Back Matter
Author:
International Monetary Fund
Search for other papers by International Monetary Fund in
Current site
Google Scholar Close
Language:
English
Keywords:
WEFS; government securities market; market participant; x index; deutsche mark; investor base; Government securities; Bonds; Securities; Securities markets; Stocks; Global; Europe; Western Hemisphere; Middle East; Asia and Pacific
Download PDF (3.9 MB)

Abstract

Government securities are the backbone of world securities markets. Their turnover far surpasses that in any other financial market, except the global foreign exchange market. In recent years, government securities markets have undergone great changes. As many industrial countries ran sustained budget deficits in the 1980s and early 1990s, the size of government securities markets mushroomed. International competition for investors has forced governments to institute a wide range of reforms aimed at deepening liquidity in their markets and broadening their investor base. These reforms have attracted foreign institutional holders of government securities. With investors continually rebalancing their portfolios between securities of different countries, these markets have grown increasingly integrated across countries.

Annex I Government Securities Markets

Government securities are the backbone of world securities markets. Their turnover far surpasses that in any other financial market, except the global foreign exchange market. In recent years, government securities markets have undergone great changes. As many industrial countries ran sustained budget deficits in the 1980s and early 1990s, the size of government securities markets mushroomed. International competition for investors has forced governments to institute a wide range of reforms aimed at deepening liquidity in their markets and broadening their investor base. These reforms have attracted foreign institutional holders of government securities. With investors continually rebalancing their portfolios between securities of different countries, these markets have grown increasingly integrated across countries.

This annex provides an overview of major elements of government securities markets in industrial countries and of issues related to those markets. Although it focuses on the marketable obligations of central governments, obligations of other levels of government and nonmarketable claims of the central government are also discussed.

Recent Trends and Developments

The Increase in Debt Issues and Debt Outstanding

The 1980s and early 1990s have witnessed a significant rise in both the amounts of debt outstanding and in issue volumes. These increased stocks and flows have challenged the governments of industrial countries to manage their debt in a way that keeps the cost of debt service as low as possible. High real interest rates during much of this period heightened this challenge for many countries.

Table 4 shows the amount of debt outstanding in the larger industrial countries between 1980 and 1992, both in dollar terms and scaled by GDP in those countries. (Because of data limitations, the table includes debt held by the domestic central bank, and it also includes nonsecurities debt.) In most of the countries shown, debt expanded considerably as a percentage of GDP over the 1980s and early 1990s. Among the major industrial countries, U.S. debt jumped from 27 percent in 1980 to 52 percent in 1992, Italian debt from 55 percent in 1980 to 109 percent in 1992, and Canadian debt from 25 percent in 1980 to 54 percent in 1992. Debt also expanded in France, Germany, and Japan, although less dramatically (Chart 2). Rapid expansions of debt also took place in Belgium, the Netherlands, Spain, and Sweden.1

Table 4

Central Government Debt Outstanding

article image
Sources: International Monetary Fund, International Financial Statistics; Bank of Japan, Economic Statistics Annual (1991) and Economic Statistics Monthly, various issues; Bank of Italy, Economic Bulletin, No. 17 (October 1993); Bank of England, Quarterly Bulletin (November 1993); United Kingdom, Central Statistical Office, Annual Abstract of Statistics 1994; and Treasurer of the Commonwealth of Australia, Government Securities on Issue at 30 June 1992. Note: The United States, Germany, France, Italy, Canada (from 1990 onward), Spain, the Netherlands, Switzerland, and Belgium contain data for December 31; Japan, the United Kingdom, and Canada (through 1989) have data for March 31 of following year; Australia and Sweden show data for June 30.
Chart 2.
Chart 2.

Government Debt Outstanding in Major Industrial Countries

(In percent of GDP)

Sources: International Monetary Fund, International Financial Statistics; Bank of Japan, Economic Statistics Annual (1991) and Economic Statistics Monthly, various issues; Bank of Italy, Economic Bulletin, No. 17 (October 1993); Bank of England, Quarterly Bulletin (November 1993); and United Kingdom, Central Statistical Office, Annual Abstract of Statistics 1994.

Table 5 turns from stocks to flows of debt, by reporting the volume of gross issues of securities with a maturity of more than one year. Most of the countries in the table show a strong upward trend in the dollar-equivalent volume of securities issued, but they also show considerable fluctuations from year to year. Such fluctuations can be due to variations in net financing requirements, in the volume of older securities maturing, and in variations in the proportion of medium-term and long-term securities relative to other debt instruments. Exchange rate fluctuations can also affect these data. A comparison of different countries shows that the greatest gross volume of new issues is in the United States and Japan, followed by Germany and Italy.

Table 5

Gross Issues of Medium- and Long-Term Government Securities1

article image
Sources: Organization for Economic Cooperation and Development, Financial Statistics Monthly, various issues; and International Monetary Fund, International Financial Statistics and World Economic Outlook.

Central government issues of bonds of at least one year in maturity; public issues plus private placements.

From September 1991, government compensation bonds (inscribed government bonds) and subscription bonds are not included.

Including German Unity Fund.

Break in the comparability of data between 1985 and 1986 is mainly because the flows are recorded on the settlement date and no longer on the issue date. Issues using accounts for the Industrial Development Funds (Codevi) are excluded.

The 1993 data are for the first three quarters.

The 1991 data are for the first three quarters.

The lower half of Table 5 scales the gross issue data by GDP, and data for the major industrial countries are depicted in Chart 3. These figures give an indication of the demands that government debt issues place on the domestic financial infrastructure. Issues in most countries do not exceed 10 percent of GDP in a given year (Belgium and Italy generally are exceptions as were Japan and Spain in 1993).

Chart 3.
Chart 3.

Gross Issues of Medium- and Long-Term Government Securities in Major Industrial Countries

(In percent of GDP)

Sources: Organization for Economic Cooperation and Development, Financial Statistics Monthly, various issues; and International Monetary Fund, International Financial Statistics and World Economic Outlook.

Table 6 provides some data on issues of short-term securities (under one year in maturity). Chart 4 shows these data scaled by GDP for the major industrial countries. Short-term securities include treasury bills and similar products. The gross issue volume of these securities can be large because the debt stock is often rolled over several times a year.

Table 6

Gross Issues of Short-Term Government Securities1

article image
Sources: Organization for Economic Cooperation and Development, Financial Statistics Monthly, various issues; and International Monetary Fund, International Financial Statistics and World Economic Outlook.

Central government securities issued on the domestic market.

Marketable bills, which have maturities up to one year.

Bills.

Bills plus treasury bills in ECUs. The 1993 data are for the first three quarters.

The 1993 data are for the first two quarters.

Bills plus notes.

Treasury paper.

Notes.

Notes plus money market debt registered claims.

Chart 4.
Chart 4.

Gross Issues of Short-Term Government Securities in Major Industrial Countries

(In percent of GDP)

Sources: Organization for Economic Cooperation and Development, Financial Statistics Monthly, various issues; and International Monetary Fund, International Financial Statistics and World Economic Outlook.

In some countries, including Japan, such bills are issued mainly for short-term cash management. While only limited data are available for short-term issues, Table 6 shows that the United States and Japan are the largest issuers of short-term debt, as they are for longer-term debt. The volume of short-term debt issued by these countries is several times the volume of longer-term debt issued. This is also true in Italy and Canada. All of this makes it important for governments to have efficient mechanisms for issuing such debt into their money markets.

This expansion in both volumes issued and the outstanding stocks has fostered change in government securities markets through two channels. First, the rise in volumes increases the stakes associated with reform of these markets. For example, benefits from changing the method of issuing government securities are magnified as issue volumes expand. Second, the expansion in government debt outstanding tends to bring about changes in the nature of holders of the debt (see below).

The Identity of Holders of Debt

The composition of government debt holdings has changed considerably in recent years. In general, the shares of banks and individuals have fallen, while the share of foreigners has increased. To the extent that behavior differs across holders, these developments may have influenced the dynamics of government debt markets. They may also have affected other markets, notably the foreign exchange market.

Data on holders of government debt are unfortunately quite limited; more is known about holdings by financial institutions (especially banks) than about other holders. Table 7 reports data for 1980, 1986, and 1992 for those countries where this information is published; Table A1 provides more detailed data for the United States.

Table 7

Holders of Central Government Debt in Selected Industrial Countries1

(In percent of total debt outstanding)

article image
article image
Sources: Bank of Canada, Bank of Canada Review (Winter 1992–93 and Winter 1993–94), Table G5; Bank of England, Statistical Abstract: Part 1 (1993), Table 17.3; Bank of Italy, Ordinary General Meeting of Shareholders Report, various issues; Bank of France, Statistiques Monétaires et Financières Annuelles (1987) and Statistiques Monétaires et Financières Trimestrielles (May 1993); Deutsche Bundesbank, Monthly Report of the Deutsche Bundesbank, various issues; France, Directorate of the Treasury; Sweden, Central Bureau of Statistics, Statistical Abstract of Sweden (1984 and 1988) and Statistical Yearbook of Sweden 1994; United States, Department of Treasury, Treasury Bulletin (March 1988 and December 1993), Table OFS-2; and IMF staff estimates.

Data are as of March 31 for the United Kingdom; end-year data for all other countries.

Securities valued at par; some savings bonds are included at current redemption value.

Includes savings bonds and other securities.

Of Federal Government, Equalization of Burdens Fund, ERP Special Fund, Länder governments, and local authorities (for 1980 and 1986); of Federal Government, German Unity Fund, Debt-Processing Fund, ERP Special Fund, west German Länder governments, east German Länder governments, west German local authorities, and east German local authorities (for 1992).

Excluding public authorities’ mutual indebtedness; and holdings of Bundesbank.

Social security funds plus other.

Of central government securities; excludes Bank of France holdings.

1981 data are presented for 1980.

Total marketable central government debt excluding the holdings of Bank of Italy and Ufficio Italiano dei Cambi.

Deposits and loans fund, special credit institutions, and social security institutions.

Total market holdings of gilts.

Including building societies and other U.K. institutional investors.

U.K. public sector, industrial and commercial companies, public trustees and various noncorporate bodies, and other (including residual).

Government of Canada direct securities and loans (underlying outstanding amounts are in par value); excludes the holdings of securities by the Bank of Canada and in Government of Canada accounts.

Life insurance companies plus other insurance companies; IMF estimates for other insurance companies in 1992.

Trust and mortgage loan companies, investment dealers, local and central credit unions and caisses populaires, and other financial institutions.

Nonfinancial corporations, provincial governments, municipal governments, all other holdings of market issues by Canadian residents (residual), and Canada savings bonds; IMF estimates for all other holdings of market issues by Canadian residents (residual) in 1992.

Holdings of national debt excluding the holdings of Riksbank, other state institutions and funds, and the National Pension Insurance Fund.

Includes cooperative banks in 1992.

Savings banks, cooperative banks, and other financial institutions.

Local governments, joint-stock companies and economic associations, enterprises, other associations, private companies, and so on.

Holders abroad (1980 and 1986); and external debt in Swedish kronor and debt in foreign currency (1992).

One trend that emerges from Table 7 is that the share of government debt held by domestic banks has declined since 1980. In the United States, the share of bank holdings fell from 18 percent in 1980 to 10 percent in 1992 (although bank holdings have risen in the United States since 1990; see Table A1).2 In Germany, the share of bank holdings fell from 70 percent in 1980 to 54 percent in 1992. Marked declines are also evident in Italy and Sweden, and in France since 1986. In the United Kingdom and Canada, the shares of bank holdings have fluctuated, rising and then falling in the United Kingdom, while doing the opposite in Canada.

Insurance companies hold a small but rising proportion of government debt in the United States and in Italy. They are major holders of government debt in the United Kingdom, and this share rose considerably after 1986, so that by 1992 insurance companies held over one third of U.K. debt. Insurance companies held a moderate and declining share of government debt in Canada and Sweden.

The data for other domestic financial institutions are sparse. Pension funds appear to have increased their shares in the United Kingdom and Canada between 1980 and 1992.

A second notable trend is that the share of government debt held directly by domestic households has fallen between 1980 and 1992. In the United States, their share fell sharply in the first half of the 1980s (from 19 percent in 1980 to 10 percent in 1986) and has stayed roughly constant since then. Holdings of individuals and private trusts in the United Kingdom fell from 16 percent in 1980 to 9 percent in 1992. Household holdings rose in Sweden in the first half of the 1980s, but have fallen considerably since 1986.

The third trend is the sharp increase in holdings of government debt by foreigners. Foreign holdings of German public debt rose from 9 percent in 1980 to 26 percent in 1992. Foreign holdings of government debt also rose sharply in France (since 1986), the United Kingdom, Canada, and Sweden.3 In the United States, however, the share of debt held abroad has fluctuated, falling from 1980 to 1985, before rising again in the second half of the 1980s to a peak of 21 percent in 1989.

What are the implications of a relative decline in holdings by individuals and by banks and a relative increase by nonresidents? The decline in individual holdings of debt implies that institutional holdings of government debt are increasing (assuming foreign holdings are primarily institutional). Institutional holders are likely to turn over their debt positions more rapidly than households, although the effect of this on the dynamics of government securities markets is unclear. Such increased turnover could stabilize markets, since institutions probably react to small return differentials more than households do. Conversely, institutions may shift funds more rapidly from one market to the other, causing greater volatility. Institutions holding government debt are also subject to widely varying levels of supervision and regulation. Large swings in the price of government debt may lead to difficulties or even insolvency at some firms.

What about the “internationalization” of government debt holding? By opening a larger market for the debt, internationalization can potentially lower the cost of debt for a country. It has also led to pressure on governments to reform their markets to make them more attractive to international investors.

Internationalization has taken other forms for some issuers, including issuing debt on the Euromarket and in some foreign domestic markets. Such an approach has tended to be popular with smaller countries and with countries that carry relatively high debt levels. This trend has underpinned the issuance of sovereign global bonds, which can be sold simultaneously in several different markets.

Maturity and Currency Composition

Governments of the larger countries issue almost all their debt in domestic currency. Germany, Japan, and the United States do not issue any foreign currency debt. France has a program of issuing European currency unit (ECU) debt, but the bulk of its debt is in French francs. The United Kingdom issues the bulk of its debt in pounds sterling, but issues some ECU debt too. During the European exchange rate mechanism (ERM) crisis, it issued some debt denominated in deutsche mark on the Euromarket, in order to augment its foreign exchange reserves. Canada issues most of its debt in Canadian dollars, but issues debt in U.S. dollars to finance its foreign exchange reserves, and has issued previously in yen and deutsche mark.

Italy is alone among the seven major industrial countries in having an extensive foreign currency borrowing program. It issues Euromarket debt in a variety of currencies, and has recently issued global bonds in U.S. dollars and yen. Italy also issues domestic debt in ECU, on top of a well-developed domestic borrowing program in lira.

While data on maturity structure are limited, Table 8 gives a summary for six countries.4 In the United States, maturities lengthened between 1980 and 1992, with most of this lengthening occurring before 1986. The proportion of debt of less than five-year maturity decreased from 81 percent in 1980 to 71 percent in 1992, while the proportion of debt of greater than ten-year maturity increased from 11 percent to 16 percent. For Germany, data are only available for total government debt, a broader category than central government debt. Like the United States, Germany has engineered a considerable lengthening in maturities. The proportion of debt of less than three-year maturity decreased from 44 percent in 1980 to 26 percent in 1992, while the proportion of debt longer than five years rose from 34 percent to 45 percent. In contrast to the United States, however, Germany has very little debt with maturities greater than ten years.

Table 8

Remaining Maturity of Domestic Currency Debt1

(As cumulative percent of total value of outstanding issues)

article image
Sources: Bank of Canada, Bank of Canada Review, various issues; Deutsche Bundesbank, Capital Market Statistics (February 1993) and Statistical Supplement to the Monthly Reports of the Deutsche Bundesbank—Series 2: Securities Statistics (February 1981 and February 1987); France, Directorate of the Treasury; Japan, Ministry of Finance; United Kingdom, Central Statistical Office, Annual Abstract of Statistics, various issues; and United States, Department of Treasury, Treasury Bulletin, various issues.

All debt valued at nominal values, not current market value.

Marketable central government debt held by private investors.

Domestic central government bonds excluding financing bills and grant-in-aid bonds.

Public bonds including bonds issued by the Federal Government, and of German Unity Fund, Currency Conversion Equalization Fund, ERP Special Fund, Treuhand privatization agency, Länder governments, local authorities, Federal and east German Railways, and Federal Post Office.

Marketable central government debt.

Market holdings of central government and government-guaranteed marketable securities.

Marketable unmatured central government securities, including government-guaranteed securities.

The maturity structure of Japanese debt shortened between 1980 and 1986. Since 1986, however, Japan has initiated a slight lengthening of maturities, especially in the very long end (more than 15 years) of the market.

The United Kingdom shows a different pattern. The proportion of debt below five years in maturity has remained roughly constant, but the proportion in the 5–15 year range has risen at the expense of longer-term debt. In March 1993, 72 percent of U.K. debt had maturities below 15 years, compared with 58 percent in March 1981. Like the United Kingdom, Canada also has experienced a shortening of maturities. The proportion of debt less than three years in maturity rose from 53 percent in 1980 to 64 percent in 1992. The proportion of debt with maturities above ten years fell from 27 percent to 14 percent over the same period.

While comparable data are not available for Italy, its effective maturity structure is short relative to other major industrial countries. About 27 percent of Italian debt is in Treasury bills of maturity one year or less, with another 35 percent indexed every six months.

Governments also issue a variety of debt that is not fixed in terms of future cash flows. One type is debt indexed to inflation. Among the seven major industrial countries, the United Kingdom is an active issuer of such debt, and Canada began a program in 1991. Some governments, including Italy and Canada, also issue floating rate debt, which has the interest rate indexed to a short-term interest rate.

Interest rate swaps can also be used to adjust the maturity composition of debt. Canada has an active program of swapping its debt. The Government enters swaps in which it agrees to make floating rate payments to its counterparty, in exchange for receiving fixed rate payments that match the Government’s liabilities on some of its bonds.

Another tool related to maturity management is strips. These are programs that separate a coupon bond into separate zero-coupon securities for each coupon interest payment and for the final maturity payment. Strips are arranged at the initiative of the market, rather than as a decision of the government. However, government policies can facilitate strips. While strips do not change the maturity composition of the government’s debt, they do allow holders to choose more precisely the cash-flow profile of their debt portfolio, and thereby increase demand for the government’s debt.

Market Structure

Primary Dealer Systems

In many countries, the authorities have designated a group of financial firms as the principal intermediaries in the government securities market. These firms receive a set of privileges in this market, in exchange for a set of obligations. The nature of these privileges varies greatly from country to country; the most common obligation is to make a secondary market in government debt. These firms sometimes operate under a special supervisory regime.5

The United States has long had a system of primary dealers in its government securities market. The Federal Reserve Bank of New York designates these firms, which currently number 39. In the wake of the Salomon scandal in 1991, U.S. authorities reduced the extent of both the privileges and the obligations of primary dealers6—in part to increase the competitiveness of government securities markets.7

Primary dealers in the United States are required to be active in both the primary and secondary markets for government securities. They must make reasonable bids in every Treasury auction. The Federal Reserve Bank of New York used to require each primary dealer to account for at least 1 percent of the total transactions of all primary dealers with customers, but it abandoned this rule in January 1992. The rule reportedly had forced some smaller primary dealers to offer steep commission discounts to their customers. Primary dealers, however, still must maintain a presence in the secondary market, because the Bank requires them to “make reasonably good markets in their trading relationships with its trading desk and provide the trading desk with market information and analysis that may be useful to the Federal Reserve in the formulation and implementation of monetary policy.”8

In October 1991, the U.S. Department of Treasury eliminated two privileges that primary dealers had enjoyed. One was the exclusive right to place bids for customers in Treasury auctions. The other was exclusive access to Treasury announcements of its borrowing plans, which the Treasury now disseminates to the entire market.

Primary dealers enjoy two remaining privileges. They have exclusive access to some of the inter-dealer broker screens, although this exclusivity also is eroding.9 They also serve as the Federal Reserve Bank of New York’s counterparties in its open market operations. Primary dealers thus are the main conduit through which the Federal Reserve conducts its monetary policy. It is sometimes argued that such a position gives primary dealers knowledge of monetary policy that other market participants do not have.

Japan does not have a primary dealer system along the U.S. lines. However, it does bestow a special role in the government securities market to a large group of firms. These firms make up the issuing syndicate for Japanese Government bonds (JGBs). Over 900 firms are members of the syndicate, including banks, securities firms, and insurance companies. Additionally, a much smaller group of about 35 firms negotiates the terms of syndicated government bond issues with the Ministry of Finance. In contrast to the U.S. system, the syndicate has a special role only in the primary market, but not in the secondary market or in central bank monetary operations.

The primary obligation of syndicate members is to accept their allocation of bonds. Forty percent of the biggest Japanese issue, the ten-year bond, is syndicated, although most other bonds—and the rest of the ten-year bonds—are auctioned. The syndicated bonds are sold at the weighted-average price from the auction.10 In exchange for accepting their share of syndicated bonds, syndicate members enjoy two benefits. First, they have exclusive access to auctions for government bonds. Second, they receive a commission from both the syndicated and auctioned portions of the ten-year bond.

Germany sells its bonds through a consortium of 109 members. Most long-term bonds in Germany are sold through a combination of a syndicated distribution through the consortium with an auction and subsequent tap sales by the Bundesbank.11 In contrast to primary dealer systems in other countries, membership in the bond consortium appears to carry no significant costs. The primary role of consortium members is to accept their share of syndicated bonds at the price determined by authorities, but there is no automatic mechanism to ensure that this price is close to the auction price. Since a selling commission is paid on these bonds (but not on those allocated through the auction or tap), these shares are probably more a privilege than an obligation. Shares are decided by the Bundesbank. In addition to their syndication shares, consortium members enjoy other privileges: they have exclusive access to government bond auctions; they hold a regular meeting, at which they exchange market information; and they have special access to Bundesbank repurchase agreement (repo) operations in order to finance their government security purchases.

The French Treasury adopted in 1987 a system of primary dealers known as Spécialistes en Valeurs du Trésor (SVTs). These firms were assigned duties in both the primary and secondary markets; SVTs, however, have no special role in central bank monetary operations. The Bank of France carries out monetary operations through a different, separately designated group of firms, although some firms are members of both groups.

The 18 current SVTs have specific, quantitative obligations in the government security auctions and in the secondary market. During the course of a year, each SVT must purchase 2 percent of the total amounts auctioned of each of the three principal types of securities: BTFs (short-term), BTANs (medium-term), and OATs (long-term).12 SVTs must be considerably stronger than this threshold in at least one of these types of securities, because the average of their annual shares in the three types of securities must exceed 3 percent. In the secondary market, each SVT must account for at least 3 percent of the transaction volume in each type of security. SVTs must also provide quotes upon request for any government security and post continuous screen quotes for the most active issues. In addition to these auction and market-making requirements, the Treasury requires SVTs to work to place securities with final investors, both in France and abroad.

In exchange for these obligations, SVTs enjoy two specific privileges. One, which is unique to the French system, is a provision for submitting noncompetitive bids. SVTs are permitted to purchase additional amounts of an auctioned security at the weighted-average auction price for up to one day after the auction. The amount an SVT may buy at this price is proportional to its average purchases in the last three auctions for the same type of security, subject to a ceiling for all SVTs of 15 percent of the issue. The second privilege is that SVTs have sole access to an interdealer broker. SVTs are also expected to advise the Treasury on market conditions and issuance policy. Such a relationship with the Treasury is likely to benefit the SVTs as well as the Treasury.

In the United Kingdom, the primary-dealer equivalent is the gilt-edged market makers (GEMMs). The GEMM system has operated since the Big Bang in 1986. In exchange for a variety of special privileges, GEMMs are required to make markets for gilts—essentially all marketable medium- and long-term U.K. Government securities. GEMMs do not have any special role in the market for short-term U.K. Treasury bills, and their role in Bank of England monetary operations is limited.

The principal obligation of GEMMs is to make continuous markets in gilts. Each GEMM must make a market in the full range of gilts, including index-linked gilts, which tend to have lower transaction volume than conventional gilts. They are also expected to participate regularly in gilt auctions, though this requirement receives less emphasis than the market-making requirement. The Bank of England does not measure the GEMMs’ secondary and primary market conduct against any explicit numerical standards.

GEMMs receive a package of privileges in exchange for their market-making activities. They have the exclusive right to deal in gilts with the Bank of England, and they may borrow from the Bank. They also have the exclusive right to borrow gilts, which they do through special stock exchange money brokers. These gilts are lent by other GEMMs, as well as by investors. Finally, GEMMs have exclusive use of special interdealer brokers. GEMMs do not have any special privileges relating to gilt auctions, except that only they may submit bids by telephone, which may give them some advantage in reacting to late-breaking market developments.

Primary dealer systems such as those described in this section clearly involve constraints on the workings of markets. These systems restrict entry of firms into parts of the government securities market and tie the participation of firms in some parts of the market to participation in other aspects of the market. Their justification is that the obligations of primary dealers satisfy public goals that would otherwise not be met. The most prominent of these goals is to maintain the liquidity of secondary markets, which arguably is increased by requirements that primary dealers make continuous markets. Such an arrangement resembles that in other financial markets, such as stock markets, where agents are charged with making markets in return for some privileges.13 Another justification for primary dealer systems is that such systems serve as devices for authorities to regulate and supervise the firms that are the major players in the domestic government securities market and which play an important role in other domestic financial markets as well.

Issuance Techniques

The methods used by many countries for issuing government securities have changed in recent years; in brief, more securities are being auctioned and fewer are being issued through syndicates. Different auction techniques have also been under consideration, especially in the United States.

Most debt securities are issued by one of four methods: auctions, direct syndications, tap issues, and underwritten syndications. The last of these methods is used for Euromarket and global issues, while the other three are used for domestic issues. Tap issues by the central bank are often used to sell parts of issues that are also sold through auctions or syndications.14 Table 9 lists the methods used by governments of the major industrial countries to sell their domestically issued government securities. Of these countries, only Italy is a regular issuer on international markets. Smaller industrial countries use techniques similar to those used by the major industrial countries. Several of these smaller countries are active borrowers on Euromarkets and other international markets and use underwritten syndications for these issues.

Table 9

Techniques for Issuing Domestic Government Securities in Major Industrial Countries

article image
Sources: Bank of England; Italy, Ministry of the Treasury; and Bröker (1993).

Auctions are used for the bulk of government debt issued worldwide. There is a variety of types of auctions. Auctions for government securities usually are sealed bid, where the government receives all bids in a single batch before the auction. With improvements in communication and information processing, however, it may also be possible to conduct interactive auctions, in which the government gradually raises prices, soliciting bids at ascending prices, until supply equals demand. U.S. policymakers have considered implementing such a system. Among sealed-bid systems, the big choice is between uniform-price and discriminatory (multiple-price) systems.15 In a uniform-price system, all units of the security are sold for the same price, generally the price of the lowest successful bidder. In a discriminatory system, each bidder pays the price bid.

For their auctioned debt, Canada, France, and Germany exclusively use discriminatory auctions, while the other major industrial countries use a mix of auction techniques. In the United States, the authorities have begun to use uniform-price auctions for sales of some Treasury notes on an experimental basis, after some incidents of manipulation in discriminatory auctions. Initially, this was a one-year experiment from September 1992 through August 1993, but the experiment has been extended for an additional year. Discriminatory auctions are also the more common auction technique in countries other than the major industrial countries.

Auctions often embody incentives to encourage bidding. In Germany, shares in the syndicate for issuing bonds are adjusted periodically, with the amounts syndicate members have purchased in auctions influencing their syndicate shares. Some incentives result in options conferred on some bidders. The noncompetitive bids available to SVTs in France are an example of such an option.

Germany and Japan have made the most prominent use of syndicates for domestic bond issues (see section on Primary Dealer Systems above). Other countries, including Canada and France, have virtually ceased using syndicates in recent years.16

Some researchers have criticized the use of discriminatory auctions and have recommended the use of uniform-price auctions instead. Although the former appears to increase revenue through price discrimination, critics argue that these auctions exacerbate the winners’ curse phenomenon, in which successful bidders suffer from having overestimated the value of the securities, and therefore, that bidders reduce their bids accordingly. It is unclear whether this reduction outweighs the increased proceeds from price discrimination, and therefore it is unclear which technique yields higher proceeds for the government. However, another consideration is that uniform-price auctions do appear to reduce risk faced by bidders. There has also been considerable debate in the United States as to which technique is less prone to manipulation, such as market cornering and collusion among bidders.17

A broader question concerns the advantages of auctions compared with other techniques. This issue has not received as much attention as the debate on auction techniques. At first blush, auctions appear to be a more efficient mechanism to elicit market demand than other techniques. However, the risk inherent in the winners’ curse problem may reduce this efficiency, although the asymmetric information between buyers that drives winners’ curse in auctions must still operate in other systems and may cause other inefficiencies.

