 1992 to restructure the main five state-owned commercial banks. In addition, the state-owned development bank, Bapindo, was recapitalized late in 1995 with an injection of Rp 3.8 trillion of capital (of which Rp 1 trillion is new money) from the government and Bank Indonesia.
For further details on these banking systems, see the more detailed discusion in Annex II.
See International Monetary Fund (1994) for a discussion of the turbulence in global bond markets.
The unwinding of yen-funded leveraged positions in two-year U.S. notes, which may have exacerbated the rise in U.S. bond yields, was the only important instance this time of market pressure from hedge funds or proprietary traders.
These currency movements, and the macroeconomic developments preceding and following them, were discussed in greater detail in International Monetary Fund (1995b).
Knock-out options are used to hedge only against moderate fluctuations in exchange rates but not extraordinarily large movements, and are canceled when spot rates reach a specified knockout level.
Veteran currency speculator George Soros, when testifying before a U.S. Senate committee on April 13, 1994, likened the destabilizing impact of knock-out options to the destabilizing impact of portfolio insurance during periods of rapid equity price movements, which would put downward pressure on prices in a declining market and upward pressure on them in a rising market. Portfolio insurance was thought to have contributed to the U.S. stock market decline (Black Monday) in 1987.
U.S. hedge funds were borrowing short-term yen at near-zero rates and investing the proceeds in U.S. Treasury securities to earn the 400-500 basis point spread.
At the end of 1995, the net assets of open-ended mutual funds amounted to $2.8 trillion, or 93 percent of U.S. commercial bank deposits.
Bank for International Settlements (1996d).
The turnover in the OTC and exchange-traded derivative markets are not directly comparable, because the exchange-traded derivatives are not adjusted for double counting.
Structured notes are derivative contracts that are tailored in their pay-off structure to fit the needs of individual investors.
The Ministry of Finance definition of nonperforming loans includes (1) loans to borrowers in legal bankrupcy, (2) loans overdue for more than 180 days, and (3) restructured loans with a restructured interest rate being below the official discount rate at the time of restructuring.