Organization of When-Issued Markets

A when-issued market is a market in securities not yet issued. Settlement normally occurs on the same date that deliveries are made in the primary market. In the United States, for example, when-issued trading begins when an auction is announced, up to ten days before the auction.18 The primary economic role of the market is price discovery. The when-issued market provides information on market demand to participants in an auction for a security. The market also plays a risk-shifting role; investors with a demand for the particular security can lock in the yield on this security in advance of the auction, and the risk of interest rate movements is shifted to their counterparties in the when-issued market. These counterparties, who have a short position in the when-issued market, are often bidders in the auction.

A when-issued market exists in all the major industrial countries other than Germany and Japan. Trading starts in these markets when the particulars of an auction are announced, including the amount to be issued and the maturity. This period ranges from as much as ten days before the auction (in the United States) to as little as two days (in France). Settlement of when-issued trades occurs when the auctioned securities are distributed. This post-auction settlement period ranges from one day (in the United Kingdom) up to over three weeks (in France).

Activity on these markets varies considerably from country to country. While volume statistics are not available, the U.S. market appears to be the most active. The when-issued security in the United States becomes the “on-the-run” (i.e., the benchmark) issue. This security is the most recently issued of its maturity range and is the most heavily traded and liquid issue. The when-issued market is also reported to be heavily used for gilts in the United Kingdom and for government securities in Canada. The Italian when-issued market, on the other hand, does not experience heavy trading.

The when-issued market came under scrutiny in the United States because of several incidents of manipulation in 1991. The market was manipulated when one dealer amassed a dominant position in the auctioned security and forced those dealers with short positions in the when-issued market to pay abnormally high prices to cover these positions. This is one example of what is known as a “short-squeeze.” In the U.S. case, rules on the amount one dealer may purchase at an auction were circumvented in order to accomplish this squeeze. The possibility of such squeezes raises risks to participants in the when-issued market and therefore reduces the liquidity and efficiency of the market, in turn reducing the information available to bidders in the primary market. This reduced information may increase the likelihood of winners’ curse and increase debt costs of the government.

There are two important unresolved questions regarding when-issued markets. The first is what are the benefits to the price discovery process of the when-issued market. Preauction trading in the when-issued market permits at least the partial aggregation of private information on the value of the issue. This reduces the asymmetry of information at the auction and therefore reduces the likelihood of winners’ curse. In turn, that is likely to make bidders bid more aggressively and increase proceeds to the government. However, the when-issued market may simply shift risk from auction participants to when-issued participants. The when-issued market may serve to reallocate risk among market participants rather than reduce it, although it will reduce the amount to which the government must compensate auction participants with lower issue prices.

The second question surrounding the when-issued market is how susceptible is it to manipulation, as discussed above. Because of the long settlement periods relative to the cash market, when-issued positions can be taken at lower cost, and it is therefore less expensive to build up market power in such a market. One possible solution to such a risk of manipulation might be to build alternative delivery options into the contract, so that if a particular security were scarce, those with short positions in when-issued securities could deliver either cash or some alternative security.19 Improved reporting of price and trade information also might reduce manipulation, as might improved supervision of the market.

Liquidity in the Secondary Market

The function of a secondary market is to provide liquidity. It is particularly important that markets are able to provide liquidity in times of stress, when buy or sell orders increase. A variety of features in secondary markets contribute to liquidity.

One rough gauge of the liquidity of secondary markets is the volume of transactions in the market. The higher this volume, the more market-makers are compensated for the fixed costs of serving their role. This in turn is likely to increase the financial resources these market-makers have at their disposal, while also drawing additional firms into the market.

Table 10 reports transaction volume in cash government securities markets for five major industrial countries.20 Generally, transaction volume in these markets is very large and has grown tremendously since 1980. The volume of transactions involving primary dealers in the United States grew from an average of $14 billion a day in 1980 to $120 billion in 1993. Volume in the Japanese market has grown even more sharply, from $1.4 billion a day in 1980 to $58 billion in 1993. Volume in the U.K. market is considerably smaller, but has also risen substantially since 1980, especially after the Big Bang in 1986. Using 1992 figures, 3 percent of U.S. debt was traded on an average trading day, compared with 2 percent for both Japan and the United Kingdom.

Table 10

Transaction Volume in Government Securities Market1

(Daily average in billions of U.S. dollars)

article image
Sources: Bank of Japan, Economic Statistics Annual and Economic Statistics Monthly; Board of Governors of the Federal Reserve System, Federal Reserve Bulletin; Deutsche Bundesbank, Capital Market Statistics (November 1993); France, Directorate of the Treasury; International Monetary Fund, International Financial Statistics; Japan, Ministry of Finance; and United Kingdom, Central Statistical Office, Financial Statistics.

Data adjusted for double counting. Daily transactions volume for Japan, Germany, and France are estimated by the IMF staff from the annual volume data.

Primary dealers transactions of Treasury securities.

Data through 1989 from eight domestic stock exchanges; since 1990, from Tokyo Stock Exchange.

Stock exchange turnover for bonds of the Federal Government, Federal Railways, and Federal Post Office.

Secondary market transactions in central government securities, including repurchase agreements.

Stock exchange transactions of U.K. Government securities.

Transaction volume in most markets tends to be concentrated in benchmark securities. In most countries, a benchmark is the most recently issued security of a particular maturity. Usually the list of benchmarks in different countries includes a ten-year bond, which forms the basis for international comparisons of yields. Nonbenchmark securities are traded less frequently than benchmarks and are more likely to be held in the portfolios of longer-term investors. Because benchmarks are traded frequently in liquid markets, market yield information is more reliable and up to date than for other securities. For this reason, markets use benchmarks to price other fixed-income securities. For example, a ten-year corporate bond in U.S. dollars might be priced at, say, a 100 basis point spread to the ten-year Treasury bond with the same maturity, or a French franc bond at a spread above the French Government bond (known as the OAT). Typically, benchmarks also carry a somewhat lower yield than similar nonbenchmark securities; this difference in yield can be thought of as the market price of the greater liquidity of benchmark issues.

In Japan, trading is heavily concentrated in the benchmark Japanese Government bond (JGB). While this benchmark is a JGB with an original maturity of ten years, the benchmark is often not the most recently issued JGB. Instead, the benchmark bond is chosen by securities dealers. Requirements for benchmark status are that the bond have a sufficient volume outstanding, between eight and ten years remaining to maturity, and a price near par (which means that the coupon rate is close to the yield to maturity). Trading in JGBs is heavily concentrated in the benchmark, with about 80 percent of total trading concentrated in this security. Bid-ask spreads are also considerably narrower for benchmarks than for nonbenchmarks, which is a sign of the greater liquidity of the benchmarks.21

Many governments have adopted a practice known as “fungibility” which involves reopening an existing security, instead of issuing a new security. By increasing the amount outstanding of particular issues, markets have available a larger inventory of the security to trade, which increases liquidity. This practice is particularly common in France, which since the mid-1980s has followed a policy of adding frequently to existing securities, while issuing new securities less often. For example, France issues a new ten-year bond only once a year, but auctions additional tranches of this bond every month. Belgium makes heavy use of this technique for its “linear bonds,” and Canada, Germany, Japan, the Netherlands, and the United Kingdom also reopen issues.22

The bulk of volume in government securities is traded over the counter (OTC), rather than on exchanges. This is true in all the major industrial countries except the United Kingdom. This exception is perhaps due to the different nature of the U.K. stock exchange, which relies on dealers’ posting quotes, rather than the order-matching processes more common in exchanges. In most countries, government securities are listed on exchanges, partly because some institutional investors are required to buy on exchanges, even though the bulk of volume occurs off the exchanges. Significant German trading takes place both on the domestic stock exchanges and OTC.

In many markets, interdealer brokers play an important role. They specialize in gathering price quotes from government securities dealers and posting them on an anonymous basis on screens to which all the dealers have access. When dealers trade, interdealer brokers do one of two things. Some interpose themselves between each side of a trade, so each counterparty actually trades with the interdealer broker. In this case, the identity of the counterparties is never revealed. Other interdealer brokers reveal the names of each counterparty to the other, and then the two counterparties complete the trade themselves. In most markets where inter-dealer brokers operate, the bulk of interdealer trades go through them. Interdealer brokers are also active in the offshore market based in London.

Offshore markets can be an important source of liquidity for government securities markets. This market is particularly active in London, where a considerable share of the volume of trading in French and German Government securities takes place.

Futures markets also contribute to the liquidity of government securities markets. Exchange-based futures contracts on government bonds from all the major industrial countries are actively traded. There also are contracts on short-term government securities from several countries. Additionally, options on many of these futures or on the underlying bonds are traded through the same exchanges. Table A2 lists contracts that are traded. In terms of contracts traded, the U.S. Treasury bond futures on the Chicago Board of Trade are the most active of all financial futures listed. Futures on U.S. Treasury bonds also are traded in Tokyo and London, although in much lower volume, making this market a 24-hour market.

Futures and options markets provide a method for market participants to adjust their exposure to government securities markets. Such markets contribute to liquidity in the underlying cash markets in two ways. First, they permit market-makers to hedge cash positions and to adjust those hedges relatively quickly. Some of this liquidity may be illusory, however, since futures markets involve two opposite positions, and the short position may be hedged or arbitraged in the cash market. However, this leads to the second way futures markets contribute to liquidity: futures markets generate trading volume in the cash market, through this arbitraging and hedging. This additional trading volume helps attract additional market-makers in the cash market. Futures markets also enhance market transparency, as contracts are traded on exchanges with published, real-time volume and trade information.

Countries vary in their systems for clearing and settling government securities transactions. Of the major industrial countries, only the United Kingdom and the United States settle transactions in long-term government securities (bonds) as quickly as the next day; other countries exhibit longer delays, although some clear short-term securities more quickly. In the United States, some Treasury bill transactions settle on the same day as the trade. Most government securities in major countries are held in a computerized book-entry system, so that there is no cumbersome physical delivery. In the United States, most government securities transactions are routed through the Government Securities Clearing Corporation (GSCC). The GSCC nets trades, and net payments and securities transfers are made through the wholesale bank payments system, Fedwire, which is connected to the book-entry system run by the Federal Reserve. This netting reduces the volume of payments made over Fedwire.

The time between trade execution and settlement exposes participants to the risk of nondelivery of securities or cash. Having a book-entry system for securities can make settlement considerably more efficient and reduces risk by facilitating delivery-versus-payment settlement. A trade netting system can also contribute to efficiency, by reducing the number of payments brokers and dealers need to make. Links between domestic clearing systems and international clearing systems, especially Cedel and Euroclear, facilitate international transactions in a country’s debt.

Markets vary according to the degree to which information on the market is disseminated. Government securities typically have a lower degree of transparency than stock markets, where real-time price and trade information is available to the public. Since most government securities transactions occur OTC, there is no exchange through which to disseminate information. In many government securities markets, market-makers post quotes and transact through interdealer broker screens. Typically, only market-makers have access to these screens. Thus, agents outside the circle of market makers have less information about the market than insiders do.

Organization of Repurchase Agreement and Securities Lending Markets

Repurchase agreements (repos) facilitate leveraged long and short positions in securities. A portfolio of long positions in securities can be financed in part by executing repos on some of the securities in the portfolio. The trader that held the original portfolio and executed the repos remains exposed to price movements on the securities used in the repos. Conversely, a trader executing a reverse repo can immediately sell the security acquired through the reverse repo, thereby creating a short position which it must cover in the future to complete the repo.23 The ability to engage in these transactions can facilitate inventory management by securities dealers, and thus reduce the cost of market-making and increase liquidity in cash markets. Securities lending is a somewhat more generalized term than repo. Repos are essentially a way to lend a security with the cash lent serving as collateral, but securities may also be lent either with other securities as collateral or without any collateral.

Repo and securities lending facilities are often cited as a key element in a liquid bond market. They allow dealers to take long and short positions in a flexible manner, buying and selling according to customer demand on a relatively small base of capital. Both repo and securities lending facilities allow dealers to acquire specific securities demanded by customers, without having to find another customer willing to sell the bonds. These transactions are often facilitated by brokers who specialize in these markets. Repos also contribute to liquidity by allowing nondealer investors to take positions in securities without putting up much capital. The market can therefore react to any price movements in securities markets, keeping prices close to their equilibrium.

In the United States, the repo market for government securities is very large. As of October 20, 1993, primary dealers had open repo contracts in government securities outstanding of $852 billion. This figure includes repos both in Treasury securities and agency securities, including mortgage-backed securities. This was equivalent to 29 percent of U.S. marketable debt outstanding at the end of the third quarter of 1993. Overnight and continuing repos ($466 billion) were more common than longer-term repos ($386 billion).24 The primary dealers also had $670 billion in reverse repos out-standing. Securities lending by primary dealers was smaller, with a total of $184 billion borrowed and $5 billion lent.25

In the mid-1980s, several government securities dealers in the United States failed. These failures resulted in the failure of several savings and loans, and precipitated the collapse of the Ohio state deposit insurance company. A factor in the failure of these dealers was that the dealers had executed repos with customers, but had failed to place the securities involved in their customers’ accounts. Instead the dealers used the same securities to execute multiple repos. These secondary dealers had been unregulated and unsupervised, since their sole business was government securities. In response to these events, the Government Securities Act of 1986 became law, placing government securities dealers under the regulation of the U.S. Treasury and the supervision of the Securities and Exchange Commission (SEC).

Repo markets in other large countries are substantially smaller than the U.S. market. In many countries, repo and securities lending markets are inhibited by regulatory restrictions or taxes. In Japan, repo transactions are known as gensaki. The market for gensaki on short-term government securities is active and accounts for a reported 28 percent of transaction volume in government securities. The main borrowers of funds in this market are securities firms. Gensaki in longer-term securities are less common, however. The reason for this is that these repos are considered bond transactions and therefore are subject to Japan’s tax on securities transactions. This transactions tax also tends to lengthen the maturity of gensaki relative to U.S. repos.26 In contrast to most U.S. repo transactions, gensaki do not permit the borrower of funds to substitute one security for another as collateral. Total repo contracts outstanding in Japan (including those on nongovernment securities) totaled ¥10.6 trillion ($101 billion) at the end of September 1993, which was 9.8 percent of total Japanese Government debt outstanding.27 While these numbers are substantial, they show less use of repo transactions than in the U.S. market. Bond lending was authorized in Japan in 1989. This type of transaction is not subject to the transaction tax, so it has become an active market and a source of market liquidity in the bond market. Bond lending can be as short as overnight, in contrast to Gensakis, for which the transaction tax makes such a short-term transaction prohibitively expensive.

In Germany, banks borrowing funds through repo transactions face a reserve requirement on these borrowed funds. As a result, a domestic repo market essentially does not exist. In contrast, a repo market in German Government securities flourishes in London, with most of the volume concentrated on the long-term bonds (Bunds). German firms reportedly are active in this market, which is linked to trading in futures contracts on German Government securities.

French Government securities repos are executed both in the domestic market and offshore. There are no tax impediments, although legal uncertainties have hampered the development of the domestic market; the French Government passed a new law intended to eliminate these uncertainties at the end of 1993. As of December 1993, there were about F 200 billion in domestic repos outstanding, and about twice that amount in international repos. Italy also has an active domestic market in repos, especially in longer-term government securities. At the end of 1993, there were Lit 92 trillion in repos outstanding, excluding interbank repos.

In the United Kingdom, a market for repos in gilts between private parties does not exist, and there is only a limited market in gilt borrowing. Only gilt-edged market-makers may borrow gilts, and they must do so through the stock exchange money brokers. The Bank of England began in January 1994 to use gilts repos as a regular open market device. In Canada, there is an active repo market in government bonds. This market has grown rapidly since 1992, after the elimination of a tax on cross-border repos. U.S. investors reportedly are active in the market and have driven much of the growth of the market.

A variety of issues arise in these markets. One is that the legal validity of repurchase contracts has been in question in some jurisdictions. This makes the status of repos uncertain under bankruptcy; while repos have the features of a collateralized loan, bankruptcy procedures may not recognize this. The tax treatment of repos can also be an issue. Such problems stem from the fact that repos combine short-term collateralized lending with a loan of a security, but not formally as a loan, but rather as a sale with a future contract to reverse the sale. Issues related to the transfer of the title to the security can also come up, especially when a dealer conducting a repo holds the security in custody for its counterparty.

Another technical issue concerns the settlement of repo and securities lending transactions. In principle, these transactions can use the same clearing and settlement facilities as regular cash transactions in securities. However, the complexity of these transactions complicates matters.

Finally, these transactions raise potential systemic concerns. Repos in particular are used by some parties, including hedge funds, to take large, leveraged positions in government securities. Dealers essentially lend to these funds using the repo as a collateralized loan. Repos typically are relatively short term (under a week), with the initial market value of the underlying security exceeding the amount of cash loan by small margin. This margin protects the dealer from fluctuations in the value of underlying security. If, however, the volatility of market prices unexpectedly increases, dealers could be left with insufficiently collateralized positions against borrowers.

International Debt Markets

The Distinctions Between Domestic Debt and Eurodebt

Governments tend to issue the bulk of their debt on their own domestic market, but many also issue on Euromarkets. In addition, some governments issue on the domestic markets of foreign countries, although less frequently. Countries normally use international markets for their foreign currency issues, while issuing domestic currency debt at home. In most major currencies, regulatory barriers generally separate the Euromarket from the domestic market. The result is that domestic residents of a country tend to hold more of domestic debt issued in that country, while nonresidents hold more of the Eurodebt. Eurodebt is typically issued in bearer, nonregistered form, which makes it difficult for the authorities to monitor and tax the holders of the debt.

While many countries auction their domestic government debt, Eurodebt normally is issued through an underwriting syndicate. Countries issuing in the domestic market of another country (most often in the United States) also use a syndicate. The nature of the syndicate varies across markets. Euro-syndicates include sellers able to place the debt internationally, while syndicates for domestic issues place more domestically. This distinction typically dictates the nationalities of members of the syndicates.

Although a Eurobond usually is listed on some exchange, most transactions occur OTC. The settlement of Eurobonds typically takes place through two international clearing agencies, Euroclear and Cedel. Domestic bonds most often clear through their domestic clearers, although international trades of these bonds sometimes use international clearers.

In the United States, SEC rules serve to differentiate bonds issued on the domestic market from Eurobonds. Bonds issued in the United States must be registered with the SEC.28 Eurobonds are exempt from SEC restrictions, but consequently they may not be sold to U.S. investors on the primary market. The SEC imposes a seasoning period of 40 days from the date of issue of a Eurobond, during which time the bonds may not be sold in the United States. Although U.S. investors may purchase Eurobonds after the seasoning period is over, the bulk of Eurobonds tend to be held outside the United States.

There are no significant regulatory barriers between deutsche mark bonds issued by foreigners on the domestic German market and Euro-deutsche mark bonds. Bonds that are sold internationally are designated Eurobonds, while others are considered domestic issues. The Bundesbank requires all deutsche mark bonds to clear through the domestic German clearing system, although international trades in Euro-deutsche mark bonds may also clear through international clearers. Euro-deutsche mark bonds arose in the 1960s as a result of a withholding tax imposed on domestic deutsche mark bonds held by foreigners, but this withholding tax has since been abolished.29

Global Bonds

As noted above, there is some separation between the Eurobond market and domestic bond markets. An issuer normally must choose on which market to issue. Demand for the bond is then constrained by the barriers between the markets. In the last few years, an instrument known as the global bond has been developed to overcome this segmentation.

The World Bank issued the first such bonds in 1989 and still remains the leading issuer of global bonds. It has issued in U.S. dollars, Japanese yen, and deutsche mark. Its first global bond issue in 1989 was in dollars, and it continues to issue regularly in dollars. It has made four issues in yen and has ¥ 900 billion in yen global bonds outstanding. Most recently, the Bank pioneered the deutsche mark global bond with a DM 3 billion issue in October 1993.

A number of sovereigns have also issued global bonds. Among the Group of Ten countries, Italy and Sweden have used this technique. Sweden issued a $2 billion U.S. dollar global bond in February 1993. More recently, Italy issued $5.5 billion in U.S. dollar global bonds in September 1993 and a ¥ 300 billion yen global bond in January 1994.

Global bonds combine SEC registration and U.S. clearing arrangements with separate clearing on the Euromarket. In some cases, there is also registration and clearing in other domestic markets, particularly in Germany and Japan for deutsche mark and yen issues. These bonds require arrangements for bridges between different clearing systems, so that the quantities of bonds in different markets can shift according to demand. If these bridges work smoothly, bonds can flow across jurisdictions easily. This enhances liquidity in the market, by allowing buyers from different jurisdictions to react to price fluctuations.

By allowing issuers to solicit demand for a variety of markets and to offer greater liquidity to investors, global bonds have the potential to reduce borrowing costs. While there does not appear to be systematic evidence on this, the World Bank has reported that the use of global bonds has reduced its interest cost of borrowing in U.S. dollars by about 18 basis points relative to the interest rates paid by agencies sponsored by the U.S. Government.30 This cost saving does not, however, take into account the fixed costs of borrowing through the global format, such as registration and clearing arrangements. These costs are presumably higher than for comparable Euro-issues, which do not require registration. For a global bond, SEC registration fees are proportional to the amount of the issue placed in the United States.

Administration and Supervision

Debt Management

In most industrial countries, the finance ministry carries out debt management. In doing so, the ministry normally consults with the central bank. In some countries, debt authority is somewhat decentralized in that other agencies can issue debt backed by the central government. In all large industrial countries, the objectives of debt management are cast only in general terms. For example, both U.S. and Japanese authorities maintain that they attempt to minimize debt costs, but they need follow no formal guidelines on the maturity structure of their issues.

In contrast, Ireland and New Zealand have recently reorganized their debt management operations. They have sought to increase the independence of these operations, while also giving debt management more clearly defined goals. Since 1988, debt management in New Zealand has been placed in a division of the Treasury that has a degree of autonomy from the rest of the government. The Debt Management Office (DMO) is responsible for increasing the Crown’s net worth and minimizing the risk to this net worth. New Zealand has developed a set of financial statements for the government, including a balance sheet. The goals of the DMO resemble those of a treasurer’s office for a corporation. It essentially attempts to reduce financial risk while maximizing return. To this end, the DMO adopted the practice of marking its liabilities to market on daily basis.

The DMO is now considering implementing a system of minimizing the risk to the Crown’s balance sheet. Under such a system, the DMO would set the duration and currency profile of its liabilities to match that of its assets. Since most of its assets consist of payments in New Zealand dollars, this strategy would entail reducing its foreign currency liabilities well below the current 45 percent of total liabilities. It would also entail a lengthening of the duration of liabilities. A rigorous matching of asset and liability risk would also call for indexing liabilities to inflation, but the DMO has said that it is unlikely to issue indexed debt.

Like New Zealand, Ireland has organized its debt management in order to provide clear performance objectives and a degree of autonomy from other government objectives. The National Treasury Management Agency (NTMA) began operations in Ireland in December 1990. It took over both the debt management operations from the Irish Department of Finance and the gilt (domestic government bond) market operations of the central bank.

The debt management objective of the NTMA is cast with reference to a benchmark portfolio. The main objective is to fund the Government’s debt at a lower cost than that of the benchmark portfolio. The benchmark consists of debt of a collection of specified maturities, with new debt issued at preset dates. The NTMA attempts to beat the benchmark both by funding at different dates than the benchmark, in order to take advantage of favorable market opportunities, and by issuing at different maturities than the maturities of the benchmark. The NTMA chooses its maturities subject to a limit on the amount of debt it is permitted to issue with maturities less than five years, and subject to guidelines on the proportions of foreign currency and floating rate debt.

Both Ireland and New Zealand have structured debt management to provide the debt manager with a degree of autonomy from the rest of the government and with explicit performance objectives. Such defined performance objectives facilitate the recruitment of personnel with the know-how to attain these objectives.

The Irish NTMA has a clear objective, which is to beat the benchmark. New Zealand’s DMO has a less clearly defined, but more comprehensive, objective, which is to optimize in a risk-return framework the Crown’s net worth. The advantage of a narrow Irish-type objective is that the task of the debt manager is clear and its performance easy to evaluate. Such a narrow objective, however, may fail to take into account other objectives, such as risk (although risk is limited by constraints on the maturity structure of the debt). A broader objective such as in New Zealand runs less risk of omitting such important considerations, but it may also be less clear operationally, and it may be more difficult to evaluate the performance of the debt manager.

Both countries offer an example of providing a government agency with a defined objective against which the performance of the agency can, at least in theory, be evaluated. Among government functions, debt management appears particularly well suited for such an approach, since its goals can be expressed in financial terms for which market prices can be used to determine results. Nonetheless, as discussed above, there is a trade-off between concreteness and comprehensiveness in the objectives. Countries not following this alternative may prefer their more comprehensive, although less concrete, approach.

The goals of debt management are easy to define in general terms, but are more difficult when specifics are examined. Typically a government seeks to minimize the costs of financing itself, perhaps with some allowance made for risk. Such a goal is. however, very difficult to implement in a coherent manner. Ex ante, only imprecise estimates can be made of the ultimate costs of different maturity and currency profiles of debt. Even if one could wait until the debt has matured, when the cost of the profile of debt chosen by the government can be determined and compared to the realized cost on other hypothetical profiles, no account can be taken of the risk of different portfolios. Debt management can also address other goals, such as providing instruments attractive to domestic savers, improving the depth and efficiency of domestic capital markets, and subsidizing and taxing various domestic and foreign entities.

The case for an independent debt management office is not clear-cut. In contrast to the interaction of monetary policy with other short-term policy objectives, such objectives do not conflict so obviously with debt management. Debt managers do not face a time consistency problem similar to that faced by monetary policymakers. However, a case for the independence of debt management is sometimes made on two grounds. First, if the debt management office is independent, its goals can be clearly defined and it can be organized to achieve those goals. Second, the human resources needed to operate a professional debt management operation are also in high demand in the private financial sector. If the debt management office is independent, it can offer salaries and career paths that can attract skilled personnel who might be unwilling to work under the conditions of ordinary government employment.

Finally, there is the interaction of debt management with other government policies. Of particular interest is the interaction of debt management with monetary policy. In theory, central banks may be tempted to manipulate financial markets to reduce the interest rates at which government debt is issued. However, the market will presumably learn about such behavior over time and demand compensation in the form of higher interest rates. Thus, the interests of policy are better served if the central bank does not engage in such manipulation. Central banks may also be tempted to inflate away some of the value of nominal debt, to which the market is also likely to react by demanding a higher interest rate. If the central bank does act in this way, the maturity and currency structure can start conveying information to the market about the central bank’s future monetary policy. Although these two considerations might suggest that monetary policy and debt management should be separate, arguments can be advanced for why coordination between the two functions is necessary. Day-to-day debt management operations affect the demand for liquidity in the economy, to which the central bank must react in order to manage liquidity in the economy. Moreover, many central banks use government securities to conduct their open market operations, and liquid government securities markets are essential for this.

Currency and Maturity Decisions

One of the basic decisions a debt administrator faces is the currency and maturity structure of its debt. The administrator has related options as well, such as issuing floating rate debt or debt indexed to inflation, and some countries also use derivative transactions to adjust their debt exposures. Countries have made very different choices both in the maturity and the currency of the debt, as well as in related matters. Market conditions in different countries, including clientele effects, are among the factors that influence such policies.

Observing that the yield curve has tended to be positively sloped over the postwar period, authorities in the United States in 1993 embarked on a policy of shortening the maturity structure of the debt. To implement this, they have reduced the volume issued of 30-year bonds, which have been the longest maturity securities issued by the U.S. Government. They have also stopped issuing seven-year notes and will compensate by issuing more Treasury bills (one-year maturity or less) and Treasury notes of three-year maturity and less.

Germany and Japan both issue most of their debt in medium-term maturity. The bulk of Japanese Government securities are the ten-year bonds, although Japan also issues a range of other maturities. Most German issues are five-year and ten-year securities. German officials have until recently followed a policy of not issuing liquid short-term issues. This policy has changed incrementally in recent years with the issue of commercial paper by the government-guaranteed Treuhandanstalt and the issue of bills by the Bundesbank.31 German authorities also have embarked on a policy of issuing longer-term debt; they issued a 30-year bond in December 1993, for the first time since 1986, and reportedly plan to issue such bonds regularly. German policy has therefore focused on expanding the range of maturities issued, but the impact of this expansion on average maturity seems uncertain.

France and the United Kingdom follow a policy of not using short-term issues (one-year maturity or less) to finance their budget deficits. In these countries, short-term securities are largely issued in order to provide a liquid instrument in the domestic money market; this is especially true in the United Kingdom, where operations in Treasury bills are the main monetary policy tool of the Bank of England. In 1993, however, the U.K. authorities suspended a full-funding rule that required the deficit to be financed by medium-term and long-term liabilities held outside the banking sector. French debt policy is to finance the deficit with BTANs and OATs, which carry maturities of two years and up.

Adjustment of a government’s maturity and currency profile of debt can reduce the cost of its debt services. Whether this has an economic impact is related to whether government deficits matter, which in turn depends on whether markets are complete.32 Long-term debt often carries a return premium that appears to be in excess of expected future short-term rates. If this term premium is in fact a risk premium, a government could reduce its expected debt costs by issuing short-term debt instead of long-term debt. In a world where debt policy did not matter, future taxes would become riskier to compensate for such a policy, and in an ideal world there might be no economic impact. However, if reducing debt costs is an independent objective, such a term shortening could still be desirable.

An active theoretical literature covers the maturity and currency structure of government debt from a different angle. In terms of maturity structure, there are two contradictory lessons of this literature. The first is that a government with short-term debt will have less of an incentive to choose an inflationary policy than will a government with long-term debt. The reason for this is that a permanent increase in the inflation rate will reduce the value of long-term debt by more than it will reduce the value of short-term debt. It follows from this that a government may in theory be able to enhance the credibility of its monetary policy by reducing the average maturity of its debt. For this to work, however, the value of the debt must be an objective of monetary policy.

On the other hand, short-term debt exposes the government to greater risk of a crisis of confidence.33 A government that is exposed to a speculative attack on its currency can defend its currency by raising domestic interest rates. If the government has to roll over its domestic currency debt during this period, however, it will have to pay the higher domestic interest rates, which will make it less willing to raise interest rates to defend the currency, thus making a successful speculative attack more likely. The solution is for the government to minimize the amount of domestic currency debt coming due at any single time, by lengthening maturities and smoothing the maturity dates over time. This smoothing of maturity dates goes against the desirability of bunching of maturity dates in order to build the size of individual issues and enhance secondary market liquidity.

This literature has a less ambiguous message about debt in foreign currency or indexed to inflation. Such debt is beneficial, because the government cannot affect its real value by changing the rate of inflation. Despite this argument, however, most countries continue to issue most of their debt in their own currency and to avoid indexation. The literature suggests that perhaps this is because nominal debt has an insurance role; if a country encounters a bad shock, it can use inflation as a means of taxing debt holders, some of whom are foreign, thereby spreading the effects of the shock. Knowing this, however, holders of nominal debt may demand a higher expected return.

Supervision of the Secondary Market

Authorities supervise the secondary market in government securities for several different reasons. One goal of supervision is to prevent systemic risk that might arise from the failure of firms operating in the market. A second goal is the protection of investors in the market from deceptive or unfair practices. Authorities may also wish to ensure that the market is liquid.

In the United States, supervision of most major firms involved in the secondary market for government securities is accomplished as part of the overall supervision of these entities. One of the federal banking supervisors (usually the Comptroller of the Currency or the Federal Reserve) supervises the banks involved in the market, which include a number of primary dealers. The main focus of this supervision is to ensure these banks’ financial soundness. Securities firms involved in the market fall under the purview of the SEC. This supervision focuses on both financial soundness and customer protection, although it does not extend to subsidiaries of the registered broker-dealers, which in some cases are unsupervised.

Entities that served as only brokers or dealers for government securities used to be exempted from regulation and supervision. After a series of failures of secondary government securities dealers in the early and mid-1980s, this regulatory gap was plugged by the Government Securities Act of 1986, which took effect in 1987. This act gave authority to regulate these brokers and dealers to the Treasury Department. The act also required them to join a self-regulatory organization, in practice, the National Association of Securities Dealers, which would conduct supervision.

The U.S. system of supervision therefore covers all entities engaged in brokering or dealing U.S. Government securities in the United States. However, firms involved in the market but not brokering or dealing may be exempt from supervision. Firms operating in the U.S. Government securities market outside the United States are also not covered by this supervisory regime. Although Treasury securities are listed on the New York Stock Exchange (NYSE), secondary market trading in U.S. Government securities is almost exclusively an OTC dealer market. This market is subject to little direct regulation or supervision. There are no daily price limits or circuit breakers in place.

Publication of quote, price, and trade information is incomplete. The bulk of secondary market trading occurs between primary dealers through the seven interdealer brokers. Primary dealers have access to broker screens, which contain the quotes and transactions data that give the best picture of the market. Only one of these brokers makes its screen public. A limited amount of information from some of the other brokers began to be made public in 1991 through the GOVPX service. The information available through GOVPX consists of real-time price and quote information for all Treasury securities, although there is no information on strips of Treasury securities (separate trading of interest and principal as zero-coupon issues) and the quote information does not contain quote sizes. Dissemination of market data has also been increased by the expansion of the customer bases of some of the interdealer brokers, which have expanded their customer bases to all netting members of the Government Securities Clearing Corporation, a status for which approximately 75 firms are eligible.

In the last two years, the Federal Reserve Bank of New York has developed a system of market surveillance. It examines price data in an attempt to detect anomalies that may be associated with market manipulation. U.S. authorities recently received authorization to request reports of large positions in securities where a pricing anomaly exists.

In Japan, banks and securities firms acting as market-makers for government bonds must receive permission to do so from the Ministry of Finance. The Ministry examines a firm’s capital level and management before granting this license. The Ministry also authorizes brokers in the market. Trading in government securities can take place either OTC or on the Tokyo Stock Exchange, where all government bonds are listed, although most trading occurs OTC. The transaction tax applies to all domestic bonds, including government securities (except short-term bills), and tends to reduce trading volume. Trades involving two foreign counterparties are exempt from the transaction tax.

Most dealers and brokers in securities markets in Germany are banks. In contrast to the United States and Japan, there is no legal distinction in Germany between securities firms and commercial banks. Banks are licensed and supervised by the Federal Banking Supervisory Office. Some brokers, however, fall outside this supervisory system and normally are licensed and supervised by one of the eight regional stock exchanges. The implementation of a new regime with a federal securities supervisor is under way, with the new supervisor likely to commence operations in 1994. Trading in German Government securities takes place both on the stock exchanges and OTC, in addition to substantial offshore trading. Most domestic trading is in the OTC market, although there is a daily fixing on the stock exchange at which the price of government bonds is set. Most exchange trading occurs at this price, and the Bundesbank sometimes intervenes at this fixing to counteract price pressure. Market activity both on the exchange and OTC is subject to little supervision; the stock exchanges have nominal authority, but their power is limited.34 Some German trading restrictions have loosened recently. The Government eliminated taxes on securities transactions in 1991 and 1992. It also abolished in 1990 a special commission that inhibited the resale of bonds in the first year following issue.

In France, the secondary market in government securities is open to a broad spectrum of participants. The main group of participants consists of the eighteen SVTs. The French Treasury requires these entities to participate in both the primary and secondary markets for these securities, in exchange for certain privileges (see section on Primary Dealer Systems above). Members of the Paris Bourse and other financial intermediaries may also participate in the market; market participants are supervised by the Société des Bourses Françaises. Trading is permitted both on the Bourse and OTC, with most trading OTC. In addition, a substantial portion of the secondary market, especially in OATs, takes place offshore. In general, France has no capital restrictions that apply separately to government securities. SVTs, however, must satisfy a minimum capital requirement of F 300 million. SVTs are supervised by the Commission Bancaire; they must also submit weekly activity reports to the Treasury. Other participants in the market do not appear to face separate supervision or reporting requirements for their government securities operations.

The market for government securities in the United Kingdom centers on the GEMMs (see section on Primary Dealer Systems above). These are firms approved by the Bank of England, which also supervises the firms’ market-making activities in gilts. Through the same system, the Bank authorizes and supervises two special types of institutions, the interdealer brokers and the stock exchange money brokers, which do business with the GEMMs. Other participants in the gilts market fall under general U.K. investment business regulation, and must generally be authorized by the Securities and Investment Board (SIB) and join a self-regulatory organization. This authorization does not apply to firms operating offshore, although it does apply to foreign firms operating in the United Kingdom. It also does not apply to investors in gilts and other securities.

Firms operating in the market for Treasury bills fall under the money markets’ supervision system. Firms conducting wholesale transactions in the sterling money market or in the foreign exchange or bullion market must be “listed” by the Bank of England under its wholesale markets supervision regime. This listing entails a series of checks on the firms’ management, reputation, and financial position. Firms conducting nonwholesale money market transactions fall under the general securities regulation and supervision of the SIB and its self-regulatory organizations.

GEMMs are under stringent capital adequacy rules. Other participants in government securities markets face capital requirements set by the Bank of England for banks and money market firms or by the SIB or a self-regulatory organization for securities firms. Banks with significant securities business may also face the SIB’s securities capital rules. These rules, as well as those of the Bank’s money markets regulation, focus on market risk as opposed to credit risk.

U.K. authorities conduct intensive monitoring of market participants. GEMMs submit daily electronic reports of their position in each gilt issue to the Bank of England. For money market firms, the Bank requires fortnightly reports on risk exposures and capital positions, except for pure brokers, which must submit reports only monthly. The Bank can also carry out spot checks of these firms at other times to ensure continuous compliance with its capital requirements.

U.K. securities firms also are subject to detailed investor protection rules. Most securities firms, including GEMMs, are subject to the rules of the Securities and Futures Authority (SFA), one of the self-regulatory organizations under the SIB. Wholesale money market firms are subject to a lighter regulatory system under the Bank of England. For GEMMs, the Bank also solicits feedback from investors to verify that the GEMMs are providing adequate investor services.

There are no formal circuit breakers for the gilts market. However, procedures are in place for the Bank of England to intervene in the market. The Bank has the authority to enter the market to react to price anomalies for particular issues. The Bank’s tap mechanism is also designed to react to market conditions. The guiding principle is for the Bank not to sell additional amounts of an issue into a falling market.

In Italy, the domestic secondary market in government securities is divided in three parts, two of them closely supervised. The bulk of trading volume occurs on the screen-based Mercato Telematico Secondario (MTS). A variety of financial firms can join the market; over 330 have signed up to be dealers on the MTS. A small proportion of Government securities transactions, mostly retail, pass through the Milan Stock Exchange. All Government securities except Treasury bills are listed on the exchange. There is also an informal OTC market that is more loosely regulated. In addition, some trading of domestic issues of Italian Government securities occurs in foreign countries. Until recently, there was a stamp tax on government securities transactions. The Government abolished this tax at the end of 1993, in part to entice trading back from offshore centers.

Primary dealers on the MTS are required to post continuous two-way quotes in a subset of Government securities issues. These quotes must be for a minimum size of Lit 5 billion, which is also the minimum trade size on the MTS. Other dealers on the MTS face no market-making requirements. There also are no market-making requirements on the OTC market.

Primary dealer status is open to both banks and nonbanks incorporated in Canada. These entities are known as “primary distributors,” with a more active subset known as “jobbers.”35 Jobbers are required to make markets in government securities. Both primary distributors and jobbers must submit weekly statistical reports on their trading activities to the Bank of Canada. The domestic market in Canadian Government securities is exclusively an OTC market, but trading also occurs internationally. Since August 1993, information from interdealer broker screens has been made available to the public. Before then, information was only available to those dealing on the screens.

Annex II Developments in the Regulation of International Banking

The impetus for recent regulatory initiatives in international banking has come from the growth of derivatives markets and the increasing involvement of wholesale banks in these markets.1 The debate over the regulation of derivatives markets has led to a redoubling of the long-standing efforts by bank regulators to improve the capital cover for risk positions entered into by international banks, without unduly distorting incentives.2 This effort resulted in the publication of three proposals in April 1993 that recommended a capital requirement against market risk, a broader recognition of netting as a means of reducing credit risk, and disclosure of interest rate risk.

Recent trends in the markets for derivative instruments are described first, followed by a discussion of the proposed extensions of the Basle capital requirements.

Growth of the OTC Derivatives Markets

The notional principal amount of exchange-traded derivative instruments increased by 69 percent in 1993, reaching $7.8 trillion, which was more than 12 times the total in 1986 (Table A3).3 Interest rate futures and options together had a notional principal value of $7.3 trillion in 1993, while currency futures and options accounted for $111 billion and stock index futures and options for $406 billion. Turnover of these contracts on organized exchanges rose by 22 percent in 1993, reaching a combined total of 774 million contracts (Table A4).4 Trading was boosted in both 1992 and 1993 by the turmoil in the European exchange rate mechanism.5

While comparisons between the markets are difficult, activity on the OTC market appears to have grown more rapidly than has activity on the exchanges. At the end of 1992, the notional principal value of outstanding interest rate swaps and currency swaps was $4.7 trillion (Table A5). This was 22 percent higher than the end-1991 notional principal and more than five times the end-1987 figure. In the second half of 1992 alone, the notional principal of new interest rate swaps was $1.5 trillion, and new currency swaps reached $146 billion (Table A6).

In more specific terms, the growth of the markets is characterized by increasing globalization, increasing involvement by banks, and increasing complexity of products.6 At the end of 1987, 79 percent of all outstanding interest rate swaps by notional value were denominated in U.S. dollars (Table A7). However, by the end of 1992, U.S. dollar interest rate swaps represented only 46 percent of outstanding interest rate swaps. Similarly, the proportion of U.S. dollar currency swaps has declined from 44 percent of the outstanding notional value in 1987 to 36 percent in 1992. A parallel trend is observed in the distribution of notional principal and turnover of exchange-traded instruments between the United States, Europe, and Japan (Tables A3 and A4).

Swaps are used to reduce funding costs and to transfer payment characteristics and interest rate or currency risk exposures. Thus, for example, an issuer of a straight bond might swap the proceeds into an obligation to pay a floating rate, or a U.S. corporation might issue a deutsche mark bond and swap the proceeds into dollars, thereby eliminating exchange rate risk. The counterparty data on interest rate swaps show that by far the most important participants in these markets are financial institutions, which held 76 percent of the outstanding contracts by notional principal value of interest rate swaps in 1992 (Table A5). In the market for currency swaps, however, financial institutions accounted for only 54 percent of the outstanding positions at the end of 1992.

The involvement of banks in the OTC markets has motivated much of the regulatory examination of derivatives. The notional value of financial derivatives (including forwards) held by U.S. bank holding companies was $5.1 trillion in June 1992, which was 27 percent higher than the notional value in September 1990.7 In June 1992, bank holding companies with less than $10 billion in capital held about $34 billion in interest rate swaps and $153 million in currency swaps. These were, respectively, more than 47 percent and 186 percent higher than the corresponding figures in September 1990. Many regional banks in the United States have started to sell derivative products to clients in addition to using derivatives as a financing and risk management tool for themselves. Many mutual funds have used derivatives to manage risk or enhance returns.8

The use of interest rate and currency swaps by nonfinancial institutions has also increased substantially.9 The amount of interest rate swaps used by governments and international institutions has increased from $47.6 billion at the end of 1987 to more than $242 billion at the end of 1992 (Table A5). The use of interest rate swaps by corporations has also increased significantly, from about $129 billion in 1987 to about $666 billion at the end of 1992. In the same five-year period, currency swaps used by governments and international institutions rose from $33.9 billion to $110.6 billion, and currency swaps used by corporations increased from $51.6 billion to $282.2 billion.

Outstanding notional principal in the sophisticated OTC products such as caps, collars, and floors increased from $468 billion at the end of 1991 to $507 billion at the end of 1992. The notional principal value of outstanding swaptions rose from $109 billion at the end of 1991 to $127 billion at the end of 1992.

Many reasons have been cited as explanations for the rapid expansion of the OTC market. These include (i) a rise in the demand for hedging instruments owing to increases in the volatility of exchange rates and interest rates; (ii) an increase in the demand for sophisticated instruments to profit from relatively large interest rate differentials across borders and across different investor groups; (iii) a rise in the demand for sophisticated products to unbundle risk and to alter the risk characteristics of portfolios in a rapidly changing investment environment; (iv) a reduction in the cost of providing OTC derivatives securities owing to advances in technology (including computing and information processing facilities); (v) a growing demand by banks to seek new business as the profitability of their traditional lending business has declined; and (vi) worldwide deregulation and financial market liberalization.

Strengthening Capital Requirements

The Basle Committee proposed a capital adequacy standard in July 1988 that provided an explicit approach by which the minimum capital requirement to cover credit risk for both on- and off-balance sheet positions could be computed. In addition, it provided guidelines on what could be counted as capital. In 1989, the Council of the European Community issued an Own Funds Directive (OFD) and a Solvency Ratio Directive (SRD) for similar purposes.

While the 1988 accord was well received, it has been long recognized that its focus on credit risk alone was too narrow. Other risk factors, such as interest rate risk and exchange rate risk, might be very important, especially for positions that are held for relatively short periods of time. Furthermore, banks might have an incentive to seek a higher return on capital by substituting interest rate or exchange rate risk for credit risk when designing their portfolios if capital requirements apply only to credit risk. Market participants have also expressed concern that the definition of capital is too restrictive. That definition was chosen to cover banks’ traditional loan positions, which tend to be held for long horizons. However, as banks’ “trading books”—which are composed of relatively short-term positions—become more important, a more flexible definition might be appropriate.

It is also widely recognized that the 1988 accord was too stringent in its treatment of netting. Under that accord, only netting by novation was recognized for the purposes of calculating the capital requirement. The 1990 Lamfalussy Report suggested that any legally enforceable form of bilateral netting should be recognized, since netting can improve the efficiency and the stability of the banking system by reducing credit exposure, liquidity risk, and transaction costs.10

The 1993 Basle Proposals

In response to these concerns, the Basle Committee released in April 1993 a set of three consultative papers that contains proposals to revise the 1988 accord. The papers consider: (i) the recognition of netting for the capital requirements for credit risk; (ii) the reporting of, and capital requirements for, market risk and foreign exchange risk; and (iii) the measurement and reporting of interest rate risk. Furthermore, a new tier of capital, called tier 3 capital, was also defined.

The 1993 Basle proposals are similar in many ways to the Capital Adequacy Directive (CAD) issued by the EC Council to modify and supplement its 1989 directives, which had been criticized for the same reasons as the 1988 Basle accord.11 The CAD and the new proposals follow the same approach in measuring interest rate risk and have the same duration-based market risk weights. Furthermore, for debt securities, the two share the same structure for aggregating market and specific risk and have the same risk weights for specific risks. However, the two differ significantly in the treatment of foreign exchange risk and position risk in equities. In those respects, the Basle proposal is more stringent than the CAD. On the other hand, the CAD applies to both banks and securities houses while the Basle proposals apply only to banks.

Capital Adequacy Standards

The proposed framework. The main conceptual breakthroughs of the proposals on the computation of capital requirements are (i) the recognition of a need to treat items in the loan book and the trading book of banks differently when setting minimum capital requirements; (ii) the recognition of a need to account for market risk and specific risk differently; and (iii) the calculation of capital requirements for general market risk from a portfolio perspective.

The trading book is composed of (i) proprietary positions in financial instruments; (ii) exposures due to unsettled transactions; and (iii) positions taken in order to hedge other elements of the trading book. Because these positions are usually held for only a short time, risks arising from adverse movements in interest rates, exchanges rates, or other prices can be very important relative to credit risk. This suggests that positions in the bank’s trading book and loan book should be treated differently. Under the new proposals, the capital requirement for items in the loan book is calculated in the same way as specified in the 1988 accord, with only some changes to reflect the more general recognition of bilateral netting. However, the capital requirement for items in the trading book is calculated based on a new framework.

In this new framework, risk exposure is separated into specific risk and a general market risk. The key distinction between the two is that the specific risk exposures of different positions are generally unrelated, which is not true for market risk. The specific risk exposure of the entire trading book can be computed simply as the sum of the specific risk exposures of each position in the book. However, since the market risk of different positions is related, the proposals suggest a portfolio approach that permits some offsets between long and short positions in the calculation of the general market risk exposure. Thus, the total capital requirement is computed as the sum of a specific risk charge (applied to the gross position of the portfolio) and a general market risk charge (applied to the net position of the portfolio). This is the so-called building block approach in the proposals.

Furthermore, since the nature of the specific risk and market risk varies across types of instruments, the Basle proposals classify positions into debt securities positions, equity positions, and foreign exchange positions and use different approaches for calculating the specific risk and general market risk charges for these groups of instruments. Under the proposed framework, a bank’s minimum capital requirement is computed as the aggregate of capital charges for the specific and market risks for each of its debt and equities positions, plus charges against foreign exchange risk and credit risk arising from its loan book.

It is important to note that, in general, the various components of general market risk are neither uncorrected nor perfectly correlated. In simply adding the capital charges for general market risk of debt securities to the capital charge for foreign exchange risk and to the capital charge for general market risk of equity positions, the Basle Committee implicitly assumed that the market risk elements are perfectly correlated. This approach is a conservative one, as the standard deviation of portfolio value computed in this way is always at least as high as the actual standard deviation of the portfolio value under less than perfect correlations among the risk factors. The approach has the merit of being easy to implement since time-varying relationships among the many risk factors are ignored. Furthermore, the capital requirement for each individual risk factor can be set separately.

Specific risk of debt securities. As in the 1988 accord, debt securities are classified into a number of broad categories by issuer; then a different risk weight for each category is assigned to capture differences in the probability of default or credit rating change for different groups of issuers. There are five such categories under the 1993 proposals: (1) government (with no risk weight); (2) qualifying securities (basically investment grade) with residual maturity of six months or less (with a risk weight of 0.25 percent); (3) qualifying securities with residual maturity between 6 and 24 months (with a risk weight of 1 percent); (4) qualifying securities with residual maturity exceeding 24 months (with a risk weight of 1.6 percent); and (5) other securities (with a risk weight of 8 percent).12 The CAD has the same definition of these categories and assigns the same risk weights.13

The definition of “qualifying” as investment grade is an important improvement over the 1988 accord in which the classification of counterparties was dependent not on credit rating but on whether they were Organization for Economic Cooperation and Development (OECD) country institutions. For example, under the 1988 accord, loans to OECD official borrowers carry a zero risk weight; claims on banks incorporated in OECD countries carry a 20 percent weight; residential mortgages have a 50 percent weight; and foreign currency loans to non-OECD governments and loans to private companies carry a 100 percent weight.14 Since the weighted exposures are multiplied by 8 percent to yield the capital requirement, the effective percentages for comparison with the risk weights in the 1993 proposal are 0 percent, 1.6 percent, 4 percent, and 8 percent, respectively. Under the 1988 accord, an interest rate contract with less than six months to maturity to a private firm, in a non-OECD country, that is rated AAA or equivalent by approved rating agencies, would carry capital charge equal to 8 percent of the marked-to-market value under the so-called current exposure approach. Under the new proposal, the capital charge would be 0.25 percent of the marked-to-market value of the position.

General market risk of debt securities. The general market risk of debt securities is mostly interest rate risk. Unlike credit risk, which is counterparty-specific (and hence position-specific), interest rate changes can affect many positions in the book simultaneously. The approach taken in the 1988 accord, which computes the total capital charge as the sum of capital reserves for individual positions, is perceived to be inappropriate by many market participants who have argued that interest rate risk should be accounted for from a portfolio perspective in which offsets at least between long and short positions should be allowed.

The interest rate risk exposure of a portfolio of debt securities is determined by the sensitivity of the value of the portfolio to interest rate changes and by the volatility of interest rates. The sensitivity of a portfolio of debt securities and derivatives to changes in interest rates can be constructed from the duration measures of the debt instruments and the deltas of the derivative instruments, which can be calculated from pricing models or from simulation techniques.15 The volatility of interest rates can be implied from options prices or calculated from historical data. However, for relatively small institutions that lack sophisticated risk management systems, these computations can be prohibitively difficult when their trading books contain a large number of positions. Hence, the Basle Committee has developed a “standard” approach, which is a compromise to facilitate implementation.

This standard “maturity ladder” approach involves setting (i) a manageable number of maturity bands (instead of a continuum of maturities); (ii) a representative duration measure for each band (instead of a duration measure for each instrument); and (iii) a representative interest rate change for each band (instead of a different change for every interest rate maturity within each band).

Under this approach, each position is converted into a combination of simple debt instruments that are then classified into 15 maturity bands. For example, a ten-year interest rate swap under which a firm is receiving floating rate interest and paying fixed is treated as a long position in a floating rate instrument of maturity equivalent to the period until the next interest fixing date (often six months) and a short position in a ten-year fixed rate bond. A long position in a three-month interest rate future maturing in one month is reported as a long position in a government security with a maturity of four months and a short position in a government security with a maturity of one month. An option on a debt instrument is treated as if it is a position equal in value to the amount of the underlying instrument, multiplied by the delta of the option determined by an approved option pricing model.

Each converted position is then assigned a risk weight, which is the product of the representative modified duration measure for the maturity band and an assumed interest rate change for the particular maturity band.16 The capital requirement is then calculated as the sum of the weighted positions (where long and short positions can offset each other).

However, positive and negative positions in the same maturity band might not be perfect hedges for each other because of the size of each maturity band and the existence of gap risk and spread risk. As an example of spread risk, a long position in a three-month U.S. Treasury bill is not a perfect hedge for a short position in a three-month Eurodollar bond since the three-month Treasury bill rate and the three-month LIBOR or Eurodollar rate are not perfectly correlated. The variation in this spread is a risk factor that should be taken into consideration. Another example is gap risk: a long position in a three-month Treasury bill is not a perfect hedge for a short position in a four-month Treasury bill (even though they are classified in the same maturity band) because the three-month yield and the four-month yield are not perfectly correlated.

To provide for such imperfections in the mutual hedging of instruments in the same maturity band, the Basle Committee proposed to impose a vertical disallowance. This disallowance is added to the net capital charge in the computation of the capital requirement for debt securities. A vertical disallowance for each maturity band is calculated as 10 percent of the smaller of the total capital charge for long positions within the band and the total capital charge for short positions within the band. The total vertical disallowance is then computed by summing the vertical disallowances across maturity bands. Without the vertical disallowance, the pairs of positions in the above examples will be treated as perfect offsets as if there is no risk at all.

Similarly, positive and negative positions in different maturity bands are not perfect hedges for each other since the yield curves need not move in a parallel fashion. In order to account for such imperfect offsets, the Basle Committee also proposed the imposition of horizontal disallowances. Since positions from distant maturity bands are the worst offsets, the capital charge should be increased appropriately. For this purpose, the Committee has proposed to group the maturity bands into three different maturity zones. Within-zone horizontal disallowances are smaller than across-zones disallowances. Also, adjacent-zones disallowances are smaller than nonadjacent-zones disallowances.

While the CAD does not explicitly mention these disallowances, the computation of capital requirements is essentially equivalent to the imposition of such disallowances, since matched and unmatched positions within and across maturity bands are assigned different weights that take into account the risk from hedging with nonidentical instruments.

The Basle Committee also proposed to allow banks with more sophisticated risk management systems to use an alternative method under which the duration of each position is computed separately. However, this is permitted only if the alternative method produces results that are consistently equivalent with the standard method.

Specific risk of equity securities. The specific risk of an equity security is the risk of an unexpected change in the price of the security that is unrelated to the general movement of the stock market. The Committee has proposed that the specific risk charge for an institution be computed as 8 percent of its gross equity position—the aggregate value of all equity positions without any offset between long and short positions. However, at the discretion of national regulators, a 4 percent weight can be applied if the equity portfolio is liquid and diversified. Under the CAD, the specific risk charge is 4 percent of the overall gross equity position instead of 8 percent. Furthermore, if the portfolio is liquid and diversified, a 2 percent weight can be allowed.

General market risk of equity securities. Since general market risk is common to all equity securities, the market risks of long and short positions can offset each other. Thus, the actual risk exposure depends on the overall net equity position rather than the gross position. Under the 1993 proposal, the capital charge for the general market risk of an equity portfolio is computed as 8 percent of the overall net position of the portfolio. The CAD follows the same approach and has the same 8 percent weight.

Foreign exchange risk. Foreign exchange risk is the risk of loss owing to fluctuations in exchange rates. The exchange rate risk exposure of a portfolio is determined by (i) the sensitivity of the value of the portfolio to changes in exchange rates; (ii) exchange rate volatilities; and (iii) correlations between exchange rates. Computing the sensitivity of the portfolio to exchange rate changes can be difficult because the values of foreign exchange derivatives are often complicated functions of the exchange rate. Strictly speaking, to determine the foreign exchange risk exposure, the behavior of all exchange rates on which the contracts depend has to be modeled, correlations need to be taken into consideration, and the future profits or losses of the entire portfolio should be simulated. However, such modeling and computation can be too complicated and expensive for smaller banks. Hence, the Basle Committee proposed a shorthand method for measuring foreign exchange risk.17

The Basle Committee also proposed that institutions with the necessary expertise, computer systems, and data could, if permitted by their national supervisors, use an alternative method based on simulations. However, an additional 3 percent charge is added to the capital charge obtained from the simulation. Consequently, the minimum capital charge can never be less than 3 percent of the overall net foreign exchange position.

The CAD is more lenient than the Basle proposal in that under a similar standard method, the capital charge is computed as 8 percent of the amount by which the overall net foreign exchange position exceeds 2 percent of the institution’s own funds. Furthermore, under the simulation method, the CAD requires only that the capital charge be larger than 2 percent of the overall net foreign exchange position.

Netting

In the 1988 accord, only netting by novation was recognized for the purpose of calculating capital requirements. Netting by novation entails a bilateral contract between two counterparties under which a new obligation to pay or receive a given currency is automatically amalgamated with all previous obligations in the same currency, thereby creating a single legally binding net position that replaces the larger number of gross obligations. The 1993 proposals extend the recognition of netting for capital requirement purposes to include any bilateral netting agreement that is legally binding in the jurisdictions of both parties and meets the minimum standards recommended by the 1990 Lamfalussy Report.

The more general recognition of bilateral netting for the computation of capital requirement is facilitated by recent legal developments on the enforceability of close-out netting. A close-out netting arrangement can eliminate cherry-picking behavior at the time of bankruptcy, and its enforceability is usually regarded as the legal basis for bilateral netting for capital adequacy purpose. In the United States, legal developments facilitate the recognition of bilateral netting.

Under the new Basle proposal, for banks using the so-called current exposure method in the 1988 Basle accord, the credit exposure on bilaterally netted forward transactions is calculated as the sum of the net marked-to-market replacement cost, if positive, plus an add-on based on the notional underlying principal. For banks now using the so-called original exposure method under the 1988 Basle accord, a reduction in the credit conversion factors applying to bilaterally netted transactions is permitted temporarily until the market risk related capital requirements are implemented. The original exposure method, which computes capital charges as percentages of notional values, will ultimately be abandoned for netted transactions, as it does not account for current and potential future exposures separately and hence is not entirely compatible with the idea of netting.

Industry Responses to the 1993 Basle Proposals

Regarding the treatment of market risk and interest rate risk, a frequent comment from market participants has been that the proposals may increase capital requirements for the affected institutions. An increase in capital requirement would be due to the size of the vertical disallowance and the computation of total capital charge as the simple sum of the capital charges for individual general market risk elements. There is also a concern that the proposals do not create incentives for banks to adopt more advanced risk measurement systems. Instead, most banks would have to maintain two risk reporting systems (one for the computation of capital requirement under the Basle proposals and a more complicated one for everyday risk management). This is financially inefficient and can reduce banks’ competitiveness relative to nonbanks. Furthermore, there is also a general feeling that the distinction between the trading book and the loan book should be clearly defined. Leaving the definition to the discretion of national authorities can potentially create unfair competition among banks in different nations.

The banking industry generally welcomed the recognition of bilateral netting since this can reduce capital requirements, which in turn will allow banks to take on more business than they are currently able to accept. The Bank for International Settlements (BIS) estimated that this proposal may reduce capital requirements for swap dealers by between 25 and 40 percent. The International Swaps and Derivatives Association (ISDA) estimated that the recognition of bilateral netting could reduce capital requirements by as much as 48 percent.

Some market participants have argued that a 10 percent vertical disallowance is too large, particularly for an institution that has a lot of interest rate swaps and short-term debt instruments. Since a floating rate instrument is at par immediately after a payment set date, the value of the floating rate side of a swap is treated as a debt instrument with maturity date equal to the next payment set date and with a principal amount equal to the notional value of the swap, which can be a huge number. As a result, a 10 percent charge can be a substantial burden.

Some participants have also expressed concern that simply adding the capital charges for individual risk elements can result in an overestimation of the amount of capital required. At the heart of the criticism is the implicit assumption of perfect correlation among the risk factors. Participants have argued that the correlation between two risk factors should be taken into account when aggregating the capital charges. If risk factors are imperfectly correlated, then less capital should be needed. In addition, the proposed aggregation approach might distort banks’ activities since risk factors having low correlations are overcharged relative to risk factors having high correlations. Some market participants have therefore recommended that actual correlations estimated from historical data should be used.

A remaining criticism deals with the calculation of add-ons and the way they are included in the computation of capital charges. The Committee’s decision to retain the rules in the 1988 accord on add-ons has been criticized as being inconsistent with the idea of netting. Under the 1988 accord, add-ons are computed as a percentage of the gross notional principal. That is, netting is allowed for potential future exposure. The Committee’s explanation is that there is no guarantee that there will be offsets of exposures in the future, so it might be prudent to keep the current scheme. Critics, on the other hand, tend to argue that this is too conservative. They have complained also that computing credit exposure as the sum of the addons and the higher of the net replacement value and zero might overstate the actual capital requirement needed. The argument is that under the proposed rule, a financial institution will be required to hold capital equal to the add-on regardless of its net replacement value. That is, the institution is even required to hold capital to safeguard a liability. The position of the Basle Committee is that the purpose of the add-ons is to cover potential future exposure; there is no reason to expect that there would be no future exposure just because the current exposure is zero.

Under the recent proposals, derivatives positions are converted into equivalent positions in the underlying instruments based on the deltas of the derivatives. However, delta can change as the price of the underlying security changes. There is therefore an additional risk element—gamma risk—related to the sensitivity of the delta with respect to the price of the underlying security. This gamma risk can be substantial. Furthermore, related to the gamma risk is the risk of changes in volatility—vega risk. Given the highly volatile nature of exchange rates and the rapid changes in volatility, many market participants have essentially taken derivatives positions to bet on volatility changes. These positions are typically created to have zero delta. As such, no capital charge is attached, yet these can be highly risky positions. The failure of the Basle proposals to cover gamma risk and vega risk can potentially lead to distortion in the banks’ activities in which these uncharged risks are substituted for those covered by the proposals in order to reduce the capital requirement.

Some practitioners have complained that the 3 percent foreign exchange risk add-on that would be added to the capital requirements obtained from simulation methods is arbitrary and can be a severe penalty for banks using a sophisticated and probably more accurate approach to risk management. This can discourage the adoption of advances in risk management technology, which goes against the spirit of the exercise.

Recommendations from Regulators

In the United States, the 1988 Basle accord was adopted by all bank regulators. In a joint report, the Federal Reserve, the Federal Deposit Insurance Corporation (FDIC), and the Office of the Comptroller of Currency (OCC) discussed the importance of the market risk component of the derivatives activities of commercial banks and argued for a portfolio-based evaluation of this risk.18

The importance of interest rate risk has also been noted by these agencies. In September 1993, they issued a proposal on measuring banks’ interest rate risk and establishing a related capital requirement.19 Under this proposal, a bank’s interest rate risk exposure can be measured by a supervisory model similar to the interest rate risk measurement model proposed by the Basle Committee in April 1993 or by the bank’s own risk management system if it is approved by the regulator. Two alternative methods were proposed for setting minimum capital requirement. Under a so-called minimum capital standard approach, an explicit capital charge based on the amount of interest rate exposure in excess of a supervisory threshold would be imposed. Under the so-called risk assessment approach, the capital requirement would be assessed on a case-by-case basis, dependent on the bank’s interest rate risk exposure, internal risk controls, and financial condition.

Bank regulators also recognized the importance of netting arrangements. In a circular issued in October 1993, the OCC recommended that each national bank should use master close-out netting agreements with its counterparties to the broadest extent possible as long as those agreements are legally enforceable. The OCC also proposed that for positions collateralized by cash or government securities, the risk weight used in the computation of risk-based capital requirement should be reduced to reflect the minimal credit risk of these positions.

Harmonization of Standards for Banks and Securities Firms

In some countries, securities houses are subject to capital requirements by securities and exchange regulators. For example, in the United States, securities houses are subjected to SEC capital requirements that utilize a different approach from that of the Basle Committee. The “Net Capital Rule” of the SEC requires a minimum level of liquid net worth after some deductions—the “haircuts”—to account for market and credit risks. The haircut for unrealized profit associated with OTC derivatives positions is 100 percent, which is viewed by many market participants as too stringent. The SEC is considering modifying the Net Capital Rule to better capture the credit and market risks of derivative products. In May 1993, it issued a proposal in which the calculation of the market risk exposure of interest rate swaps was considered. Two alternative approaches to dealing with interest rate swaps were suggested. Under the first approach, a swap book is treated as a portfolio of debt securities with the same interest rate sensitivity. These “converted” debt securities are then treated as simple bond positions in the net capital calculation. Under the second approach, swaps are assigned to maturity bands currently used by government securities, and then a capital charge of 0 percent to 6 percent of the notional value, depending on the maturity, is applied to each swap. The capital charges for long and short positions in swaps are allowed to offset each other depending on the relative maturities of the swaps. In March 1994, the SEC issued a proposal to modify the capital requirements for brokers or dealers by allowing them to use a theoretical pricing model developed by the Options Clearing Corporation when calculating capital charges for listed options and related positions.

The Commodity Futures Trading Commission (CFTC) follows a similar haircut approach to setting capital requirements for brokers and dealers engaged in commodity futures transactions. The approaches used by the SEC and the CFTC are in contrast to the “risk-weighted capital approach” in the 1988 Basle accord for banks and the recently proposed “building-block approach” by the Basle Committee. Generally, there are also differences in the definition of capital between bank and securities firm regulators.

In order to promote harmonization of regulations applied to securities companies in different countries, the International Organization of Securities Commissions has devoted substantial effort to setting up global capital standards. To date, the effort has not been brought to completion because of disagreements within its Technical Committee.

Accounting and Disclosure

As a result of the complicated nature of the regulatory structure for the OTC derivative markets, data on OTC derivative market activities are collected by a large number of regulators and concerned parties under different regulatory and accounting regimes in many countries. However, given the lack of uniformity of reporting documents and the discrepancies in reporting requirements, it is very difficult to aggregate all the information to give a clear picture of the market situation. This is further complicated by the fact that OTC derivative positions are generally off-balance sheet items and are not readily available from institutions’ financial statements.

Given that all major securities houses in the United States are engaged in OTC derivatives activities, either directly or through subsidiaries, information on the risk exposure of these companies from the derivatives activities of their subsidiaries is needed. To this end, both the SEC and the CFTC have required brokers and dealers with affiliates engaging in OTC derivatives trading to report on the portfolios of these units. In February 1994, the CFTC proposed a set of rules to further improve disclosure. These rules require a company unit that engages in swaps transactions to file quarterly and annual reports on its financial situation and to provide information on how it monitors and controls the risks related to swap trading. Some regulators have emphasized the importance of identifying the source of revenue for market participants, especially dealers. The concern is that the current lack of information has made it impossible for the regulators to tell whether the many profitable dealers are serving a financial intermediation function or not—or if most of their profits are from proprietary trading or speculation.

To improve the international coordination of disclosure requirements, the SEC, the CFTC, and the Securities and Investment Board in the United Kingdom announced in March 1994 that they will share their information on the OTC derivative markets.

The broad range of OTC derivatives products and the ease with which new tailor-made OTC products can be created has significantly complicated the application of existing instrument-specific accounting standards, many of which were issued well before many OTC derivative products were created. Furthermore, the accounting method used can be affected by whether or not the transaction is used for hedging purposes. Under the so-called hedge accounting approach, hedges are reported as related to the position being hedged and are not marked to market. The issue is complicated if the hedge is imperfect.

Efforts are under way to write new accounting rules for OTC derivative transactions. In the United States, the Financial Accounting Standards Board (FASB) has introduced Financial Accounting Standard (FAS) numbers 105 and 107, and Interpretation number 39. FAS 105 requires that the notional amount of OTC derivative positions be reported in the firm’s financial statement or in the accompanying notes together with their nature and terms. It also requires that information on the concentration of credit risk to groups of counterparties be reported. FAS 107 requires the disclosure of the fair value of related on- and off-balance sheet derivative instruments. FASB Interpretation 39 allows the offsetting of derivatives positions with the same counterparty under a master netting arrangement. The FASB is still working on improving the accounting and disclosure of OTC derivative transactions.

In its July 1993 report, the Group of Thirty recommended that the agencies that set national accounting standards should (i) provide comprehensive guidance on the accounting and reporting of transactions in derivatives and (ii) work toward international harmonization of standards.20 The Group also suggested that firms should disclose the following information: (i) management’s attitude toward financial risks; (ii) how instruments are used; (iii) how risk is monitored and controlled; (iv) accounting policies; (v) analysis of credit risk; and (vi) the extent of dealers’ activities in financial instruments.

The Group of Thirty further recommended that dealers should regularly perform simulations to determine the market risk of their portfolio. It is important for the market participants to have a risk management unit, which is independent of the trading units, to monitor transactions, develop risk limits, identify various risk components, and review the pricing models and valuation systems used by the traders.21

In its study of the OTC derivatives markets, the Group of Thirty proposed that regulators should recognize netting arrangements to the extent that they are legally enforceable. The Group also recommended that dealers and end-users should be encouraged to use a common master agreement with all counterparties to document existing and future derivatives transactions, including foreign exchange forwards and options.

Annex III Developments in International Financial Markets

This annex discusses recent developments in international financial markets, beginning with the markets for fixed-income securities. The first section chronicles the decline in yields in 1993 and the remarkable growth in the market. The second section discusses developments in the international equity market, where activity reached record nominal levels. The third section discusses international banking activity in 1993, including the market for syndicated loans. The annex concludes with an update of developments in the banking sectors of some of the Nordic countries and of Japan.

International Bond Market Developments in 1993

A continued decline in long-term interest rates, exchange rate turmoil in Europe, and increased demand for official financing contributed to a record increase in bond issues in 1993. Yields on ten-year government bonds issued by the major industrial countries ended the year lower than their opening levels (Chart 5), as monetary authorities in each of these countries attempted to reverse a general slowdown in growth through declines in policy interest rates. In the United States, the federal funds rate did not decline further in 1993 but remained at a low level of 3.0 percent. Yields on U.S. bonds declined by 102 basis points from December 1992 to December 1993, while those on French Government bonds fell by 245 basis points and Italian yields fell by 504 basis points. These interest rate declines were reflected in the international bond markets. Yields on deutsche mark Eurobonds fell by 140 basis points from the end of 1992 to the end of 1993, while yields on dollar Eurobonds fell by 110 basis points (Table 11).

Chart 5.
Chart 5.

Long-Term Government Bond Yield for Seven Major Industrial Countries, January 1990–March 19941

(In percent)

Source: International Monetary Fund, World Economic Outlook.1 United States: 10-year treasury bond. Japan: Before January 6, 1992, refers to over-the-counter 10-year government bonds with longest residual maturity; thereafter, refers to benchmark 10-year government bond. Germany: Government bonds with maturities of 9–10 years. France: Long-term (7–10 year) government bond (Emprunts d’Etat à long terme). Italy: Before June 1991, refers to government bonds with 2–4 years’ residual maturities; thereafter, refers to 10-year government bonds. United Kingdom: Medium-dated (10 year) government stock. Canada: Government bonds with residual maturities of over 10 years.
Table 11

Developments in International Bond Markets

article image
Sources: Bank for International Settlements, International Banking and Financial Market Developments; Organization for Economic Cooperation and Development, Financial Market Trends and Financial Statistics Monthly; and IMF staff estimates.

Gross issues less scheduled repayments and early redemption.

All developing countries except the seven offshore centers (listed in footnote 3).

The Bahamas, Bahrain, the Cayman Islands, Hong Kong, the Netherlands Antilles, Panama, and Singapore.

European currency unit.

Three-month deposits, at end of period.

Bonds with remaining maturity of 7–15 years, at end of period.

Overall, gross bond issues increased for the third consecutive year in 1993, reaching a record $481 billion, an increase of 44 percent over the previous record in 1992 (Table 11). Since redemptions and early repayments increased by only 27 percent over the 1992 level, net issues of bonds increased by 78 percent to $198 billion. Among total bond issues, straight (fixed rate) bonds accounted for a record $369 billion, a 77 percent share, compared with $265 billion in 1992, as borrowers took advantage of the low interest rates (Table 12). A sharp decline in amortization resulted in an increase of more than 200 percent in net issues of such bonds.

Table 12

Sources of International Capital Markets Financing

article image
Sources: Bank of New York; Baring Securities; and Organization for Economic Cooperation and Development, Financial Market Trends, various issues.

Data for 1993 are estimates.

The floating rate note (FRN) sector saw even more impressive growth in 1993, with gross issues rising 60 percent to $70 billion and net issues increasing by approximately 270 percent. Particularly during the latter part of the year, the anticipation of an increase in official interest rates in the United States provided an incentive for lenders to avoid committing to fixed rates. The FRN sector also saw an increased use of structured issues such as “collars”—in which the lender is assured of receiving a minimum interest rate while the borrower is protected by a ceiling on the rate it would pay in the event of large increases in the reference rate. Other variants on the structured-note theme included “corridors” or “range floaters” and SURFs (step-up recovery floaters), in which the coupon is linked to the Constant Maturity Treasuries index rather than the shorter-term LIBOR. SURFs provide higher returns to investors as long as the yield curve is upward-sloping and getting steeper. As interest rate uncertainty increased, FRN issues grew in popularity at the expense of straight bonds. This development was particularly pronounced in the first quarter of 1994 as straight bond issues reportedly declined by 30 percent while FRN issues expanded by 130 percent.

The search for higher yields contributed to a modest recovery in the equity-related bond market. Such bonds were popular with Japanese borrowers in the mid- to late-1980s, but the decline in the Japanese stock market in 1990 and the ensuing recession in Japan resulted in a decline in activity in that segment of the bond market. The increase in equity returns in some of the industrial countries in 1993 provided an opportunity for this vehicle to expand (Chart 6). Gross issues of convertible bonds and bonds with warrants increased by 85 percent to $39 billion in 1993. However, a bunching of redemptions resulted in a net decline of $58 billion in outstanding issues.

Chart 6.
Chart 6.

Major Industrial Countries: Changes in Stock Market Indices, January 1990–May 1994

(Twelve-month changes, in percent)

Source: International Monetary Fund, World Economic Outlook.

With long-term interest rates falling sharply and yield curves in the United States and the United Kingdom flattening out during the second half of the year (Chart 7), borrowers sought to lock in the low interest rates for a longer period of time. Indeed, one of the notable achievements of the bond markets in 1993 was a lengthening of maturities. While, previously, Eurobond issues of longer than 10 years had been relatively uncommon, bonds with maturities as long as 30 years were very successfully issued in 1993. This development was most pronounced in the U.S. domestic market, which saw three issues of 100-year bonds. Long-term bonds benefited from strong demand from prudential institutions, which were keen to improve the maturity match between their assets and liabilities. Even narrow spreads above government securities were sufficient to attract this type of investor.

Chart 7.
Chart 7.

Major Industrial Countries: Yield Differentials, January 1990–April 19941

(In percent)

Sources: Organization for Economic Cooperation and Development, Financial Statistics Monthly; Nikkei Telecom; and IMF staff estimates.1 Government bond yield minus three-month treasury bill rate, except for Germany and Japan.2Government bond yield minus three-month Gensaki rate.3Government bond yield minus three-month FIBOR rate.

The turmoil in European currency markets in September 1992 gave rise to a significant increase in sovereign borrowing on the international market as the countries that had defended their currencies attempted to replenish their foreign exchange reserves. To this source of demand was added a cyclical demand arising from weakening fiscal positions. Sovereign borrowers raised $104 billion in international bonds in 1993, up from $64 billion in 1992, and increased their share of gross bond issues to 22 percent from 19 percent (Table 11). Banks also increased their share of total issues by raising $110 billion in the bond markets in 1993, compared with $67 billion in 1992.

Another broad characteristic of the market in 1993 was the ease with which very large issues were marketed. Liquidity concerns have plagued the international bond market since its early days, and in 1993 some issuers took action to try to improve the liquidity of their bonds—and thereby capture the liquidity premium investors appear willing to pay—by issuing larger bonds and by offering to provide a secondary market. Much as governments have discovered that they can lower their borrowing costs in domestic markets by providing large benchmark issues rather than issuing a larger number of smaller-sized bonds with different characteristics, issuers in the international bond market have increasingly turned to “jumbo” bonds. Whereas only a couple of years ago most bonds were in the $50–100 million range, in 1993 bonds exceeding $1 billion were common in the international market, and there were some very large individual issues. For example, the World Bank issued a DM 3 billion global bond and Italy issued $5.5 billion in 10- and 30-year global bonds together. The World Bank deutsche mark global bond can be traded on both the German and U.S. markets despite their very different clearing systems. Some issuers deliberately created new benchmarks out of existing outstanding stocks—Italy, for example, offered to exchange all its outstanding straight bonds for new FRNs with a common maturity-while others attempted to provide a secondary market. The World Bank, for example, has required the seven underwriters of its $5 billion European medium-term notes (EMTNs) to post public bid prices.

Turnover on the secondary market increased significantly in 1993, exceeding the rate of increase in net bonds outstanding. Eurobond trading through Euro-clear and Cedel increased by 11 percent to $4,426 billion in 1993, while the net outstanding stock of bonds increased by 9 percent to $1,850 billion.

Another important development in 1993 was the globalization of the bond market both in terms of investor base and currency of denomination. The market for global bonds—bonds registered with the SEC and marketed and traded in North America, Europe, and Asia—grew markedly in 1993, with new issues raising $34 billion—including the first sovereign FRN global bond—compared with approximately $25 billion raised in 1992 and $15 billion in 1991.1 Formerly the preserve of industrial country governments and agencies and supranational borrowers, the global bond market opened up to private issuers and to issuers from developing countries—including Argentina and China—in 1993. The attraction for this vehicle has continued in 1994, with over $10 billion in new issues in the first two months alone. Somewhat paradoxically perhaps, 1993 also witnessed the coming of age of regional bond markets in which borrowers seek to tap a local rather than global market. The Dragon bond market in particular saw an increase in activity with issues totaling $3.2 billion.

With the emphasis on long-term issues both from issuers and investors, the shorter end of the market registered less impressive growth in 1993, with gross Euronote issues increasing by 18 percent to $159 billion.2 The most dynamic segment of this market continued to be the EMTN facilities, gross issues of which increased to a record $113 billion. Since 1991, the EMTN sector has rapidly overtaken the Euro-commercial paper (ECP) market. Since EMTNs are generally relatively short-term, small issues, they provide an attractive point of entry into the international bond market for new borrowers. The key to the success of the EMTN market is its highly flexible structure provided by the ability to register facilities without having to specify exactly how the issues will be structured. Thus, once the program is registered, it can be accessed very quickly in a wide range of currencies and maturities and with a seemingly unlimited range of credit enhancements. This flexibility has given rise to substantial “reverse enquiry” issues in which investors will communicate their preferences for maturity, currency of denomination, and embedded derivatives to issuers through their program managers. Thus, notes can be designed to fit precisely the requirements of particular investors and then placed directly with them. In 1993 between 50 percent and 60 percent of EMTN issues are believed to have included some kind of structure allowing investors to take positions in other markets. The use of structured issues—which link the return on the bond to prices of other assets such as equity or commodities—is used to provide more attractive returns to investors seeking to combine some minimum yield with a play on a particular market. At the same time, the increased use of underwriting methods to distribute EMTNs with sizes and maturities similar to those of Eurobonds means that the expansion of the EMTN market is increasingly at the expense of both the commercial paper and the bond markets.

As in previous years, the U.S. dollar was the most frequent currency of denomination for bonds in 1993. However, with the expansion of currency swap markets, the importance of this statistic is less clear, since many dollar issues are swapped into other currencies. With the virtual collapse of the ECU bond market in the last quarter of 1992 following the first ERM crisis, issues of bonds denominated in ECU declined from 6.4 percent of total gross issues in 1992 to only 1.5 percent in 1993. The share of issues in deutsche mark increased as European governments restored their foreign exchange reserves and German regional governments increased their use of international bond markets. The share of bonds issued in pounds sterling also rose as U.K. interest rates fell after sterling’s suspension from the ERM. The main development in terms of currency distribution of new issues, however, was the increased use of currencies other than the dollar, the yen, the deutsche mark, and the pound sterling. The expansion of issues into a broader range of currencies was facilitated by deregulation in a number of markets. For example, the French Government approved the use of the franc in the EMTN market; the Italian Government eliminated the stamp tax on government bond trading on regulated exchanges and introduced a more efficient system of withholding tax reimbursements for investors from countries with double taxation agreements with Italy; and Portugal abolished withholding taxes on foreign investors in government bonds. The most significant liberalization, however, was implemented in Japan, where financial deregulation has proceeded rapidly in recent years. The Japanese Government eliminated the 90-day “lockup” on Euro-yen bonds issued by governments and international institutions and reduced the credit rating requirements for corporate issuers in the domestic and foreign markets.

International Equity Market Developments

Reflecting the greater recourse to securities markets as an alternative to bank financing—and benefiting from price rises on most exchanges—cross-border equity issues increased by 45 percent in 1993 to $37 billion (Table 12). Of this total, the United States alone accounted for approximately $10 billion in 1993, while the developing countries issued $11 billion in new equity, up from about $7 billion in 1992. An increasingly important segment of the international equity market is the market for depository receipts. Total net issues of these instruments by all countries reached $9.5 billion in 1993, an increase of 81 percent over $5.3 billion in 1992 (Table 12).

Privatization has played an important role in expanding the international primary market for equity. In 1993, an estimated $8.9 billion in privatized equity was sold to foreign investors. The central governments of OECD countries alone have announced approximately $196 billion in privatizations to be completed by the end of the century.

More important perhaps than the expansion in the international primary market, the significant price increases on stock exchanges worldwide—particularly those of the emerging market economies-increased the attractiveness of international portfolio investment.3 For a U.S.-based investor, for example, investment in foreign markets proved highly lucrative in 1993. All of the emerging stock markets tracked by the International Finance Corporation (IFC), except for China, Nigeria, and Venezuela, outperformed the Standard and Poor’s (S&P) 500 index in dollar terms. In addition, many of the European markets performed relatively well; however, the highest returns were found in the emerging markets. As discussed in Annex IV, the proliferation of country funds has greatly facilitated access by retail investors in industrial countries to foreign markets, especially the emerging markets.

In the United States in particular, the search for higher returns on savings than have been offered by banks led individuals increasingly to invest in mutual funds and, among them, country funds.

Net international equity trading reached an estimated $159 billion in 1993, up sharply from $53 billion in 1992, and by far the highest volume of cross-border trading estimated since 1986 (Table 13). U.S. investors once again provided most of the demand, investing an unprecedented $66 billion in foreign equity while benefiting from inward investment of only $21 billion. Although European investors showed the largest annual increase in investment outflows, from $7.6 billion in 1992 to $45.2 billion in 1993, the European markets again benefited the most from international portfolio flows, with sales to foreign investors rising from $25 billion in 1992 to $56 billion in 1993.

Table 13

Net Cross-Border Equity Flows1

(In billions of U.S. dollars)

article image
Sources: Baring Securities; and Howell and others (1993).

Data for 1993 are estimates.

Africa, Middle East, and Eastern Europe.

It was, however, the scale of investment in the emerging markets that attracted the most attention in 1993. Sales of equity to nonresidents more than doubled in 1993 to $52 billion. The increase was greatest among the Asian markets, which recorded net equity sales of $30 billion, up from $11 billion in 1992. Latin American markets also more than doubled their sales of equity to foreign investors. Emerging market investors ventured abroad on a greater scale in 1993 than they had previously. Investors in these markets spent $21 billion on equities listed on the exchanges of the more developed markets.

International Banking Activity

Banks’ cross-border activities continued to expand slowly in 1993—as they did in 1992—after having contracted sharply in 1991.4 The aggregate figures, however, conceal markedly different experiences between different types of counterparties in 1993. Lending to nonbank borrowers increased by the second largest margin since 1982, while interbank lending rose by the smallest amount in any year other than 1991 (when the interbank market contracted sharply). A feature common to both market segments since 1990 is the slowdown in deposit growth relative to the experience of the mid- to late-1980s. This difference between lending to banks and nonbanks was observed for industrial countries, but not for developing countries, where lending to nonbanks rose by a much smaller amount than in 1992. Reporting banks continued to reduce their net exposure to all counterparties in the countries in transition.

Aggregate cross-border bank lending increased by $257 billion in 1993, down slightly from an increase of $262 billion in 1992 and well below the increases achieved in the years since 1985 (Table 14). Lending to industrial country counterparties increased by $190 billion, more than in 1992 but still considerably lower than the increases observed in the late 1980s, while claims on developing country borrowers increased by $38 billion, which was $7 billion less than in 1992. With cross-border liabilities increasing by $195 billion in 1993, net claims actually increased by $63 billion, compared with an increase of $95 billion a year earlier.

Table 14

Changes in Banks’ Cross-Border Claims and Liabilities1

(In billions of U. S. dollars)

article image
Sources: Bank for International Settlements, data reported to the IMF on currency distribution of banks’ external accounts; International Monetary Fund, International Financial Statistics; and IMF staff estimates.

Data on changes in claims and liabilities are derived from stock data on the reporting countries’ assets and liabilities and are net of changes due to exchange rate movements.

As measured by changes in the outstanding liabilities of borrowing countries defined as cross-border interbank accounts by residence of borrowing bank plus international bank credit to nonbanks by residence of borrower.

Excluding offshore centers. Data include some accumulation of interest arrears and reduction in bank claims resulting from debt conversions, sales, and write-offs.

The Bahamas, Bahrain, the Cayman Islands, Hong Kong, the Netherlands Antilles, Panama, and Singapore.

The difference between the amount that countries report as their banks’ positions with nonresident nonbanks in their monetary statistics and the amount that all banks in major financial centers report as their positions with nonbanks in each country.

As measured by changes in the outstanding assets of depositing countries, defined as cross-border interbank accounts by residence of lending bank, plus cross-border bank deposits of nonbanks by residence of depositor.

Change in claims minus change in liabilities.

While the overall figures suggest a declining importance for international bank lending, in fact lending to nonbank borrowers has been fairly robust during the 1990s. Net international claims on non-banks rose by $147 billion in 1993, almost twice the increase in 1992 and the largest increase since data became available in 1983 (Table A8). This expansion in net claims was due both to a rapid rise in gross claims and a slowdown in liabilities. Total lending rose by $191 billion, more than in any year since 1982 except 1990. The increase in lending was most visible in the U.S., German, and U.K. banks. Cross-border lending by Japanese banks, on the other hand, fell in the second half of the year as these institutions continued their retrenchment from international lending in response to domestic balance sheet difficulties.

On the other side of the balance sheet, nonbanks’ cross-border deposits grew by only $43 billion in 1993, compared with an increase of $65 billion in 1992. Liabilities of nonbanks in Japan and the United States, in particular, declined for the third consecutive year. Banks in France also saw a decline in cross-border nonbank liabilities. The decline in cross-border deposits by nonbanks parallels the recent trend toward weaker deposit bases for banks in their domestic markets. In the United States, for example, there has been a tendency for nonbanks, including households, to look for higher returns on their savings in nonbank financial institutions such as mutual funds (see Box 2). Part of the explanation for the coexistence of a slowdown in deposits and an expansion in credit lies in the significant increase in banks’ purchases of international corporate bonds. Hence, banks are increasingly extending credit not through loans but through the securities markets.5

In sharp contrast to the experience of the nonbank sector, gross interbank claims rose by a modest $67 billion in 1993. Industrial countries accounted for most of the slowdown, with cross-border lending growing by only $31 billion, less than one tenth of the increases observed during 1986–90 (Table A9). Lending to developing country banks actually picked up slightly, increasing by $34 billion compared with $22 billion in 1992. A sharp rise in interbank liabilities—particularly to industrial countries—resulted in a strong contraction in net cross-border interbank claims of $85 billion in 1993 after a $20 billion increase in 1992. This reversal was due mostly to a decline of $73 billion in net lending to industrial country banks, although net claims on developing country banks fell by $10 billion, continuing the series of annual decreases dating back to 1987.

Developments in the interbank market in 1993 reflected a number of influences. The large buildup of cross-border interbank claims within Europe during the attack on the European ERM in September 1992—which saw intra-European interbank claims rise by $146 billion—was partially reversed during the subsequent three quarters. The unwinding of such positions in the first quarter also resulted in a large fall in net claims of U.S. banks. However, renewed tensions in the foreign exchange market led to a further accumulation of cross-border claims within Europe in the third quarter of 1993. Interbank activity within Europe was also buoyed by an increase in securities transactions and large currency transfers in the fourth quarter. For example, German banks’ net claims on other, mostly European, banks increased strongly in the fourth quarter, as a result of the decision to apply withholding tax to interest earned on foreign mutual funds. German residents responded by liquidating investments in offshore funds. The Japanese banking industry continued its retrenchment from the international interbank market, with the third consecutive year-on-year decline in total claims. Finally, the favorable borrowing conditions available on international securities markets caused banks to rely less on the interbank market. Bond issues by banks reached record levels in 1993, increasing by 64 percent to $110 billion (Table 11).

Reflecting the weakening of overall international bank lending, the syndicated loan market continued a trend of declining net issues that began in 1991. Although the total value of syndicated loans signed in 1993 rose by 10 percent to $130 billion, an increase in redemptions resulted in a 20 percent decline in net new lending to $70 billion. This development is attributable to the reduction in long-term bond rates, which made such issues more attractive and increased concern for credit quality among lending banks. New lending in the first quarter of 1994 reportedly declined by 25 percent over the same period in 1993, owing in part to increased uncertainty over interest rates.

Banking Developments in Three Nordic Countries and in Japan

As discussed in last year’s report, International Capital Markets: Part II, rapid financial liberalization in some industrial countries had led banks in the 1980s to enter into new activities, as well as to increase their exposure to real estate. A decline in asset prices at the end of the decade and in the early 1990s reduced both borrowers’ ability to repay loans and the value of loan collateral, resulting in significant losses on bank balance sheets. This section provides a brief update on efforts undertaken since the first quarter of 1993 to resolve these difficulties in the banking sectors in three Nordic countries and in Japan. This year, Venezuelan banks have encountered severe difficulties, while some individual large banks in France and Spain have also registered large losses (Box 4 discusses these events).

New Episodes of Stress in Banking Systems

A banking crisis emerged in Venezuela in early 1994. The liberalization of interest rates in 1989 led to intense competition among banks for market share in deposits. In the anticipation of further reforms that would introduce universal banking and open up the sector to foreign competition, many banks attempted to increase their share of the deposit and loan markets. The most aggressive competitor was Banco Latino, which had become the second-largest bank in Venezuela by the end of 1993. By offering higher interest rates than the other banks, Banco Latino nearly doubled its deposits in 1991 alone. This strategy contributed to operating losses in 1992, since it resulted in narrower interest margins. On the other side of the balance sheet, the rapid expansion of lending—including to other members of the same financial group and to bank insiders—led to increasing loan losses. By the end of 1992, 7 percent of the bank’s loans were nonperforming, a much higher ratio than for other banks. Liquidity problems emerged in December 1993, which led to a run on deposits. On January 13, 1994, the central bank announced that Banco Latino had a deficit of Bs 26.5 billion in the check-clearing system. The bank was placed under receivership by the Superintendent of Banks and Other Financial Institutions.

Problems at eight smaller Venezuelan banks developed in the ensuing weeks as a result of a flight to quality by depositors. The deposit guarantee agency, Fogade, responded by providing Bs 446 billion in low-interest loans in exchange for at least 51 percent of the capital of these banks. This amount is in addition to the Bs 313 billion lent to Banco Latino prior to its reopening on April 4, 1994.1 The Bs 759 billion ($6.4 billion) total represents approximately 14 percent of Venezuela’s 1993 GDR To try to limit the political, monetary, and fiscal impacts of this rescue operation, the Government has increased deposit insurance coverage to Bs 4 million (effectively covering almost all of Banco Latino’s deposits), issued zero-coupon bonds to soak up the extra liquidity, and imposed a 0.75 percent tax on bank withdrawals. The authorities have proposed that the eight banks, which are now effectively state owned, will be liquidated or merged with stronger banks.

Serious problems also surfaced at two of the largest banks in Europe. On December 28, 1993, the Bank of Spain announced that it was intervening to support Banco Español de Credito (Banesto), the fourth largest bank in Spain. After two years of very rapid expansion, the bank’s problem loans grew too large to be covered by investment income; at the end of 1993, approximately 20 percent of the bank’s loans were nonperforming. This news came only six months after the bank had successfully raised in June 1993 Ptas 95 billion in capital, at which time nonperforming loans were an estimated 9 percent of total loans. The interim management of Banesto, installed in December 1993, estimated that provisions and write-offs totaling Ptas 605 billion (approximately $4.3 billion) were needed. Under an agreement with the Bank of Spain, Banesto assumed Ptas 320 billion of the loss, wiping out its reserves and almost half its share capital. The Bank of Spain and the other commercial banks shared equally in Ptas 285 billion in losses. In addition, Ptas 600 billion in problem loans with an estimated market value of Ptas 315 billion was transferred to the Deposit Guarantee Fund. On April 25, 1994, Banesto was sold at auction to Banco Santander for Ptas 313 billion, making the latter the largest Spanish bank with total assets of Ptas 17.1 trillion.

Difficulties at Crédit Lyonnais, the largest banking group in Europe, became more widely known in February 1994, although the replacement of its executive chairman in November 1993 had signaled that the bank was in trouble. Here again, the problem appears to have stemmed from a rapid expansion of lending in the late 1980s and early 1990s, a downturn in the economy, and a decline in prices for commercial property used as collateral for loans. In early 1994, the bank reported a loss of F 6.9 billion for 1993 and a significant shortfall in capital was recognized. In March 1994, an agreement was reached that called for a capital injection of F 4.9 billion ($850 million) from the French Government and two state-owned companies, the transfer of F 40 billion in doubtful real estate loans—partially guaranteed by the Government—to the Treasury, and asset sales of over F 35 billion over two years. Once again, problems appear to have come to a head relatively quickly. At the end of June 1993, problem loans were reported to be only F 24 billion.

1 Banco Latino also benefited from Bs 48 billion in loans from other banks to resolve its liquidity problems in January 1994.

Developments in Nordic Banking

In three Nordic countries—Finland, Norway, and Sweden—a rapid increase in credit and the expansion of bank activities into riskier, less familiar markets—such as real estate lending—was accompanied by the liberalization of the financial sectors in the early to mid-1980s.6 There, as in Japan, a decline in real estate prices, combined with other country-specific factors, triggered the emergence of substantial stocks of nonperforming loans. In these three countries, the Governments declared their willingness to provide the necessary support for the banking system. In Finland and Sweden, unlike in Norway and Japan, the Governments responded with explicit guarantees of the banks’ liabilities and, in many cases, with equity injections, which have resulted in governments becoming majority shareholders in a number of institutions.

The crises that afflicted the banking industries in Finland, Norway, and Sweden appeared to have eased significantly in 1993, as declining interest rates and exchange rate depreciation contributed to increased profits, which allowed many banks to raise capital on domestic and international markets.7 These crises have been costly. The Governments of Finland, Norway, and Sweden have together committed about $20 billion in capital injections and guarantees in support of their banks since the onset of the crisis.8 The individual country amounts represent 8.2 percent of 1992 GDP in Finland, 4.0 percent in Norway, and 6.4 percent in Sweden; in each case, the commitments have exceeded the value of equity in the banking system when the crisis emerged.

In late 1992 and into early 1993, interest rates in these three Nordic countries fell sharply, in part following a global trend, but partly also because of a change in policy away from defending fixed exchange rates. The lower rates resulted in a widening of interest margins, which directly led to an improvement in operating profits. In addition, lower interest rates translated into capital gains on banks’ bond and equity portfolios, which in turn provided additional income. Finally, securities trading commissions in some banks were favorably affected both by the European exchange rate turmoil in late 1992 and early 1993, and by a general increase in securities trading activity (as bond and share prices rose).

Table 15 shows the improvement in Nordic banks’ income statements in 1993. With one exception, all of the major banks in these countries recorded increases in operating profits—or declines in losses—in 1993. In most instances, the most important source of increased income was the securities trading account. For example, in Norway, securities trading earned profits of NKr 1.4 billion in 1993, after a NKr 600 million loss in 1992—accompanied by modest increases in net interest income. The turnaround in profits is most noticeable in Norway where net income increased by NKr 9 billion. The 1993 results for Sweden are not directly comparable to the 1992 results because of the substantial restructuring of the banking system in that country. Both Nordbanken and Gota Bank had most of their nonperforming loans removed from their balance sheets and received large capital injections from the Bank Support Authority (BSA). With the exception of Gota Bank, all of the major Swedish banks enjoyed improved profits in 1993. In Finland, although only one of the major banks, Okobank, was profitable, losses at the other institutions declined considerably—by more than 40 percent at Postipankki and Skopbank.

Table 15

Nordic Countries: Performance of Major Banks1

(In billions of local currency)

article image
Sources: IBCA Limited; and IMF staff estimates.

Includes the following banks: for Finland, Kansallis-Osake-Pankki, Okobank, Postipankki, Skopbank, and Union Bank of Finland; for Norway, Christiania Bank, Den norske Bank, Fokus Bank, and Union Bank of Norway; and for Sweden, Gota Bank (data not yet available for 1993; preliminary estimates have net interest income of SKr 1.7 billion and operating profit of SKr–13.57 billion), Nordbanken, Skandinaviska Enskilda Banken, Swedbank, and Svenska Handelsbanken (Securum and Retriva are not included).

In Finland, the increase in profits made it unnecessary for some banks to seek government assistance. The largest Finnish bank, Kansallis-Osake-Pankki (KOP), announced in February 1993 a Fmk 3 billion capital restructuring proposal. This involved cutting the nominal value of existing shares in half, and then issuing bonus shares and a rights issue with a combined value equal to the new value of old equity. In October 1993, after the parliament’s decision to adopt the “good bank/bad bank” model, KOP announced that its investment banking operations were to be spun off into a new unit, called Prospectus. KOP raised a total of Fmk 4.5 billion in new equity in 1993—38 percent of its end-1992 capital—through sales of shares to the Government, rights issues, and international offerings. The Union Bank of Finland also announced an ambitious capital raising campaign in August 1993 that involved a Fmk 1.158 billion rights issue, a Fmk 300 million direct offering, and finally the sale of Fmk 1 billion in new shares.

On the other hand, the first bank taken over by the Finnish Government, Skopbank, continued to require government support in 1993. In May 1993 the Government Guarantee Fund (GGF) injected Fmk 700 million in preference capital into the bank. Skopbank went to the international capital markets in July and September 1993 to borrow a total of $150 million (by issuing a three-year bond), but in December it announced that it needed a further capital injection of Fmk 350 million in order to meet its capital requirements. This brought the total amount of GGF funds invested in this bank to Fmk 3.7 billion, and the GGF now owns 53 percent of Skop-bank’s equity and 63 percent of its voting shares. The Bank of Finland had provided Fmk 11.1 billion in assistance to Skopbank by the end of 1993.

In October 1993, the Government announced that the GGF had negotiated the takeover of the Savings Bank of Finland by KOP, Unitas, Postipankki, and the Okobank group.9 Each of these four groups would acquire a quarter of the Savings Bank for a price of Fmk 1.4 billion; each will receive about Fmk 12 billion of assets. All of the Savings Bank’s bad loans will be transferred to a separate company to be wholly owned by the GGF.

Following on favorable midyear results and the success of KOP’s equity issues, the Norwegian banks have also raised additional equity, which has in some cases reduced the Government’s equity share to 70 percent. Swedish commercial banks reported significantly improved results in the first half of 1993—particularly Nordbanken which, as a result of the divestiture of most of its nonperforming loans, recorded a SKr 12.6 billion turnaround in pretax profits to SKr 2.5 billion. Nordbanken’s improved position provided the necessary support for the issue of a $30 million bond. Following the announcement of an increase in net profit to SKr 837 million from SKr 658 million during the first half of 1992, Svenska Handelsbanken, the largest commercial bank, announced a SKr 2.5 billion rights offering. Most significantly, the improved operating environment reduced net losses at Skandinaviska Enskilda Banken (SEB) to SKr 298 million from SKr 2.5 billion in the first half of 1993. As a result, in August 1993 SEB withdrew its application for government support and announced plans for a SKr 5.3 billion rights issue. SEB had also employed the “good bank/bad bank” model in 1993, by transferring 513 real estate properties, worth SKr 12.4 billion, to its real estate unit, Diligentia Fastigheter.

Likewise, the remaining two Swedish applicants for government assistance withdrew their requests after posting improved results for the first half of 1993. Foreningsbanken, a former rural cooperative institution, successfully launched a SKr 3.5 billion initial public offering in December 1993, which was followed early in 1994 by a $100 million bond issue.10 Swedbank followed suit with the announcement of a proposed SKr 8 billion restructuring plan in December 1993, which will include a SKr 2.1 billion stock issue.

The situation at Gota Bank took somewhat longer to be resolved. It was taken over by the Government in December 1992 after its parent company, Gota AB, declared bankruptcy. As a result of the government guarantee, Gota was able to issue $200 million in two equal tranches of five- and eight-year bonds. In the fall of 1993, the Government announced its intention to sell Gota Bank and began approaching other banks, domestic and foreign. Eventually all of the banks that expressed interest, except Nordbanken, declined to bid. Consequently, in December 1993, these two banks were merged. Prior to the merger, however, SKr 38 billion of Gota Bank’s bad assets—all bad loans greater than SKr 5 million—were transferred to a new company, called Retriva, which was capitalized by the Government to the amount of SKr 3.8 billion and received loan guarantees of SKr 3.5 billion. As a result of this operation, Gota Bank’s stock of nonperforming loans was reduced to 4 percent of total loans, from 36 percent. In addition, the BSA provided a capital injection of SKr 20 billion to Gota Bank in order to cover remaining loan losses.

Balance Sheet Difficulties in Japan

Steep declines in equity and real estate prices after 1990 were the key factors that precipitated banking difficulties in Japan. Since a significant proportion of bank loans were secured directly or indirectly by real estate collateral or extended to real estate companies and developers, the decline in land prices soon resulted in the accumulation of increasing stocks of nonperforming loans on the balance sheets of the major banks. These difficulties were compounded since 1991 by a general slowdown in the real economy, which led to further additions to the stock of bad loans.

As discussed in Part II of last year’s International Capital Markets report, the banks’ ability to strengthen their balance sheets depends to a considerable extent on external factors such as developments in the equity and real estate markets and in the real economy more generally. While equity prices have recovered from cyclical lows and a bottoming out of real estate prices may be in sight, banks must rely on improved revenue—in particular, an increase in interest earnings—in order to be able to increase their reserves against bad loans.

With economic growth slowing sharply from mid-1991 on, monetary and fiscal actions were taken to support activity. The official discount rate was reduced from 6 percent in mid-1991 to a historic low of 1.75 percent in September 1993, and fiscal expenditure was increased in a series of four supplementary budgets announced since August 1992, the latest of which was announced in February 1994.11 Banks’ interest rate margins, which had, if anything, narrowed slightly in the early part of 1993, widened appreciably in the fourth quarter. The spread between the average long-term loan rate and the interest rate paid on large (in excess of ¥ 10 million) one- to two-year term deposits fluctuated between a high of 211 basis points and a low of 185 basis points during the first eight months, but it jumped to 256 basis points in September and finished the calendar year at 286 basis points. A similar, although less dramatic, increase is observed in spreads between short-term lending and deposit rates, and in spreads over many other sources of funding.

While these wider spreads improved banks’ operating revenue—particularly in the second half of the fiscal year—commercial property values continued to decline in 1993 and in the first quarter of 1994. Commercial real estate prices fell by 18.3 percent in Tokyo and by 19.1 percent in Osaka in 1993; further declines of 4 percent and 3.9 percent, respectively, were registered in the first quarter of 1994. Prices for commercial real estate in these cities have fallen by approximately 40 percent and 60 percent, respectively, from the peak level in 1990.

Developments in the equity and property markets are relevant for a number of reasons. As regards equity prices, banks are allowed to claim 45 percent of the unrealized gains on their equity portfolios as tier 2 capital. Thus, a decline in equity prices results in a direct decline in capital. This factor has been considerably alleviated by the permission banks received in July 1992 to include perpetual subordinated debt in tier 2 capital. Between March 1992 and March 1994, the major banks increased their subordinated debt by ¥ 2.9 trillion. The accounting for unrealized gains or losses on equity holdings, however, exposes banks to a potential problem. Since banks must report the value of their holdings at the lesser of the market value and the book value, a decline in the price of these shares must be recorded as a loss on their income statement. Land prices affect the banks’ balance sheets by changing the value of their loans. A significant proportion of banks’ lending has been extended to real estate developers, to construction firms, or to housing loan companies; real estate serves as the collateral for such loans.

Equity prices fluctuated widely in 1993. The key Nikkei 225 index rose from ¥ 16,925 at the end of 1992 to a peak of ¥ 21,148 in mid-September 1993 before falling to a low of ¥ 16,079 at the end of November. Since early January 1994, the index has fluctuated within a relatively narrower band of ¥ 19,000 to ¥ 21,000.

The continued slowdown in economic activity, combined with declines in land prices, contributed to an increase in nonperforming loans in the fiscal year 1993/94. At the end of March 1994, the 21 major banks reported nonperforming loans of ¥ 13,573 billion, up from ¥ 12,775 billion at the end of March 1993 (Table 16).12 Most of the problem assets are on the books of the city banks, whose reported nonperforming loans increased by 6.1 percent to ¥ 8,974 billion. The problems are most pronounced, however, at the trust banks, whose ¥ 2,712 billion in nonperforming loans represent almost 4 percent of their lending.

Table 16

Japan: Indicators of Banking Performance

(In billions of yen, except where indicated)

article image
Sources: Bloomberg Financial Markets; IBCA Limited; Japan, Ministry of Finance; and IMF staff estimates

Adjusted to exclude investment bond profits and losses.

Specific provisions and charge-offs against nondeveloping country loans plus losses on loans sold to the Cooperative Credit Purchasing Company (CCPC).

The banks’ first line of defense against balance sheet weakness is an increase in loan-loss reserves. There are two kinds of reserves: (i) general reserves, additions to which are exempt from tax up to the legal maximum of 0.3 percent of the value of loans made; and (ii) specific reserves held against individual loans. Until recently, banks could only create specific reserves with the approval of the Ministry of Finance’s Banking Bureau. This could be a time-consuming process because the bank would have to submit each loan for consideration individually to a Banking Bureau examiner. The tax liability of these provisions was determined by the Tax Bureau, which generally required proof of legal bankruptcy before exempting specific reserves from taxation. The February 1994 stimulus package eased these constraints somewhat by allowing banks more discretion in creating specific reserves: banks were freed of the need to present each loan for consideration and could make provisions against loans that they decided were nonperforming or doubtful.

Banks are permitted to write off losses equal to the difference between the book value and the market value of a loan when it is sold. In addition, when losses arise from loans against which sufficient provisions have not been made, they must be charged against current income; interest accrued but not paid in two years must also be charged off. However, while banks’ foreign subsidiaries have begun to make use of the secondary market for loans elsewhere, there is no such market in Japan as of yet. In the absence of a market for loans, the major Japanese banks, together with many regional banks, insurance companies, and cooperative credit institutions, set up the Cooperative Credit Purchasing Company (CCPC) in early 1993. This essentially serves as a vehicle through which banks can realize the losses on their nonperforming loans and thereby make them eligible for tax deductibility. In the 1993/94 fiscal year, ¥ 3,838 billion in loans were sold to the CCPC; the discount on these loans averaged 54 percent (Table 17).13 The 21 major banks claimed tax deductions of ¥ 1,775 billion. During the same period, banks made specific provisions of ¥ 1,378 billion and charged off ¥ 385 billion against nondeveloping-country loans.

Table 17

Business Results of the Cooperative Credit Purchasing Company (CCPC)

(In billions of local currency)

article image
Sources: Bloomberg Financial Markets; IBCA Limited; Japan, Ministry of Finance; and IMF staff estimates.

Loans sold to the CCPC are evaluated by a pricing committee that determines an approximate market value for the collateral. The loans are then purchased by the CCPC at this price, with the financing provided by a loan from the selling bank. Interest and principal payments on these loans commence only after the collateral is sold. Hence the bank essentially exchanges a nonperforming loan for a performing loan to the CCPC and obtains a tax deduction for the difference. However, since the bank remains responsible for losses incurred by the CCPC if it is unable to sell the collateral (at its new estimated value), the bank is not as clear of the problem as it would be if it had sold the loan on a secondary market.

The Ministry of Finance has recently encouraged banks to make more aggressive use of the CCPC and has permitted banks to set up similar units to purchase restructured loans of nonbank affiliates-loans that often are not already included in the estimate of nonperforming loans. Reportedly, the first such entity has been proposed by a group of banks to purchase the restructured loans due to ten non-bank affiliates of Hyogo Bank.

As of March 31, 1994, the 21 major banks reported adjusted operating earnings, defined as net interest revenue plus noninterest revenue less non-interest expenses, of ¥ 2,725 billion, 12 percent less than for 1992/93.14 Among the three groups of banks, the adjusted operating earnings of long-term credit banks suffered the largest decline, falling by 48 percent. Contrary to the experience of the other two groups, the earnings of the trust banks increased by 52 percent. Changes in net interest revenues were the driving force for changes in adjusted operating income: city banks’ net interest revenues declined by ¥218 billion and long-term credit banks’ net interest revenue declined by ¥ 185 billion. For the trust banks, in contrast, net interest revenues increased by some ¥ 75 billion.

Net operating income is an important element of banks’ efforts to resolve the bad loans problem, because higher operating income allows for faster provisioning and write-offs. Assuming an average discount of 50 percent on the value of the nonperforming loans remaining at end-March 1994, the noncollateralized portion of these bad loans was approximately ¥ 6.8 trillion;15 the combined specific reserves of the major banks at that time was ¥ 3 trillion. In 1993/94, banks’ provisions and charge-offs, including losses on CCPC loans, amounted to ¥ 3.5 trillion—130 percent of net operating income. With similar levels of provisioning, banks could increase reserves by the additional ¥ 3.8 trillion needed to fully cover the decline in portfolio value in just over a year.

However, the banks were only able to make such large provisions and write-offs by drawing on their unrealized gains on securities holdings—particularly equities. In 1993/94, the 21 major banks recorded ¥ 1.8 trillion in net profits on their equity holdings. As of March 31, 1994, the major banks had unrealized gains on equities of ¥ 19.6 trillion. While these gains can be realized by some institutions and used as a means to increase provisions, it would be difficult for all banks to do so since they hold a significant share of Japanese equities. At the end of September 1993, the 21 major banks’ equity portfolios were valued at an estimated ¥ 36 trillion, or 10 percent of the market capitalization of the first two sections of the Tokyo Stock Exchange.16

Annex IV Developments in Private Market Financing for Developing Countries

Private financing to developing countries increased substantially in 1993; it was associated with a further improvement in the terms of borrowing and strong increases in equity prices in many local stock markets. Developing country borrowers issued almost $60 billion of bonds in 1993, by far their most widely used financing instrument. Much of this inflow represented net financing, because most bonds issued so far have bullet repayments that have not yet fallen due. A large share of these bond flows went to borrowers in about six to eight countries in Asia, Europe, and Latin America, but the range of countries accessing this market continued to expand in 1993. In contrast, international equity placements by developing country issuers rose only moderately from $9 billion in 1992 to $12 billion in 1993, with an important share of these issues made through American Depository Receipts (ADRs) or Global Depository Receipts (GDRs). This estimate of equity placements, however, misses direct purchases of equity in the local stock markets, which reportedly grew significantly in 1993. Banks continued to provide financing primarily through short-term trade credits, although their medium- and long-term lending to most developing countries picked up somewhat in 1993.

In the first quarter of 1994, the increase in U.S. interest rates, inter alia, helped spark a drop in demand for emerging markets assets (relative to their peaks in the fourth quarter of 1993), as financing to developing countries tapered off, together with strong declines in bond and equity prices. Developing country bond issuance had accelerated throughout 1993 to $24 billion in the fourth quarter and then fell off to $18 billion in the first quarter of 1994. A similar pattern occurred with international equity placements, which had reached over $5 billion in the fourth quarter of 1993 before declining to $4 billion in the first quarter of 1994.

The first part of this section reviews the recent experience with securitized capital flows (international bonds, short-term debt instruments, and equities) and with bank lending. The second part discusses two aspects of the development of the market in 1993. First, it explains the broadening of the investor base and the reasons behind the expansion. Second, it discusses the factors behind the assessment of risk.

Recent Experience

International Bonds

Issuance by developing country borrowers in international bond markets continued to show impressive growth, as bond placements more than doubled to $59 billion in 1993 following a doubling in the volume of bond issues in each year since 1990 (Table A10). During the course of the year, total bond issues accelerated from $10 billion in the first quarter to over $23 billion in the fourth quarter. The average size of bonds issued increased to $135 million in 1993, somewhat higher than the average of $110 million in 1992. In May, Cementos Mexicanos (Cemex), Mexico’s largest cement producer, successfully placed a $1 billion issue, the largest Eurobond issued by a Latin American borrower to that date. In December, the Republic of Argentina launched the first global bond issue by an emerging markets issuer, a $1 billion bond payable in ten years. The share of developing countries in total bond issuance in international markets rose further to 12 percent in 1993, three times the share of 4 percent in 1991; and this share more than doubled from 7 percent in the first quarter of 1993 to 20 percent by the fourth quarter.

Bond issuance in 1993 was concentrated in three regions. Borrowers in the Western Hemisphere raised some $27 billion, with Mexico once again emerging as the leading borrower, raising over $10 billion. Besides the issue by Cemex, a wide range of Mexican corporations, such as Petroleos Mexicanos (Pemex) and several banks, placed sizable bond issues. Argentine issuers quadrupled their access to bond markets to $6 billion, which included several large issues by Telecomunicaciones. International bond offerings by Brazilian and Venezuelan entities also rose considerably, in spite of continued uncertainty about the course of economic policies and restrictions on the maturity of Brazilian bond issues. One of the Venezuelan issues was launched simultaneously in Colombia and outside Latin America, the first intra-regional bond issue in Latin America. Chile, Trinidad and Tobago, and Uruguay maintained their presence in these markets through a moderate amount of borrowing, and Colombia, Guatemala, and Peru tapped the market for the first time in many years.

Asian borrowers more than tripled their borrowing in international bond markets to $20 billion, reflecting in large part a sixfold increase in bonds issued by China and Hong Kong.1 Borrowers in Korea and Thailand sharply increased their presence in international bond markets. Both the Philippines and Malaysia entered the market for the first time in many years, with the Philippines raising over $1 billion. India also returned to the market after being absent in 1992, with a sharp pickup in issuance in the fourth quarter of 1993.

Among European developing countries, Hungary stepped up its bond issues to the equivalent of $4.8 billion in 1993, in an effort to cover its present and future large external financing needs, while Turkey borrowed $3.9 billion on international bond markets. Other regions were relatively inactive in international bond markets in 1993, with the exception of the Middle East, where Israel floated $2 billion of bonds in 1993 as part of a loan-guarantee program granted by the United States.

All types of borrowers took part in the surge in bond financing in 1993. Private sector issuers more than doubled their bond issuance from $10 billion in 1992 to $27 billion in 1993, which accounted for 46 percent of total bond issues by developing countries. Almost three fourths of the private sector bond issues were placed by issuers in four countries or regions (Hong Kong and Mexico with about $6 billion each, followed by Brazil with $4.8 billion and Argentina with $3.8 billion). Sovereign borrowers roughly tripled their bond placements between 1992 and 1993, accounting for $16 billion, or 28 percent of bonds issued in 1993. Hungary and Turkey led sovereign bond issuers with $4.5 billion and $3.7 billion of bond placements, respectively, in 1993. Other public sector borrowers issued roughly the same amount of bonds in 1993 as the sovereign borrowers.

Terms on primary issues improved for many developing country borrowers during the year. The average yield spread against comparable U.S. Treasury securities at launch for all borrowers fell from 288 basis points in the first quarter of 1993 to 241 basis points by the fourth quarter of 1993. The average yield spread for sovereigns, however, actually increased slightly, reflecting the entrance of sovereign borrowing by the Philippines, the Slovak Republic, and Venezuela at spreads of over 300 basis points. For countries with a track record in the markets, sovereign yield spreads came down. In 1993, average yield spreads for other public sector borrowers were below the spread on sovereign issues for the first time, reflecting the experience of Mexico, the Philippines, and Thailand. There was a strong variation in yield spreads across countries, with the lowest spreads for Asian borrowers, such as China, Korea, and Thailand (all below 100 basis points), while the private sector in Latin America typically paid a spread of 300 to 500 basis points. The average maturity of bond issues continued to lengthen in 1993 to about 6½ years. Maturities were shortest in the private sector. Some sovereign borrowers pursued a strategy of lengthening the maturity structure of their country’s debt; Hungary placed a 20-year bond issue, while a number of countries such as Argentina and Mexico placed 10-year bond issues.

Bond issues continued to be concentrated in three currency sectors, with the share of issues denominated in U.S. dollars, deutsche mark, or Japanese yen amounting to 95 percent of the total. The U.S. dollar sector continued to be the major funding source for developing country borrowers, even for non-U.S. investors. Most investors reportedly hedge their currency exposure and it is easier to hedge instruments denominated in U.S. dollars. German investors exhibit a strong home currency preference and tend to buy deutsche mark-denominated issues to avoid any exposure to exchange rate risk. Other reasons for the predominance of these three currencies would include their widespread use in international payments and the ease of settlement. While most borrowers—especially those in the Western Hemisphere—placed dollar-denominated bond issues, a number of countries—particularly in Europe—continued to try to diversify the currencies of denomination of their borrowings, as a means of broadening their investor base and in an effort to match the currency composition of their external assets and liabilities. Mexico has issued bonds in nine different currencies, and is followed by Hungary with bonds issued in eight currencies, including its first Matador bond and a forint medium-term note facility.2

Enhancement techniques continued to be employed by developing country issuers in 1993 to help reduce borrowing costs, and as in past years the pattern of enhancements differed among regions. Asian borrowers enhanced roughly $6 billion (35 percent) of their bond issues, relying principally on equity conversion options, while Western Hemisphere borrowers enhanced only $3 billion (12 percent) of their bond issues, mainly through put options or collateralization.

The bond market began to taper off somewhat in January and early February 1994, but experienced a more pronounced setback after U.S. interest rates were increased starting in early February. For the quarter as a whole, a decline in bond issuance compared with the last quarter of 1993 took place in all regions, and the fall-off was substantial for Hungary, Turkey, and several other major borrowers. Nonetheless, certain countries, such as China, Thailand, and Mexico, actually increased their bond issuance. Sovereign borrowers increased their level of bond issuance, while other public sector and private sector issuers registered sizable declines. The average maturity of the bonds became shorter in the first quarter, and a growing share of bonds relied on floating interest rates. Yield spreads at launch continued to improve, suggesting that lower quality borrowers lost access to the market, while the secondary market spreads rose for some countries but declined for others.

International Equity Placements

The market for international equity placements grew sevenfold between 1990 and 1992, reflecting in part the wave of privatization in several countries. In contrast to bonds, this expansion of the equity market moderated in 1993, as international equity placements reached $11.9 billion, up from $9.3 billion in 1992 (Table A11). In 1993, developing countries accounted for only 23 percent of all international equity placements, well below their share of 41 percent in 1992. Since 1990, companies from developing countries have raised over $28 billion through the international equity markets.

Latin American and Asian companies accounted for almost all of the international equity placements in 1993. In Mexico, share prices and issuance activity were subdued in the first three quarters because of uncertainty about the passage of the North American Free Trade Agreement (NAFTA), but following the passage of the Agreement in the last quarter, Mexican corporations raised $1.7 billion in international equities, while local share prices rose rapidly (Chart 8). The shares sold by Mexican corporations in the fourth quarter were placed through ADR/ GDR programs, with Grupo Televisa offering the largest GDR issue ever at $822 million (Table 18). For the year, Mexican firms issued $2.5 billion in international equities, down from the $3 billion issued in 1992. International equity issues from Argentina rose sharply in 1993 to $2.8 billion, reflecting a strong increase in share prices and the $2 billion privatization of Yacimientos Petroleros Fiscales in the second quarter. China and Hong Kong accounted for over half the equity issues from Asia.

Chart 8.
Chart 8.

Share Price Indices for Selected Emerging Markets1

(In U.S. dollar terms, December 1988 = 100)

Source: International Finance Corporation, Emerging Markets Data Base.1 IFC weekly investable price indices.2Argentina, Brazil, Chile, Colombia, Mexico, and Venezuela.3India, Indonesia, Korea, Malaysia, Pakistan, the Philippines, Taiwan Province of China, and Thailand.
Table 18

International Equity Issues by Developing Countries: Depository Receipts and Other Issues

(In billions of U.S. dollars)

article image
Sources: Euroweek; and International Financing Review.

For a number of developing countries, cross-border equity inflows have occurred through direct purchases on local exchanges. Although comprehensive statistics across countries are not available, recent estimates suggest that secondary market purchases in emerging markets by international investors amounted to some $14 billion in 1992, and many market participants believe that these flows increased substantially in 1993, with one source reporting an amount of $40 billion.3 For India, direct purchases of equity amounted to an estimated $1.4 billion in 1993, following no purchases in 1992, while in Mexico these inflows are estimated to have held steady at about $6 billion in 1992 and 1993.

In the first quarter of 1994, international equity issues declined to $4 billion, but accounted for about one third of total international equity issues, up from roughly one fifth for 1993. Enterprises in both Asia and the Western Hemisphere issued less equity, except for India which experienced about a sevenfold increase in the volume of equity placements.

Bank Lending

Banks began to show a renewed interest in lending to developing countries in 1993, although bank activity remained subdued in comparison with the financing through bonds. To limit their risk exposure, banks restricted new lending to short-term credits (typically trade credits), project finance, and loans structured using a variety of risk-mitigating techniques, including asset securitization. Medium-and long-term bank loan commitments to capital importing developing countries rose from $14.1 billion in 1992 to $18.5 billion in 1993 (Table 19).

Table 19

Bank Credit Commitments by Country or Region of Destination

(In billions of local currency)

article image
Source: Organization for Economic Cooperation and Development, Financial Statistics Monthly.

Excluding offshore banking centers.

Most notable was the increase in loan commitments to Latin America, which rose from $0.9 billion in 1992 to $2.4 billion in 1993. Each of the major restructuring countries (Argentina, Brazil, Mexico, and Venezuela) received loans ranging from $0.3 billion to $0.8 billion. Asian borrowers continued to account for three fourths of new syndicated bank credit commitments to capital importing countries, with China borrowing $3.8 billion, followed by Thailand ($3.3 billion), Korea ($2.1 billion), and Indonesia ($2.0 billion). In Europe, Turkey obtained close to $2 billion in bank loan commitments in 1993, while European countries in transition had little access to nonguaranteed medium- and long-term bank lending. Bank lending to the Middle East (including Kuwait and Saudi Arabia) fell to virtually nothing in 1993, down sharply from the historical high of $10 billion in 1991. Bank lending activity to Africa continued to be very low.

This pickup in bank lending to developing countries in 1993 was associated with a widening of the average spread on voluntary loans to these countries to 106 basis points and a fall in the average maturity to 5.6 years in 1993, compared with 6.7 years in the previous year. The spreads differed widely according to the credit quality of the borrowing country, with spreads ranging from 57 basis points on loans to Malaysia to over 200 basis points on loans to India and Venezuela.

In contrast to the primary market for bank loans, the secondary market for bank claims on developing countries was very active. In October 1993, the Emerging Markets Traders Association issued the results of the first Trading Volume Survey. This report revealed that total trading volumes in the secondary market for developing country instruments exceeded $730 billion in 1992. Latin American instruments represented more than 80 percent of the volume, led by $209 billion for Brazil, $189 billion for Mexico, and $156 billion for Argentina. It is believed that the trading volume in 1993 was on the order of $1 trillion. By the end of 1993, securitized bank debt in developing countries had increased to some $90 billion, which includes about $25 billion of bonds issued in Argentina’s 1993 debt restructuring.

Issues in Market Access

Broadening of the Investor Base

The rapid growth in private financing to developing countries in 1993 reflected a considerable broadening of the investor base. Prior to 1993, the investor base in this market displayed a regional specialization. Asian countries that maintained access to the international markets continued to attract a moderate level of investment from the mainstream institutional investors such as pension funds and insurance companies from industrial countries, especially the United States and the United Kingdom. In contrast, Latin American countries, which began to regain market access in 1989, received inflows primarily from flight capital investors—who in many cases were repatriating money to their home countries—and from wealthy individuals. Hedge funds and other highly leveraged speculators have generally remained on the sidelines of this market, but have entered for short periods when they have perceived a good profit opportunity.

Starting in late 1992, some U.S. pension funds, such as ARCO and GTE, began to purchase investments, including Brady bonds, in Latin America.4 According to market participants, U.S. mutual funds significantly increased their participation in all segments of this market in mid-1993 and were followed by another round of buying by pension funds and insurance companies. At the end of 1993, U.S. mutual funds held about 2 percent of their assets (roughly $30 billion) in emerging markets, principally in the form of equity.5 The 200 largest pension funds in the United States increased the share of their portfolio placed in investments outside the country from 5.2 percent at the end of September 1992 to 7 percent at the end of September 1993, with slightly over one half of this increase due to new cash investments and with the remainder deriving from appreciation in the market value of the assets.6 U.S. pension funds engaging in international investment tended to be defined benefit plans and included public as well as private pension plans. There is little evidence on the investments of U.S. insurance companies in emerging markets in 1993, but they probably allocated an even smaller share than pension funds to foreign assets. European institutional investors also increased their participation in this market, although more moderately, perhaps taking a cue from the heightened interest of the U.S. institutions. While the shares of the international investments allocated to emerging markets are difficult to obtain, a recent survey of institutional fund managers reports that these managers allocated 13 percent of their international portfolios in 1993 to emerging markets assets, up from 10 percent in 1992 and 2.5 percent in 1989.7 The major institutions manage very sizable portfolios. The portfolios of U.S. mutual funds amount to $2 trillion.8 U.S. pension funds and insurance companies manage almost $6 trillion, while the assets of these institutional investors in France, Germany, Japan, and the United Kingdom amounted to $5.7 trillion at the end of 1991.9

Investor preferences continue to vary widely across countries. U.S. investors continue to play the largest role and purchase debt and equity principally from Latin American and Asian issuers. U.K. investors are also active in this market and tend to buy assets in Asia and to a lesser extent in Latin America. German investors focus principally on Eastern Europe, although their interest in Latin American instruments is picking up, and they prefer deutsche mark-denominated bond issues to avoid any exchange rate risk. Japanese investors have invested a small share of their assets in securities issued by the fast-growing economies of East and Southeast Asia and by other developing countries that maintained a good debt-servicing record during the 1980s. Investors outside these four countries have largely stayed on the sidelines of this market, although they have shown modest interest, mostly for opportunities in Asia.

Factors Behind the Expansion of the Investor Base

Starting in 1989, a number of developing countries proved they could sustain a program of sound macroeconomic policies and structural reforms (especially privatization), which helped open up investment opportunities with rates of return sufficiently high to attract international investors. In addition, several of the developing countries, including China and Mexico, began to improve financial reporting and the supervision of their financial markets, which helped boost investor confidence further. The reputation of the strong performing countries rubbed off on neighboring countries that embarked on a reform path later and made investors more willing to invest at an earlier stage in the reform process.

In 1993, high returns in emerging markets relative to those in industrial countries put pressure on some institutional investors, who face short-term performance goals, to enter these markets. The dollar return on equity investments in emerging markets reached 80 percent in 1993, ten times higher than the 8 percent return on U.S. equities and almost three times the gain in other industrial country stock markets (Table 20). In 1990–92, emerging market equity returns only matched those of the U.S. stock market after a strong relative performance in 1989. In addition, the equities in a number of emerging markets have price-earnings and price-book value ratios that are low in relation to the rest of the world (Table A12); and although a great deal of caution is needed in interpreting these indicators, these may lead investors to view emerging markets equities as undervalued. The average return for equities in all emerging markets, however, masks the strong variation in returns across countries as well as the volatility over time of returns in each country (Table 20). Similar to the returns on emerging market equity, the total return on Latin American bonds rose from about 10 percent in 1992 to 18 percent in 1993.10 Total returns on bonds appear to show a much greater degree of stability than emerging market equity returns.

Table 20

Total Return on Equity in Selected Emerging Markets

(In percent)

article image
Source: International Finance Corporation.

The high rate of return compensates the investors for the fact that the returns on emerging markets equities have been considerably more volatile than those in industrial countries. Investors may also be attracted by diversification arguments since returns in emerging markets tend to be relatively uncorrelated with returns in developed countries.11

Institutional Issues

The investment decisions of insurance companies and pension funds in industrial countries are also affected by a variety of prudential limits on their ability to invest in foreign assets. Nonetheless, it is important to stress that these constraints do not appear to have been binding so far for most of these institutional investors. A recent World Bank study reviewed these restrictions in five major industrial countries and found the intensity of these restrictions varied across countries and differed for insurance companies as opposed to pension funds.12 Of these five countries, Germany imposes the strictest limits on both insurance companies and pension funds, setting a maximum portfolio share of 5 percent for foreign investments and prohibiting net exposure in any foreign currency (Table A13). In the other four countries, pension funds are treated differently from the insurance companies. In the United States, private pension funds are free from mandatory ceilings on holdings of foreign assets, but are subject to a “prudent man” rule and review by their boards or shareholders.13 Public pension funds in the United States are often subject to binding limits. In the United Kingdom, there are no limits, while in Japan there is a limit of 30 percent. Canada sets a ceiling that will reach 20 percent in 1994 on the foreign asset share of pension funds. Insurance companies in Canada, Japan, and the United Kingdom are free from any mandatory ceilings on their holdings of foreign assets, while the ceilings in the United States are set by state insurance regulators. Not surprisingly, insurance companies and pension funds in all five countries must meet certain minimum standards on the credit quality of their assets, which are often self-imposed, and in several countries these funds are subject to a prudent man rule.

The expansion of the investor base for emerging markets assets has been facilitated by the development of an infrastructure of the market. In recent years, many emerging market countries have improved the information that is available about their markets and have established investor safeguards, such as tougher laws against insider trading. Nonetheless, investments in these markets are affected by the host country’s macroeconomic policies as well as its regulations, taxes and settlement, and custodial procedures. As a result, pension funds and insurance companies tend to rely on specialized fund managers to select investments in emerging markets. Almost four fifths of the U.S. pension funds decide on an allocation for international investments and let a specialized manager place these funds as they see fit among non-U.S. investments.14

One common way to invest in emerging markets is to purchase shares in a country fund, a mutual fund that invests in a variety of emerging markets or in just a single country. The first country fund—the Mexico Fund—was launched in 1981, and by the end of 1993 there were nearly 500 country funds listed in a number of major financial centers. There is a wide variety of country funds, ranging from global funds that may invest in any emerging market, to funds dedicated to a specific region, to funds that specialize in a single country. Country funds require relatively low minimum investment and offer more liquidity than directly investing in the local market of the developing country, because the funds are traded in major financial markets. Also, a number of emerging market countries restrict portfolio capital inflows, and country funds may be the only vehicle for investing in such countries. Multi-country funds lower the cost of diversifying across emerging markets.

Many country funds are closed-end mutual funds in which a fixed number of shares are issued and shares may be sold only if another investor is willing to buy them, meaning that there are no net redemptions on the fund. In 1993, these funds invested $4 billion in emerging markets, bringing the combined portfolio of all emerging market closed-end funds to $33 billion at the end of 1993 (Table A14). The structure of a single-country closed-end fund protects the country from sudden swings in capital flows, because the shares invested in the country remain relatively stable. This also frees the fund from the risk of large net redemptions, allowing the manager to invest in less liquid assets. With a fixed number of shares, the market price of the share may trade at a discount or a premium from the net asset value of the fund, which is the total value of the assets divided by the number of shares. A discount or premium can persist for several reasons, including restrictions on the access of nonresidents to domestic capital markets of the issuing countries. At the end of 1993, closed-end equity funds on average had a discount of 11 percent.

Depository receipts are another instrument that facilitates investment in emerging markets. These instruments offer several advantages: they are denominated in and pay interest or dividends in U.S. dollars; settlement occurs in five days in the United States, which may be faster than in the issuer’s home market; tax payments on the underlying asset may be simpler, particularly when the host country has a withholding requirement; and the investor avoids global custodian safekeeping charges. Companies from many countries, both developing and industrial, rely on this mechanism to raise capital, and the number of depository receipts currently trading exceeds 900.

There are three levels of depository receipt programs, which differ in the degree of disclosure required.15 A level I program must obtain an exemption from the SEC’s registration and periodic reporting requirements, which allows it to trade these instruments only on the OTC market. This type of program is useful to get investors accustomed to trading in a particular stock, and most depository receipts use a level I program. A level II program is subject to a fairly complete registration and reporting requirement and is used mainly by issuers wishing to sell new shares through ADRs on NASDAQ or an exchange. A level III program requires full compliance with disclosure requirements of the SEC and is for foreign firms issuing new shares through a public offering.

Investors may also purchase a private placement of shares issued by a non-U.S. firm, and this private placement may take place through an ADR program. A private placement may qualify for an exemption from SEC reporting requirements if it meets a certain number of conditions, such as whether the potential investors have access to the kind of information that would be available in a registered public offering and whether they are sufficiently sophisticated. Rule 144a was adopted in 1990 to make securities privately placed under this exemption more liquid. Rule 144a permits holders of these securities to sell them freely to qualified institutional buyers (QIBs) under certain conditions without being subject to the two-year minimum holding period. Rule 144a does not apply to securities that are of the same class as “listed securities,” that is, securities that are listed on NASDAQ or an exchange. Although not necessarily required by U.S. securities law, non-U.S. companies selling newly issued stock or debt securities under Rule 144a have typically prepared extensive placement memoranda or offering circulars, and the amount of disclosure contained in such material is not much less than that required for complete disclosure under the U.S. Securities Act.

Assessment and Pricing of Risk

The nature of the credit risk associated with developing country debt instruments makes risk assessment and pricing more complex than for bonds issued in industrial countries. The likelihood of repayment for a developing country sovereign bond issue is affected by the country’s macro-economic policies and in particular by the government’s ability to service its debt obligations. The political situation also matters because of its impact on a country’s ability to sustain sound fiscal and other economic policies. Other factors also count, such as the market for the product of the issuer, the financial structure of the issuer, and the domestic legal and regulatory environment of the issuer. These factors are of course relevant for assessing risk more generally. But the recent wide-ranging structural changes in a number of these countries diminish the value of historical information about an issuer. Also, transfer risk—the possibility of restrictions on a corporation’s access to foreign exchange—can be important for many developing country bonds. Because the bond market for developing countries has expanded so quickly, investors, especially those who entered the market in 1993, are still learning to understand the available information. Likewise, many corporate issuers are new to the market and are just becoming acquainted with the needs of their investor base.

The market measures the degree of risk of a particular bond in terms of the spread over the comparable U.S. Treasury obligations—that is, the difference between the yields to maturity on the bond and on the U.S. Treasury instrument with the same maturity.16 Bond issues are sold initially at a particular spread, which may subsequently change over time through secondary market transactions. Because of the complexities associated with processing and evaluating the information, the market looks for certain benchmarks and arrives at a spread through a process of trial and error. Mexico, as the first debt-restructuring country to regain access to voluntary financing in recent years, has come to serve as the benchmark for measuring the risk of new sovereign debt issues from other developing countries in Latin America and in other regions, particularly for those with subinvestment grade ratings. In 1989, Mexico placed a bond at a spread of about 800 basis points over the comparable U.S. Treasury instrument. Mexico’s economic performance improved steadily since 1989, and over time the market was willing to accept a larger stock of Mexican sovereign bonds at spreads that declined to around 200 basis points by late 1993.

Some analysts suggest that investors evaluate developing country bonds by reference to U.S. corporate bonds with comparable credit ratings. According to Moody’s Investor Service, a U.S. Aaa corporate bond trades at a yield to maturity about 60 basis points above the yield on a long-term U.S. Treasury instrument, although this spread has at times reached 100 basis points.17 A U.S. Baa corporate bond—the lowest investment grade category—pays a spread of about 75 basis points above the Aaa bond, or 135 basis points above long-term U.S. Treasury instruments. Like the spreads on developing country bonds, these spreads are determined by market forces. But because the U.S. corporate bond market is stable and established, bond investors have access to financial information about the borrower that meets the investors’ standards, and the spreads can more accurately reflect the costs associated with delinquent payments or outright defaults. Bond default rates for different classes of U.S. borrowers are known by the market. It has been estimated, for example, that AAA bonds experienced a default rate of 0.21 percent in the ten years after issuance, while the ten-year default rate for the lowest investment grade bonds was 2.1 percent. Default was much more common in subinvestment grade issues, with 10.7 percent of BB bonds and 30.9 percent of B bonds defaulting in the ten years after issuance.18

In January, the market was anticipating that Standard and Poor’s would upgrade Mexico from the highest subinvestment grade rating, which for a U.S. corporate bond reportedly trades at a spread of about 200 basis points, to the lowest investment grade rating, which in the U.S. market would trade at a spread of about 130 basis points. As a result, the spread on Mexican sovereign issues fell to about 150 basis points, before the current market correction pushed the spread back up to about 200 basis points.

Argentina’s debt trades at a spread about 50 to 100 basis points above Mexico’s. Argentina is perceived as having made substantial progress in controlling inflation and implementing structural reforms, but is regarded as being at an earlier stage in the reform process and as having not yet resolved doubts about its external competitiveness. The market clearly regards both Brazil and Venezuela as much greater risks, and these countries’ bonds trade at spreads of about 200 to 300 basis points above Mexico’s. With regard to Eastern Europe, bonds of the Czech Republic currently obtain spreads below Mexico’s, mainly because of the Czech Republic’s investment grade rating, while Hungary’s spread has risen above Mexico’s because of the former’s high and increasing external debt. China’s $1 billion global bond issue was priced at a spread of about 80 basis points, in part because China received an investment grade rating and has relatively little external debt compared with Mexico and other countries.

Since 1989, Mexico has consistently paid the lowest spread of the major Latin American borrowers; the spread fell to below 200 basis points after NAFTA was approved in November 1993. The spread on Argentine bonds peaked at over 400 basis points in early 1993, but has hovered around 300 basis points since the completion of its bank debt restructuring in April 1993. The spread on Venezuelan bond issues has been the most volatile, rising from less than Argentina’s spread in early 1992 to more than Brazil’s spread of about 500 basis points by mid-1993. Brazilian bonds have generally paid the highest spread in Latin America.

The bonds of many countries that avoided debt restructurings in the 1980s generally have paid lower spreads compared with the Latin American borrowers other than Mexico (Chart 9). Korea has consistently paid a spread of less than 100 basis points. Turkey’s spread was 250 basis points at the end of 1993, but fluctuated considerably in early 1994, reaching 900 basis points. In the wake of the uncertainties surrounding the dissolution of Czechoslovakia, spreads on the bond of the Czech Republic reached almost 500 basis points in late 1992, even though its external debt was relatively low, but its spreads have fallen sharply since then to about 150 basis points.

Chart 9.
Chart 9.

Yield Spreads at Launch for Selected Developing Countries1

(In basis points)

Sources: International Financing Review and Financial Times.1 Yield spread measured as the difference between the bond yield of U.S. dollar-denominated bonds and the corresponding U.S. Treasury security. Figures are weighted averages for sovereign and other public issuers.2Argentina, Brazil, Mexico, and Venezuela.3Czech Republic, Hungary, and Turkey.4Korea, Malaysia, and Thailand.

These spreads are linked to the secondary market prices for Brady bonds and other bank claims on developing countries. The market arbitrages away differences between the yields to maturities on new issue and Brady bonds, but the arbitrage possibilities are often complicated by the different characteristics of these two types of instruments. For example, Brady bonds are typically collateralized, while new issues usually carry no collateral and instead may feature different types of credit enhancements, such as put options. Brady bonds also have much longer maturities than the new issues. Some market participants use Mexico and other countries that have completed Brady operations as a benchmark for setting the price of claims on countries (such as Peru and Russia) still to complete bank debt restructurings.

Sovereign bond issues in each country serve as a benchmark for the bond issues by other borrowers in that country. Mexico pursued a deliberate strategy of issuing sovereign bonds at different points along the yield curve to facilitate the pricing of bonds issued by Mexican public and private enterprises.19 In 1990, private bond issues paid spreads of 400 to 500 basis points above sovereigns, but by 1993, this margin had come down to 100 to 150 basis points.

Annex V Role of Capital Markets in Financing Chinese Enterprises

The development of securities markets in China has become one of the most visible aspects of that country’s 15-year reform program. Drawn by the lure of double-digit growth rates in the most populous country in the world, foreign firms and investors rushed to establish a presence in China. Foreign direct investment commitments soared to $100 million in 1993, while portfolio investors responded eagerly to opportunities to invest in mainland Chinese firms. Equity placements on the Stock Exchange of Hong Kong (SEHK) amounted to $1.1 billion in 1993. International bond issuance increased to $2.9 billion, compared with $1.3 billion in 1992 and only $0.4 billion between 1989 and 1991. Demand for securities issued by Chinese enterprises was even stronger within China as individual Chinese, once given opportunities to invest in assets whose returns were market determined, proved eager to do so.

This annex examines the development of Chinese securities markets from their beginnings in the early 1980s. The Chinese approach to reform in the financial sector, as in other sectors, has been one of cautious experimentation, gradual relaxation of control, and decentralization of responsibility. Recently, however, it became necessary to retrench temporarily as the pace and direction of recent market activity exceeded the authorities’ intentions.1

Development of Securities Markets in China

Although government bonds had been issued between 1952 and 1958, the systematic development of a bond market in China began in the early 1980s, when the Central Government recognized that an alternative to central bank financing would be needed to minimize the inflationary consequences of budget deficits. In 1981, the Government issued Y 4.9 billion in ten-year Treasury bonds (Table 21). These were allocated to state-owned enterprises, collectives, and local governments, which then passed them on to individuals through compulsory subscriptions.

Table 21

Issues of Securities in China

(In billions of Chinese yuan)

article image
Sources: Data provided by the Chinese authorities; Almanac of China’s Finance and Banking (Beijing: China Financial Publishing House, 1992); Chen (1991); and IMF staff estimates.

Data for earliest entry include all previous issues.

In 1985, issuing privileges were extended to financial institutions as a means of supplementing the funds provided by the Ministry of Finance. These bonds were issued in shorter maturities—between one and five years—and with higher interest rates than those on Treasury bonds. Subscription to these bonds was usually voluntary, although local governments occasionally resorted to forced allocation methods similar to those employed to distribute Treasury bonds. From the mid-1980s, the range of bonds broadened, as different types of bonds were introduced—each issued by a different group of institutions and often targeted to different investors—to finance particular kinds of activities. In 1988, the first bonds issued directly by the Ministry of Finance and placed with financial institutions appeared.

The issuance of corporate bonds progressed much more slowly than that of government bonds and financial bonds. As a result of the financial reforms introduced in the early 1980s, it became more difficult for enterprises, particularly those outside the state sector, to obtain bank financing for investment or for working capital. Approval for enterprise bond issues was first made in 1982 on a very limited basis. Initially, the enterprise bond market was only loosely regulated, a situation that allowed enterprises to promise interest rates much higher than those paid on bank deposits. This led to a drain of resources away from the banks and caused the People’s Bank of China (PBOC) to respond with regulations in 1986; these required that enterprise bonds be approved by the PBOC, subject to an overall quota and a 15 percent ceiling on their interest rates. Enterprise bonds had relatively short maturities of two or three years and were usually issued to the enterprise’s employees—often by forced subscription—and were nontransferable.

Through the mid-1980s, as inflationary pressures built up, resistance to the forced allocation of Treasury bonds with negative real interest rates grew, particularly since some nongovernment issuers, especially enterprises, greatly increased the interest rates they paid on their bonds. Consequently, the Ministry of Finance was induced gradually to reduce the maturity of Treasury bonds, first to five years in 1985 and then to three years in 1988, and to increase the interest rates (Table A15).2 Even in 1985 however, these rates still compared unfavorably to the rates paid on bank deposits.

As a result of the widening of bond issuing privileges, the amount of bonds issued each year rose substantially after 1985 (see Table 21). However, Treasury bonds became a smaller proportion of total issues as they were supplanted by bonds issued by financial institutions and enterprises, in particular certificates of deposit. That being said, enterprise bonds still amounted to only 15 percent of total securities issued up to the end of 1990 (or 11 percent of the outstanding stock at year-end).

Chinese equity markets also have their roots in the reforms initiated in 1979. As part of the Government’s policy of financial decentralization, enterprises were given limited permission to issue shares. These securities had several unusual characteristics. For one thing, they frequently offered a guaranteed minimum annual rate of return. In principle, this return would consist of some fixed rate plus a share of the firm’s profits. However, in practice, most enterprises simply promised the maximum allowable rate of return and in some cases were able to provide much greater returns. During 1985–86, returns of 20–40 percent a year (four to six times the annual rates of interest on bank deposits or government bonds) were common, and in 1988 some enterprise equities earned 50–100 percent. A second characteristic of shares in the earlier years was that they carried a maturity date, usually between one and five years, and their holders frequently had the option of early redemption. Finally, enterprise shares generally carried no ownership rights.3 Consequently, these securities were more like bonds or preference shares with an embedded redemption option than straight equities. A further complicating factor was that in many cases shares, like bonds, gave investors special advantages such as priority in obtaining enterprise-provided housing or health care.

Central control over the equity market was first exerted in 1984 when the Central Government approved share issues on an experimental basis in five cities. In the same year, the municipal government in Shanghai—which was not one of the approved cities—permitted state-owned enterprises to issue shares. This made it easier for state-owned enterprises in these locations to issue equity, and as credit was tightened in late 1984 and in 1985, this became a more common means of raising capital.

Most shares, even those issued publicly rather than placed privately, were purchased by state and collective enterprises. Hence, effective state ownership was maintained.

Secondary markets for debt and equity developed slowly. In the early 1980s, all securities were officially nontransferable. After 1985, Treasury bonds could be discounted at the PBOC or used as collateral for loans. However, since these bonds were unattractive as investment vehicles, a thriving illegal market soon developed in which individuals who had been forced to purchase bonds sought to sell them. Since bonds held by individuals were in bearer form, speculators could purchase them at steep discounts, hold them to maturity, and then redeem them for substantial profits. Similarly, early enterprise bonds were mostly nontransferable; however, some issues included a tranche that was made available to individuals on a voluntary basis, and these were transferable.

Officially sanctioned securities trading did not begin until August 1986, when a secondary market was established on an experimental basis in Shenyang. This was an OTC market in which two corporate bonds were available for trading at prices determined on a daily basis by the authorities. However, the market was quite illiquid because of the lack of supply. Since enterprise bonds, like equities, offered fringe benefits to investors, there was little incentive to sell them once they were acquired.

An official market in government debt securities was established on a trial basis in seven cities in April 1988. This experiment proved so successful that it was extended to more than 60 cities by June 1988 and the range of securities eligible for trading was expanded to include key construction bonds, enterprise and financial bonds, shares, commercial paper, and certificates of deposit. Turnover of Treasury bonds in 1988 was Y 2.4 billion (Table 22). The legitimization of trading led to the creation of a number of securities companies and a renewed drive to recognize securities markets throughout the country. However, the events of June 1989 resulted in a temporary freeze on securities market development. It was not until late in 1990 that the central authorities resumed their experimentation with securities, and it was only in March 1991 that securities trading was legalized throughout the country.

Table 22

Transactions in Debt Securities in China

(In billions of Chinese yuan)

article image
Source: Almanac of China’s Finance and Banking (Beijing: China Financial Publishing House, 1992).

Although the authorities had allowed secondary markets for government securities to be established, they had not allowed a truly national market to develop. While individuals were free to trade eligible bonds in any one of these cities, they were not permitted to transfer bonds from one trading center to another until October 1990.4 As a result of different market conditions around the country, large differences in prices for the same issues developed. A comparison of the market prices of the five-year Treasury bonds issued in June 1986 in the ten most active trading centers during 1990 reveals that the yields to maturity differed greatly between these trading centers. The difference between the highest and lowest yields ranged from a low of 221 basis points in April to a high of 716 in December.

Official secondary markets for equity also took form in 1986. In September 1986, the Shanghai branch of the Industrial and Commercial Bank of China (ICBC) opened a securities trading counter and offered to buy and sell enterprise shares at transaction prices set each day by the municipal government. However, there were practically no sellers. In 1987, when six equities with a total value of Y 60 million were available, turnover reached only Y 6 million.5 By the end of 1988, there were 33 trading centers in Shanghai, trading seven enterprise shares and eight bonds. Turnover in that year was Y 535 million, although equity trading was only Y 11 million. Even in 1989, when total turnover reached Y 803 million, equity trading was still only a small share. In late 1990, the Central Government decided that it would extend the stock market experiment throughout the country, not by giving approval for a large number of exchanges to open, but by concentrating activity in a small number of centers and by providing investors around the country with access to that market. The first recognized stock market was in Shanghai, where the exchange opened officially on December 19, 1990 with seven equities, five state bonds, eight enterprise bonds, and nine financial bonds. The total value of equity issued by these seven enterprises amounted to Y 235.9 million, of which the state share was Y 156.5 million, institutions held Y 16.3 million, and individuals held Y 63.1 million. Only the last amount was available for trading on the exchange. Combined turnover in bonds and equity in Shanghai reached Y 2.7 billion in 1990.

The Shenzhen Securities Market began operations with the initial public equity offering in April 1987 of the Shenzhen Development Bank (SDB). Following official approval for securities trading in 1988, an OTC market for SDB shares emerged in Shenzhen. Turnover of both bonds and equity was only Y 4 million in 1988. By 1989, four more enterprises had issued shares and two more brokerage firms had opened up. Interest in securities waned as a result of an increase in deposit interest rates and a 20 percent tax on securities trading imposed by the Shenzhen government. However, later in that year, the SDB reported unusually strong profits and paid a dividend of Y 10 per share, which raised great public interest in equities. Turnover in equities in Shenzhen in 1989 was Y 23 million, and in 1990, with the addition of more issues, turnover jumped to Y 1.8 billion. The official OTC market could not handle the sudden increase in volume, and a large curb market emerged. In response to this, the municipal government opened a centralized trading center, which began trading in equities on December 1, 1990. The Shenzhen Stock Exchange did not receive official recognition until April 14, 1991 and opened officially on July 3, 1991 with five enterprises listed. These five enterprises had issued Y 270 million of equity, of which only Y 91.7 million was held by individuals and therefore available for trading.

At the end of 1991, there were 9 stocks and 28 bonds listed on the Shanghai exchange, with a total capitalization of Y 2.9 billion. The Shenzhen exchange had 6 stocks listed, with a capitalized value of Y 7.4 billion. Despite liquidity problems and high volatility, the market continued to grow in terms of issues and capitalization through 1993. At the end of the year, there were 183 issues of both A and B shares listed on the two exchanges with a combined value of Y 347.4 billion ($39.9 billion).6 New issues of shares reached Y 644 million in 1991, Y 7.15 billion in 1992, and Y 29.5 billion in 1993. Turnover in both classes of shares was Y 331.9 billion ($38.5 billion) in 1993, most of which was in A shares, despite a decline in turnover during the middle two quarters. During most of the year liquidity in the B share market was extremely low. At the end of February 1994, there were 234 companies listed—including 45 B shares—with a combined capitalization of Y 416.7 billion.

Steps toward the creation of a national securities trading network came with the introduction of the Securities Trading Automated Quotation System (STAQS).7 The STAQS came on line on December 5, 1990, and provided a satellite computer link for 17 securities companies in six cities. STAQS provides on-screen pricing information and a centralized clearing and settlement system and is unique among securities markets in China in that it adopted the market-maker trading structure. At the end of 1991, the five circulating Treasury bonds—those issued between 1986 and 1990—were listed on STAQS. STAQS helped to narrow but did not eliminate the price differences between the regions. Also in 1990, Treasury bonds were listed on the Shanghai Securities Exchange (SSE).8 In October 1993, turnover in the five-year 1992 bond on STAQS was Y 133 million, while trading in Wuhan, the most important regional trading center, amounted to Y 22.4 million, and turnover in SSE was Y 19.9 million.

Recently, a system similar to the STAQS has been established under the auspices of the PBOC and with the support of the specialized banks and the three national securities companies. The National Electronic Trading System (NETS) was officially opened in June 1993 to provide a nationwide electronic trading system for stocks and bonds. Trading is based on an order-driven, book-entry system in which orders are relayed through a network of more than 100 satellite-linked ground stations. At present no bonds are traded, only institutional shares, in which the market is relatively illiquid.

The development of a secondary bond market and particularly the creation of the STAQS system, which allowed for the issuance of bonds on a paperless, book-entry basis, facilitated an increase in government bond issues. In 1991, Treasury bond issuance doubled to Y 19.9 billion, compared with Y 9.3 billion in 1990. The 1991 Treasury bond issue is important because it represented the first attempt to distribute government debt to individuals through voluntary purchases. A syndicate of 70 financial institutions was appointed to market Y 2.5 billion of the planned Y 10 billion amount to be issued to households.9 The bond issue was successful and the syndication method was repeated in 1992 when a total of Y 40 billion was issued.

The government bond market suffered a setback in 1993 as a result of the availability of more attractive returns in other markets—particularly the equity market. In March, the Ministry of Finance decided to issue, by voluntary subscription through a syndicate of 91 financial institutions, Y 30 billion in three- and five-year Treasury bonds, at interest rates of 10 percent and 12.52 percent, respectively. However, by the end of the subscription period in April, only about Y 4 billion of the issue had been taken up, and by the end of May only Y 8.3 billion had been sold. Investors were increasingly drawn to the equity markets (in which capital gains were much higher than bond yields), to financial and enterprise bonds issued by local governments and nonbank financial institutions, and to deposits, all of which paid much higher rates of interest. As a result, the Ministry of Finance raised the interest rates and offered holders of the five-year bond the option to cash in after only three years (Table A15). Still, demand did not increase sufficiently, and the Government was forced to return to forced subscriptions. Regional and local governments were assigned quotas of bond subscriptions and were told that new equity issues would not be approved until they met their quota. Employees of state enterprises and government offices were then forced to purchase bonds (Y 5 billion) as were employees in private enterprises (Y 3 billion); institutional holders of maturing Treasury bonds were forced to roll them over into the new issue (Y 16 billion).

In 1993, the PBOC issued short-term financing bills for the first time. Another innovation in the domestic bond market in 1993 was the issue of bonds denominated in foreign currency. In June, the National Metallurgical Import and Export Company issued $40 million in one- and two-year bonds. In July, the China Investment Bank issued $50 million in one-year bonds. In the same month, the Ministry of Finance announced the adoption of a primary dealer system, to which 19 institutions-including only one bank—had been appointed. These dealers would be required to make a market in Treasury bonds.

For 1994, the Government has announced plans to sell about Y 100 billion in bonds with maturities between six months and ten years. Sales of long-term bonds—eight and ten years—are for the first time to be conducted through auctions. About Y 15 billion in short-term bonds—with maturities of less than one year—were sold through a syndicate of 35 financial institutions led by the ICBC by March 1994. Two-year bonds carry a 13 percent coupon, while the three-year bonds carry a rate of 14 percent—the three-year deposit rate plus 1 percent—which exceeds the forecast 10 percent inflation rate. Despite the fact that these yields are below the actual urban inflation rate, sales of Treasury bonds have been brisk. The target volume of two-year bond sales was actually met ahead of schedule.

Futures exchanges were first officially recognized in 1992 in Shenzhen. Since then, at least 30 recognized commodities futures exchanges have been established. Financial futures have also been introduced. In January 1993, the SSE listed its first futures contracts on government bonds, but the market was limited to the members of the exchange. In October participation was opened up to anyone, and turnover increased greatly to a daily average of Y 30 million.10 However, bond futures are also traded on the Beijing Commodity Exchange, where turnover is reportedly much higher.

Recent Developments in External Financing

China’s access to international bond and equity markets increased markedly in 1993, emulating the recent experience of many other developing countries. The supply of resources has shifted from commercial banks to the securities markets. Commercial banks’ lending to developing countries has been constrained in some countries by capital weakness arising from losses on loans to other groups of borrowers, slow growth in deposits, and the capital constraints imposed by the implementation of the Basle capital adequacy guidelines.11 As a consequence, banks have generally become more interested in loans that have lower capital requirements (such as short-term credits and officially guaranteed or structured lending) and in fee-generating, off-balance sheet business, such as underwriting the issuance of bonds. Moreover, some market participants noted that some banks had by the end of 1992 reached their internal limits on credit exposure to Chinese borrowers.

This combination of factors has been reflected in increased interest rates and shorter maturities on bank loans to China. Interest rate spreads appear to have increased by some 25–50 basis points during 1991–93 to around a yield of LIBOR plus 95 basis points. At the same time, average maturities in 1993 fell to 5 years, from 5.8 years in 1992 and over 10 years in 1991.

Bond Financing

China first issued a foreign bond in 1982 and was quite active up to 1989. In 1987 and 1988, various International Trust and Investment Corporations (ITICs) and banks raised $2.3 billion, with placements denominated in deutsche mark, yen, and dollars. The terms were in fact better than those achieved recently by Chinese issuers: interest rates were near LIBOR and maturities were rarely below seven years. Only one bond was issued in 1989, which raised $163 million. No bonds were sold in 1990, and in 1991 China resumed borrowing with two issues worth a combined $273 million. In 1992, activity picked up, with $1.3 billion in new bond issues (Table A16). At the end of 1992, international bond placements stood at about $6 billion and constituted approximately 10 percent of China’s total external medium- and long-term debt, a ratio that has remained relatively constant over the recent past.12 However, in 1993 China stepped up its external financing from the international bond markets, placing close to $3.0 billion.

Prior to 1993, China’s borrowing from international bond markets was undertaken exclusively by the so-called ten windows, which consist of the ITICs and certain financial institutions. The ten windows’ monopoly of the issuance of bonds in the international markets has been led by the China International Trust and Investment Corporation (OTIC) (25 percent of total Chinese bond issuance in 1992–93) and the Bank of China (BOC) (17 percent).13 The ten windows act as intermediaries, borrowing abroad with the advantage of an (implicit) government guarantee and then on-lending to domestic borrowers. Authorization for even these public sector institutions to borrow has been stringently regulated by the State Administration of Exchange Control (SAEC)—China’s debt management supervisor.

In order to prevent lenders in the market becoming overexposed to the same borrowers—and therefore increasingly reluctant to lend—the ten windows have attempted to diversify their investor base since China’s re-entry into the market in 1992. Although most borrowing has previously been concentrated in the Japanese domestic market—73 percent in 1992 and 43 percent in 1993—reflecting the familiarity of those investors with the Chinese market, the investor base has been diversified through re-entry into several major international bond markets.

The ten windows system has been recognized by the authorities as being limited, since the repetitive use of borrowers may eventually reach the limit of investor demand. Moreover, the balance sheets of the ten windows, in many cases, are in poorer condition than those of some of the enterprises seeking to access the international capital markets, and the ten windows’ on-lending to domestic enterprises has reportedly been inefficient. In addition, as the number of bond issues multiplies, the case-by-case approval by the SAEC has also become cumbersome.

Meanwhile, nonstate enterprises have reportedly placed unauthorized issues privately with regional investors, especially in Hong Kong. The volumes are understandably hard to calculate, but the terms are reportedly much less favorable than those for the ten windows. For example, Hong Kong-based Guangdong Enterprises—which reportedly issued the first Chinese bond in the U.S. market not to carry a government guarantee—paid a spread of 300 basis points on its Rule 144a ADR in December 1993.

From 1993, the authorities planned to experiment with a more pragmatic approach, including the necessary first steps to giving enterprises direct access to the international bond markets. Following the lead of many other developing countries, especially in Latin America, the Chinese authorities intend increasingly to allow direct access by the enterprises themselves—rather than through the ten intermediaries—and will use sovereign issues to establish benchmarks against which these enterprises’ issues can be priced. In this regard, the Government entered the market directly in 1993 and early 1994 after a six-year absence and set three benchmark issues (Table A16).14 Under the new approach, borrowing entities will be screened by an internal rating agency—ostensibly to provide potential lenders with better information—and then be provided a borrowing quota. To test the new approach, in 1993 OTIC was allowed to borrow up to a fixed limit without requiring case-by-case approval from the SAEC. To date, however, the SAEC has not opened up access to nonpublic enterprises, and at this initial stage access to the international markets continues to remain under the control of the authorities.

The Chinese authorities expect to place more emphasis on bond financing relative to bank financing, in view of the lower interest rates on bonds and the availability of longer maturities, which they believe to be more appropriate for their prospective infrastructure development.15 The spread over government instruments of comparable maturity for most Chinese issues has been consistently tight, at about 100 basis points. Maturities have averaged 5.7 years and have exceeded the average maturity on uninsured bank credits of five years. Moreover, there is potential for lengthening maturities. For the group of developing country borrowers, average maturities in the bond market for (unenhanced) bonds—7.1 years—exceed those in the credit markets for uninsured credits—4.9 years; furthermore, the range is wider in the bond market.16 The authorities place a special emphasis on the U.S. market, as they view it as the deepest market and most likely to produce the longest maturities, a fact already established by the recent elongation of maturities in the dollar market, to seven years in November 1992 (People’s Construction Bank of China (PCBC)) and to ten years in July 1993 (CITIC).

Equity Financing

The stock markets were transformed in February 1992 with the issues of class B shares in Shanghai and Shenzhen.17 These shares are reserved for foreign investors; they are denominated in U.S. dollars in Shanghai and in Hong Kong dollars in Shenzhen. In all other respects, they are identical to the class A shares reserved for mainland Chinese investors. The listing of B shares attracted the attention of international investors. The opportunity to invest in Chinese companies and thereby to benefit from the opening up of the Chinese economy proved to be highly attractive. By the end of 1992, 18 issues of B shares had been listed on the two exchanges, compared with 52 issues of A shares. In Shanghai, of the total capitalization (including bonds of Y 55.8 billion at the end of 1992), Y 3.1 billion was in A shares and Y 800 million in B shares. In Shenzhen, the corresponding figures were Y 48.3 billion, Y 2.6 billion, and Y 416 million, respectively. Enthusiasm for investment in Chinese equities was bolstered by the remarkable returns: an index of all Chinese equities—A and B shares combined-gained over 200 percent in the final quarter of 1992 alone, far outstripping returns in all other emerging markets.18

By early 1993 international investors apparently began to reconsider their headlong rush into B share investment. There were some reports that these investors were concerned about the adequacy of information disclosure and about the uses to which the funds that had been raised were being put. In addition, it was well known that the Chinese authorities were preparing to list nine enterprises on foreign stock markets where information disclosure requirements were more demanding. By the end of the first quarter of 1993, liquidity in B shares, never very high to begin with, declined, and with it average prices also fell (Chart 10).

Chart 10.
Chart 10.

Selected Stock Exchange Indices in China and Hong Kong, 1993–February 1994

(January 2, 1993 = 100)

Source: Bloomberg Financial Markets.

The initial euphoria, occasioned by what was for many people the first opportunity to invest in equity since 1949, and the remarkable capital gains recorded in 1992 resulted in a surge of investment in late 1992 and early 1993. Of particular concern to the authorities was the fact that funds were being diverted from the banking sector—often by the banks themselves—into securities and real estate investment through bank-affiliated trust and investment companies and securities companies. The drainage of funds became so serious that shortages of working capital in the state-enterprise sector became widespread, and funds were unavailable to pay farmers for their crops. The authorities responded in early July with a 16-point austerity program in which banks were ordered to cease their involvement with securities markets and securities companies and to call in loans made for unauthorized investments in real estate and securities.

After a modest increase in the first half of 1993, the overall trend in the A share markets was downward in the second half. This more than offset the increase in B share prices after the authorities stepped in to strengthen regulation in May.19 As a result of these conflicting effects, at the end of 1993, the composite market index showed only a 7 percent gain over 1992, the worst performance among the emerging markets being followed by the IFC—a marked contrast to 1992. Prices of both A and B shares fell in the first quarter of 1994, resulting in a 20 percent decline in the market index.

Reflecting the problems affecting the B share market and the domestic credit market, many enterprises reverted to unauthorized listings (“backdoor listings”) on the SEHK. Backdoor listings are takeovers by Chinese enterprises of publicly quoted Hong Kong companies. The primary incentives for such takeovers were that a Hong Kong listing enabled the enterprise to raise financing through new rights issues and that a Chinese enterprise could halve its effective tax rate by setting up a joint venture with the overseas-registered entity that it controls.20 Initially, most backdoor listings were undertaken by well-established and powerful mainland enterprises, such as China Resources, China Travel, China Overseas Land and Investment, and OTIC, which are referred to as “red chips.” Market participants have estimated that Chinese companies injected HK$21 billion into Hong Kong during 1991–93 through 29 listings (with a current market capitalization of HK$100 billion or $13.2 billion). The largest are OTIC Pacific (HK$15.4 billion) and Guangdong Investment (HK$1.8 billion), a property developer.

The Hong Kong and Chinese authorities did not initially discourage these practices. However, many other enterprises that followed suit had more muddled accounts without clear title to mainland assets and poor standards of investor disclosure. In many cases, it was not clear if the resources raised from rights issues were eventually used for their intended purpose. There was also concern that enterprise assets were being transferred to the Hong Kong companies to protect the assets from a weakening of the yuan. Moreover, the extent of questionable takeovers—many of which were based on inflated real estate property in China—started to tarnish the reputation of the SEHK and led to unauthorized transfers of assets out of China.

On May 11, 1993, the Hong Kong authorities prohibited unlisted companies from taking control of listed firms in order to obtain an automatic listing. In August, newly listed firms were prohibited from making rights issues for 12 months after obtaining their listing and, in September, the minimum market capitalization for new issuers was increased from HK$50 million to HK$100 million, of which one half has to be in public hands.

The tightening up on backdoor listings in mid-1993 coincided with the listing on the SEHK of shares issued to foreign investors by mainland Chinese enterprises—H shares. On June 19, 1993, representatives of the Hong Kong and Chinese regulatory authorities and stock exchanges signed a Memorandum of Regulatory Cooperation, which paved the way for listings by nine selected Chinese companies after they met its standards for primary listings. The memorandum requires (i) the full, accurate, and immediate disclosure of information relevant to investors; (ii) action to be taken against insider trading; and (iii) the supervision of trading and settlement. Of key concern were a higher standard of disclosure requirements, which were to be cross-checked by international auditors for compliance with international accounting standards, and the possibility of recourse to arbitration in Hong Kong, based on Hong Kong legislation, both of which were considered to be the main misgivings of foreign investors concerning the B share market. The initial listings reportedly required hundreds of thousands of hours of work by international accounting firms to construct acceptable financial accounts and to separate the noneconomic activities (schools, hospitals, and so on). Obtaining adequate investor protection also required addendums to the listing firm’s rules of incorporation in order to redress inadequacies in China’s company laws.

Between July 1993 and May 1994, nine Chinese firms issued H shares in Hong Kong exchange, raising $1.5 billion (Table 23). Reflecting the gradual and experimental role of other reforms, the firms chosen for initial H share listings represented some of the highest-quality Chinese enterprises, mostly companies in heavy industry with good track records, comparatively good management, and high growth potentials.21

Table 23

Chinese H Shares

article image
Sources: Euroweek; International Financing Review; and IMF staff estimates.

Including international tranches.

Daily trading volume as percent of market capitalization.

Standard deviation of daily price changes, at daily rate.

The Hong Kong and Chinese authorities recently announced the next group of enterprises (22 in total) for international listings, with 4 envisaged to have primary listings on the New York Stock Exchange (NYSE) and the other 18 on the SEHK. Though the first nine listings had an even geographic distribution, the subsequent H share listings appear to be more explicitly based on economic criteria, favoring especially infrastructure development—for example, the power generating sector. Once again, this second set of enterprises contains some of the highest quality firms in China. The attractions of the NYSE over the SEHK for Chinese companies are reportedly the much higher price/ earnings ratios that Chinese companies have obtained on the former—which permits them to raise funds more cheaply—as well as intangibles such as the scarcity value and prestige of a NYSE listing. The highest quality enterprises are expected to list on the NYSE, specifically two power supply companies and two regional airlines.

Major constraints on the number of firms that can be listed include the legal framework—a national securities law has yet to be passed—and the speed of adoption of appropriate accounting and disclosure standards. The latter problem may be alleviated by China’s new accounting standards, which are reportedly not much different from international standards. As the enterprises adopt these new standards, it will be easier for them to meet the listing requirements of the SEHK.

Though investor interest in Chinese equity has been heightened in the past two years, it has focused on indirect or diversified channels. Most investment in China is from retail investors (such as country funds), with institutional investors having shown the greatest enthusiasm for the NYSE listing of the Shanghai Petrochemical ADR. Country funds have three main channels for investing in China—H shares, B shares, and companies controlled by mainland Chinese corporations (backdoor listings and red chips) or companies closely related with Chinese interests. There are about 40 country funds that invest in China—all of which were established in the past two years—with a total of some $1.5–2.0 billion of assets under management. The B share listings have a market capitalization in the order of $5 billion, with average daily turnover varying from the high of $13 million to very low levels. Only about one fifth of fund managers invest more than 40 percent of their assets in B shares, whereas some three fifths invest over 40 percent of their assets in Hong Kong China plays. The preference for investing in China through companies listed in Hong Kong reportedly reflects the quality of the company managers, a familiar accounting system, and the perception that the local intermediaries have a comparative advantage in monitoring the quality of the investments.

Hong Kong’s role as conduit for investing in China arises from the fact that most foreign direct investment in China and most external bank credits are advanced by Hong Kong enterprises—reportedly in search of a low cost manufacturing base—and banks based in Hong Kong. Foreign direct investment to China almost tripled in 1992 to $12.0 billion and for the first time exceeded loan financing; for 1993 it is estimated to have increased further to an estimated $26 billion in disbursements and over $100 billion in commitments. In 1992, foreign direct investment from Hong Kong amounted to $7.7 billion (64 percent of the total).22

The Chinese Approach to Securities Market Development

Despite appearances to the contrary since the beginning of 1993, the development of securities markets has been a relatively gradual one in China. In the early stages, the authorities reacted to developments in the markets rather than anticipating or leading them. For example, official approval of the issue of enterprise securities was given only after a large number of enterprises had done so on their own initiative; secondary markets for securities were recognized only after unofficial curb markets had been operating for some years; and securities were traded on the Shanghai and Shenzhen exchanges long before they were officially opened in 1990 and 1991. Since 1990 the authorities have been more active in guiding market developments. New initiatives include official recognition of secondary markets for bonds and equity and efforts to create a national market. The Central Government has consistently attempted to maintain strict control over issuance by means of quotas for new share and bond issues by enterprises.

Banks’ Involvement in Securities Markets

Overall, the securities markets remain relatively small as sources of funds. The data up to 1991 show that the substitution of securities issues for bank loans has been very modest. The ratio of securities issues by enterprises (enterprise bonds plus stock issues) to bank loans to enterprises rose from 0.4 percent in 1987 to 1.5 percent in 1991. The total stock of securities outstanding at the end of 1992—Y 330 billion—represented only 11 percent of all financial assets, although that proportion is increasing at a rate of 1 or 2 percentage points a year.

Until recently, banks were intimately involved in the securities markets. Many of the securities firms were bank subsidiaries or affiliates, and the markets were regulated by the central bank. In early 1993, however, when it became apparent that this mixing of banking and securities activities was resulting in the diversion of credit away from the productive sectors of the economy, the authorities introduced regulations providing for a complete separation of the industries. Banks had to close or sell their securities subsidiaries and call in “speculative” loans, and they were forbidden to have any further involvement with securities markets. As a result there is, for example, no margin lending, and securities firms and exchanges are not permitted to seek insurance from banks in the form of lines of credit. This of course has had an effect on liquidity in the secondary market. Individuals must put cash up front for securities they purchase and must retain minimum balances in accounts with securities firms. The firms themselves must keep relatively large balances in their accounts with the exchanges in order to finance their proprietary trading activities and to protect themselves against settlement failures.

The Approach to Equity Market Development

In China, the selection of issuing firms thus far has been determined on an administrative basis. Rather than simply having to meet the exchange’s listing requirements and some national regulatory standards, Chinese firms that want to list must be transformed into joint-stock companies have their land and other assets valued by the State Assets Bureau and State Land Administration and have their financial statements audited and prepared according to international standards. They then must apply to the local government for approval to issue shares, and if that is successful, they apply to the China Securities Regulatory Commission (CSRC), which re-examines the application (in part to ensure that the exchange’s listing requirements have been satisfied). One of the conditions of eligibility is that the company has earned positive profits for the three preceding years. Applications by enterprises owned by the Central Government for bond issues are examined by the PBOC and the State Planning Commission, whereas applications from private enterprises or enterprises owned by local governments are examined by the relevant local government.

The Central Government has a quota of A share listings for the country as a whole (this may be subdivided into quotas for different exchanges), which allows it to control market development. Compared with an estimated Y 10.9 billion in new issues in 1992, the quota in 1993 was Y 5 billion, and that for 1994 is Y 5.5 billion. Subject to this limit, approval will be given for one or two listings from each province or municipality on each exchange. The recent weakness in the markets has delayed scheduled issues, with the result that some issues that have been approved for 1994 may not be marketed until 1995. There is no quota for B shares, but the approval process is a little more complicated (at the very least enterprises have to prove a need for foreign exchange). There are also other mechanisms by which the center can control these issues—for example, in 1993 B share issues from provinces that had not taken up and distributed their allotment of Treasury bonds were prohibited.

The marketing of approved issues differs from practices followed elsewhere. A shares are underwritten by a securities firm—usually a domestic firm, but often with a foreign firm as an advisor—which sells share-order forms in which the holder applies to purchase a fixed number of shares. The price is generally not known at this stage. This modified book-building system allows the underwriter to gauge market demand for the issue and to better price the shares. In fact, public offers have generally been greatly oversubscribed, so the successful applications have been selected by lottery, and the underwriter bears minimal risk. Although the Government no longer sets the issue price—as the PBOC had done in the past—it apparently still has influence over the determination of the issue price. By contrast, B shares are generally marketed by private placement, most often in Hong Kong.

Because of their different investor base, A and B shares have very different liquidity characteristics. For example, B shares generally are more thinly traded. As a result, prices of A and B shares for the same company (which differ only in their investor base and currency risk, but not in dividend or shareholder rights, and so on) frequently have very different prices. For example, on March 16, 1994, B shares in Shanghai traded at a 62 percent discount to the price of the same companies’ A shares, while in Shenzhen the discount was 39 percent. For individual companies, the discounts reach as high as 84 percent, while only one company’s B shares traded at a premium. It is not uncommon for developing countries to discriminate between foreign and domestic investors in their stock markets, but China may be alone in having the shares reserved for foreign investors trade at a discount.

In addition to strong “investor sentiment” pressures, one possible explanation for the B shares discount is that the required return for Chinese investors may be lower than for foreign investors because they have fewer alternative investment opportunities.23 While the B share discounts for a selection of companies during March 1992–March 1993 did not respond to international interest rates and risk premiums, they do appear to be significantly correlated with the premiums on shares reserved for foreigners in Singapore and Thailand. As the discount on B shares falls, the premiums on foreign shares in Singapore and Thailand rise. This suggests that foreign investment in China may be motivated by similar factors as investment in these other two markets.

The segmentation of the equity market into A, B, and H shares is a distinguishing characteristic of securities market development in China. As mentioned earlier, although discrimination between domestic and foreign investors is not unusual—this occurs, for example, in Singapore and Thailand—the unique feature of the Chinese approach is that these classes of market differ in their information requirements. Thus, it was recognized early on that issuers of B shares would have to provide better information to the market than would issuers of A shares, and that H share issuers would need to disclose even more information. Thus, issuers of B shares have to prepare financial statements according to international accounting standards, which is not required of issuers of A shares. Moreover, issuers of H shares, have the added responsibility of meeting listing requirements on the SEHK, which has stricter information disclosure requirements than the two Chinese exchanges.

The informational asymmetries reflected the recognition that international investors would not be prepared to invest in companies in which they did not have sufficient information (which was felt to be the case for the A shares). The local regulations in Shanghai and Shenzhen, for example, did not require issuers of A shares to use international accounting standards. The rules for B share issues were more demanding, and the regulators accepted foreign advice that accounts prepared according to Chinese standards would not then have been acceptable to foreign investors. The standards for H shares were set by the SEHK and are generally recognized to be the strictest.

The experience during 1993 has demonstrated that investors generally prefer markets with better disclosure. Foreign investors, with a choice of purchasing B or H shares, have gravitated toward the latter. Initially, when only B shares were available, the international investment community showed considerable interest in these shares, and turnover, while initially low, appeared to be improving. However, in the spring of 1993 when enterprises that had issued B shares began reporting their 1992 results, some of these firms chose not to prepare their reports according to international accounting standards, thus violating B share regulations. Moreover, it soon became apparent that some of them had misused the funds they had raised by speculating in real estate or securities markets, or simply by on-lending the foreign exchange. By the end of June, B share liquidity had all but disappeared. The emergence of Chinese companies listing on the SEHK, beginning in July, provided a superior alternative to B shares, and international investors quickly turned their attention to that market. With one exception, H share issues have been very well received in the primary market, and the secondary market for these shares has been very liquid since their listing.

A second element of the Chinese authorities’ approach to stock market development is their decision to pick which firms will issue shares on the two regulated markets, subject to an overall quota and a concern for regional equity. The selection of H share issuers is not subject to a quota; however, the choice of companies is almost entirely made by the central authorities at the highest level. Rather than letting the market determine the issuers, subject to some minimum standards contained in securities and company law and in the exchange’s listing requirements, the authorities are determined to follow a strategy of “picking winners.” To a certain extent this approach was dictated by the absence of adequate disclosure, market regulation, and market-determined prices for competing assets. Since China did not have a market economy, a market-based approach to equity issues might not have been feasible. Perhaps more important in the authorities’ opinion, there was a need to maintain control over the liberalization process. This is consistent with the general approach to reforms in China in which measures are implemented on a trial basis at first.

Suffice to say that given the large structural changes under way and the nature of the remaining distortions, it would not be easy for an outsider to identify the most profitable or best-managed firms. Local authorities may in fact be in the best position to do so. Government intervention of this sort may resolve a marked information asymmetry between domestic firms and foreign investors.

This strategy could run into trouble if non-economic factors, such as a desire to spread listings geographically, were to receive too much attention in the selection process. This could result in firms being allowed to issue shares, despite the fact more profitable firms in the same industry located elsewhere could not.

Corporate Governance

Another motivation for the development of corporate securities markets is to create a system of corporate governance that does not rely on bureaucrats. A first step in this process is to remove the enterprises from direct government control. Thus, the authorities have encouraged enterprises to transform themselves into joint-stock companies with majority (initially usually 100 percent) state ownership, but with independent management.24 To date, more than 13,000 enterprises have taken on this status—9,500 in 1993 alone.

The second step is to allow enterprises to obtain their own financing. While the largest enterprises can still rely on fairly automatic bank credit, smaller ones would have to negotiate terms with the bank. Moreover, the intent has been that these loans should be repaid. In addition, subject to the approval of regulatory authorities, firms are allowed to issue debt and equity securities to the public. However, so far, equity issues have generally comprised less than 30 percent of the enterprises’ enlarged capital.

Having released enterprises from direct bureaucratic control, the Government has recognized that some new source of discipline for managers must be found. It hopes to obtain this discipline from a reformed banking sector and from the securities markets. In assessing the scope for securities markets to exercise discipline, several factors need to be taken into account.25

Consider first the contribution of the banks. Banks can impose financial discipline on firms primarily through the requirement to make regular payments on loans. If a firm does not obtain a regular cash flow from the investments financed by the borrowed funds, they may be unable to service the debt and be forced into bankruptcy. If bankruptcy is costly to managers, they will have an incentive to ensure timely service of their debts. Moreover, this obligation may mitigate any incentive managers may have to divert enterprise resources to their own personal uses.

Banks also have access to fairly detailed information on the financial condition of the firm—both as a condition for making or renewing a loan and from observing the firm’s cash flow through its current account, which is often required to be maintained at the lending bank. Of course, banks will only enforce this discipline if they themselves face a cost to bad lending decisions. At this point, the only domestic financial institutions that have lent to the state-owned enterprises are themselves state owned. Since their cost of funds and lending interest rates are beyond their control, and since banks face no apparent penalty from accumulating bad loans, they have limited incentive to enforce their lending contracts. For example, although a bankruptcy act was passed in 1988, by the end of 1993 only 20 petitions for bankruptcy had been brought to court. Bank managers acknowledge that they would rather carry a bad loan on their accounts—effectively capitalizing unpaid interest—than force bankruptcy and closure of the firm.

Equity markets provide an alternative—and complementary—source of discipline through two possible avenues. In the first place, since shareholders own a claim to the firm’s residual profits, they have an incentive to monitor the management of the firm and to ensure that the most profitable investments are undertaken in order to maximize the value of their shares. The second source of discipline derives from the marketability of the shares. At the very least, if shareholders become dissatisfied with the performance of the firm they can sell their shares. If this results in a significant decline in the market price of the equity, then the firm may find it more difficult to find new investors. If the price falls by a large enough amount, a rival manager may attempt a takeover, believing that he is more capable of managing the firm efficiently. Thus, managers for whom the loss of their position would be costly have an incentive to avoid this by managing the firm as efficiently as possible.

Equity markets in China face several constraints on their disciplining role. Investors need reliable information about the firms in which they are investing if they are to be able to closely monitor the firms’ activities. Moreover, since private institutional investors have not yet emerged as significant shareholders, the shares the public holds are widely dispersed; it is not clear therefore that any one individual has a large enough stake to compensate himself for the monitoring costs that would be incurred before effecting any change in management. Rather, it is possible that any indication of poor performance would result in a sale of shares. However, takeover regulations are drafted in a way that makes it difficult to rely on this mechanism for disciplining poor managers. Investors must make a public announcement when their stake reaches 5 percent of a firm’s outstanding shares, and at every 2 percent increment thereafter. Moreover, the investor is barred from trading in that stock for two days after each announcement. Foreigners cannot mount a takeover through the secondary market because they are barred from purchasing any shares other than B and H shares, which generally amount to no more than 30 percent of the firm’s capital.

Under present conditions, therefore, the main source of discipline is likely to come from the actual listing process itself. Both the authorities and Chinese and foreign securities houses report that preparing audited financial reports in line with international accounting standards and restructuring the operations of state-owned enterprises are highly instructive to management.

Supervision and Regulation of Domestic Capital Markets

The regulation of securities markets was rearranged in early 1993, which resulted in greater centralization and rationalization. Prior to April 1993, the PBOC took the initiative in developing and regulating securities markets through its securities regulatory office in Beijing. In this they were supported by the Ministry of Finance and the State Council for the Reform of the Economic System (SCRES). Because of the speed with which securities market activities emerged in centers all across the country, these responsibilities were soon decentralized. The local municipal governments and PBOC branches had the most important direct supervisory roles. The PBOC licensed securities firms and markets and approved new listings, while the local authorities were responsible for regulating the markets. In most cases, central government approval for a particular development—secondary market trading and issuance of a new kind of security—came long after the local authorities had approved them.

As a consequence of the decentralized regulatory structure, various markets formulated their own rules and regulations and created parallel systems for trading, clearing, and settlement. This duplication of effort made clear the need for a consistent, central regulatory structure. This was created in October 1992 when the State Council decided to centralize regulation. A two-tier structure was established—the State Council Securities Policy Committee (SCSPC) and its executive arm, the CSRC—and became operational in April 1993. The SCSPC consists of representatives of 14 government ministries, including the Ministry of Finance and the SCRES, and is primarily responsible for drafting securities laws and regulations (or authorizing the CSRC to do so) and for formulating guidelines and rules governing securities market development. The CSRC is responsible primarily for implementing regulations and for supervising securities firms and markets. The exchanges continue to set their own listing requirements and operate as self-regulatory agencies.

The CSRC began operations in April 1993 by issuing the Interim Regulations on the Administration of the Issue and Trading of Shares, which replaced the provisional regulations promulgated by the various local authorities—particularly in Shanghai and Shenzhen—and formed the foundation for future national legislation, which is being drafted. These regulations prescribe issuing requirements—among others, the requirements that companies show at least three consecutive years of profits prior to issue and that intangible assets represent less than 20 percent of total assets—and acceptable practices in the primary and secondary markets. They specify the operating requirements for securities firms and explain how the firms will be supervised. They also include penalties for securities fraud and cover mergers and acquisitions.

The PBOC has retained its role of licensing securities firms, but the CSRC sets the conditions of eligibility and capital requirements and supervises the daily operations of these firms. Similarly, the Ministry of Finance licenses accounting firms and their professionals, but the CSRC sets the eligibility requirements and supervises their activity. The Ministry of Justice licenses lawyers, but the CSRC monitors their securities activities. The CSRC also monitors listed companies’ compliance with reporting and other requirements.

Issues in External Financing

The Chinese authorities have chosen to list firms in Hong Kong rather than access international markets through issues of ADRs/GDRs—as is common in many other developing countries (see Annex IV). Both these vehicles, direct listing and ADRs/ GDRs, raise an issue’s liquidity and concomitantly its price through the broadening of the investor base, the reduction in settlement time and risk, the introduction of trade and settlement according to international standards and more established currencies, and the avoidance of foreign investment restrictions. However, ADRs/GDRs differ from direct placements in international markets in that they provide extra liquidity to the domestic market and not just the individual issue.26 However, this route is made more difficult by the more rigorous accounting and disclosure standards required for a public listing of an ADR/GDR. Moreover, China has a close relationship with Hong Kong, which may increase the attraction of listing there rather than elsewhere.27

The pricing of Chinese equities highlights the market’s concerns with the B share market. During 1993, the IFC’s Asia investable index, the Hang Seng stock market index, and Standard Charter’s B share index increased by 98 percent, 119 percent, and 35 percent, respectively. An index of red chip firms outperformed the Hang Seng index until the introduction of H shares in July 1993, with the former rising by 40 percent, compared with a 30 percent increase in the Hang Seng index. The Hang Seng and the B share indexes were substantially more volatile than the IFC index during 1993. Moreover, liquidity is relatively equal among all the China plays except for the B share market, where liquidity has been lower.

Broadening the analysis to include H shares is complicated by the relatively small period for which they have been issued, as well as by the upheavals in the Hong Kong market during this period. The first H share was issued in July 1993, while the largest H share started trading only in November 1993. The Hang Seng soared to record heights during the last quarter of the year and turned sharply downward in early January 1994. Nevertheless, an H share index greatly outperformed the Hang Seng index over November 3–December 31, 1993, but fell slightly more than the Hang Seng over December 31, 1993–March 3, 1994.28

In the short run, China’s access to the international securities markets, especially by entities carrying government guarantees in the bond markets, will be facilitated by the absence of debt-servicing difficulties in recent years and the opening of the economy under the reform program—along with the concomitant economic development. In recognition of these positive aspects, the major credit rating agencies having granted China an investment grade status. In setting this rating, Standard and Poor’s cited China’s demonstrated record and long-term prospects for strong economic growth, moderate external debt burdens, and cautious policy approach to managing economic development. Reconfirming the positive outlook, Moody’s upgraded China from Baal to single A in September 1993.29 These sovereign ratings exceed those of most other developing countries, other than a few of the other dynamic Asian economies. As the rating of a country’s non-sovereign issues cannot exceed their respective sovereign rating, China’s high rating should facilitate access by qualified enterprises to the markets when this is authorized. It is interesting to note that industrial country corporations with similar ratings can borrow at yields some 150 basis points over government paper of 20–25 years maturity, which—based on an extrapolation—is comparable to China’s spreads on issues by the ten windows.

Annex VI Glossary of Financial Terms

Auction

Discriminatory:

Auction in which each successful bidder pays the price bid. Also known as a multiple-price auction.
Uniform-price:

Auction in which all successful bidders pay the same price, usually the price of the lowest successful bid. Sometimes called a Dutch auction.
Sealed-bid:

Auction in which all bids are submitted secretly, before some deadline, with no opportunity for revision.

Benchmark:

Security used as the basis for interest rate calculations and for pricing other securities. Also denotes the most heavily traded and liquid security of a particular class.
Broker:

Financial intermediary that solicits orders from buyers and sellers and then orchestrates trades, either by passing the identity of each party to the other or by making offsetting, simultaneous trades with each party. Brokers typically maintain computer screens with anonymous bid and offer quotes from dealers.
BTANs (Bons du Trésor à taux annuel):

French fixed-rate, two- or five-year Treasury notes.
BTFs (Bons du Trésor à taux fixe):

French fixed-rate, short-term discount Treasury bills.
Bunds (Bundesanleihen):

German long-term Federal Government bonds.
Cash market:

Market for sale of a security against immediate delivery, as opposed to the futures market.
Cherry-picking:

The practice of a firm selectively enforcing only those contracts with positive net present value to itself in the event of the bankruptcy of its counterparty.
Coupon:

Periodic interest payment on a bond. Some bonds have physical coupons that must be clipped and submitted to a bank. A coupon rate is the stated interest rate on the bond, equivalent to annual coupon payments as a percentage of the principal of the bond (par value).
Dealer:

A financial intermediary that buys and sells securities or other instruments, by setting bid and offer quotes. Unlike brokers, dealers take positions in the instruments.
Depository receipt:

A negotiable certificate that is issued by a U.S. bank and that is fully backed by shares that in turn, represent claims on the publich traded debt or equity securities of a company. These receipts trade freely on an exchange or an OTC market. American Depository Receipts (ADRs) and Global Depository Receipts (GDRs) are identical from an operational point of view and the terms are used interchangeably, depending on the marketing strategy.
Dragon bonds:

Bonds that are priced, launched, and syndicated in the Asia-Pacific region outside Japan and listed in at least two of the Hong Kong. Taiwan Province of China, or Singapore markets.
Duration:

A weighted average of the terms of all cash flows from a debt instrument. The present values of these cash flows are used as the weights, with the yield to maturity used to compute the present values. Duration also represents the elasticity of the value of a bond with respect to changes in its yield to maturity. Modified duration is duration divided by a factor of one plus the interest rate.
EMTN (Euro medium-term note):

Medium-term bonds issued on a tap basis on the Euromarkets.
Gilts (or gilt-edged securities):

Irish or U.K. Government medium- and long-term debt securities
Good bank/bad bank model:

Method of restoring a bank with a large amount of bad loans by placing bad loans in a separate institution (the “bad bank”) and adequately capitalizing the remaining bank (the “good bank”).
GOVPX:

Service in the United States that distributes real time price and quote information for all U.S. Treasury securities.
Margin:

Amount of cash that must be provided when borrowing to purchase a security. For example, U.S. regulations limit margin to 50 percent for equity purchases, so that a customer may only borrow half the value of a stock purchased. Margin also refers to the amount by which value of a security in a repurchase agreement exceeds cash lent.
Market-maker:

A dealer that posts ongoing quotes in a particular instrument.
Marking to market:

Expressing assets or liabilities at current market values.
MATIF (Marché à Terme International de France):

Financial futures exchange in Paris.
Netting by novation:

A bilateral contract between two counterparties under which a new obligation to pay or receive a given currency is automatically amalgamated with all previous obligations in the same currency, thereby creating a single legally binding net position that replaces the larger number of gross obligations.
OATs (Obligations assimilables du Trésor):

French fungible, long-term Treasury bonds.
On-the-run:

Term used in the United States for the most recently issued Treasury security of a particular maturity class, which is also the most liquid and heavily traded (see benchmark). On-the-run status begins with when-issued trading.
Par:

The principal of a bond.
Primary dealers:

Group of dealers in the United States with a formal, ongoing trading relationship with the Federal Reserve Bank of New York and with certain obligations in the primary and secondary market for Treasury securities. Term also applies to similar entities in other countries.
Primary market:

Market in which a security is first sold by issuer.
Price discovery:

A general term for the process by which financial markets attain an equilibrium price, especially in the primary market. Usually refers to the incorporation of information into the price.
Pure discount:

Zero-coupon (see below).
Repurchase agreement (repo):

An agreement to sell a security and then repurchase it at a particular time and price.
Secondary market:

Market in which a security is sold by one investor to another, as opposed to the primary market.
Settlement risk:

Risk that one party or another in a securities trade will fail to deliver, especially when the other party has already delivered.
Short squeeze:

Situation in which traders with short positions seeking to close their positions are required to pay an abnormally high price for the instrument because another trader has amassed a dominant long position in the instrument.
Strip:

A pure-discount security created by the decomposition of a bond into separate securities for each coupon payment and for the final principal payment. The term strip comes from the U.S. Treasury acronym for “separate trading of registered interest and principal.”
Syndicate:

Group of intermediaries that purchase prearranged shares of a security in the primary market and sell the security to other investors.
Tap sales:

Sales by a central bank of a new issue of government securities, usually on a gradual basis.
When-issued:

The market for a security before it is sold on the primary market.
Winners’ curse:

Losses incurred by successful bidders at an auction, due to those bidders’ having inaccurate, overoptimistic information on the value of the item auctioned.
Yield to maturity:

Interest rate that makes a bond’s present value equal to its market price. If the price of a bond is below par, the yield to maturity is greater than the bond’s coupon rate, and the bond is said to trade at a discount.
Zero-coupon:

A bond with no coupons, only a single principal payment at maturity.

Statistical Appendix

Table A1

United States: Estimated Ownership of Public Debt Securities by Private Investors1

(In percent of total privately held securities)2

article image
Sources: United States, Department of Treasury, Treasury Bulletin (March 1988, December 1993, and March 1994), Table OFS-2.

As of end-December.

Securities valued at par; some savings bonds are included at current redemption value.

Exclusive of banks and insurance companies.

Includes savings and loan associations, credit unions, nonprofit institutions, mutual savings banks, corporate pension trust funds, dealers and brokers, certain government deposit accounts, and government-sponsored enterprises.

Table A2

Financial Futures and Options: Exchanges, Contracts, and Volume of Contracts Traded

article image
article image
article image
article image
article image
article image
article image
article image
article image
Sources: American Stock Exchange; Battley (1993); Belgian Futures and Options Exchange; Chicago Board of Trade; Chicago Board Options Exchange; Chicago Mercantile Exchange; Deutsche Terminbörse; European Options Exchange; Financial Futures Market; Finnish Options Market Exchange and Clearing House; Futures Industry Association; Futures Industry Institute, Fact Book 1993; Guarantee Fund for Danish Options and Futures; Hong Kong Futures Exchange; London International Financial Futures Exchange; Marché à Terme International de France; Mid-America Commodity Exchange; Montreal Exchange; New York Cotton Exchange; New York Stock Exchange; Pacific Stock Exchange; Philadelphia Stock Exchange; Singapore International Monetary Exchange; Stockholm Options Market; and Toronto Futures Exchange. Note: n.t. = not traded; — = either zero or less than 500 contracts; $A = Australian dollar; S = Austrian schilling; BF = Belgian franc; Can$ = Canadian dollar; DKr = Danish krone; DM = deutsche mark; ECU = European currency unit; Fmk = Finnish markka; F = French franc; HK$ = Hong Kong dollar; £Ir = Irish pound; Lit = Italian lira; ¥ = Japanese yen; f. = Netherlands guilder; $NZ = New Zealand dollar; NKr = Norwegian krone; £ = pound sterling; Ptas = Spanish pesetas; SKr = Swedish krona; Sw F = Swiss franc; and $ = U.S. dollar; Options volume is puts and calls combined.

Blanks in this column indicate that information is not available.

Data on 6½–10-year and 5-year notes for 1988 and 1989; from 1990 onward it also includes 2-year notes.

Face value of contract is $100,000 for 6½–10-year and 5-year notes, and $200,000 for 2-year notes.

There are five types: mortgage-backed 7.5 percent, 8.0 percent, 8.5 percent, 9.0 percent, and 9.5 percent. The value presented for each year is the total volume of contracts traded for those types existed in the respective year.

Data on 10-year notes for 1988–89; 10- and 5-year notes for 1990–91; and 10-, 5-, and 2-year notes from 1992 onward. Face value of contract is $100,000 for 5- and 10-year notes; and $200,000 for 2-year notes.

CME deutsche mark, Eurodollar, Japanese yen, and pound sterling contracts are listed on a mutual offset link with the Singapore International Monetary Exchange.

Five-year notes for 1988; and 2- and 5-year notes from 1989 onward.

Commodity size is $250,000 for 5-year notes and $500,000 for 2-year notes.

NYFE is a subsidiary of the New York Stock Exchange.

Includes NYSE composite index and NYSE beta index (discontinued trading in 1988).

Includes AMEX major market index, AMEX institutional index, AMEX computer technology index, AMEX oil index, Japan index (trading began September 1990), major market index LEAPS (major market index divided by 10; trading began November 1990), international market index (options on an index of American depository receipts; delisted June 21, 1991), AMEX major market index capped options (delisted June 22, 1992), AMEX institutional index capped options (a.m. settled index capped options were delisted September 21, 1992), S&P MidCap index options (trading began February 13, 1992), Eurotop index options (trading began October 26, 1992), Pharmaceutical index options (trading began June 26, 1992), Biotechnology index options (trading began October 9, 1992), Morgan Stanley cyclical and consumer index options (trading began September 21, 1993), and North American telecommunications index options (trading began November 23, 1993).

Covers Australian dollar, Canadian dollar, deutsche mark, ECU, French franc, Japanese yen, pound sterling, and Swiss franc.

Include American and European options.

PHLX value line index, PHLX national OTC index, and bank index (since 1992).

A contract identical to the MATIF treasury bond future is also traded on an over-the-counter basis outside exchange hours and is cleared by the clearinghouse.

BTP stands for Buoni del Tesoro Poliennali.

Trading began on July 13, 1987 for the old JGB. A new JGB was launched on April 3, 1991, with modified specifications.

The 1988 data also contain 54,108 contracts traded on medium gilts (£50,000).

Include one- and three-month futures. Face value of contract for the one-month futures is Can$3 million and for the three-month futures is Can$1 million.

Includes 90-day MIBOR. Beginning from 1993, it also has data for 360-day MIBOR.

The MMI option is also listed on the American Stock Exchange.

Table A3

Markets for Selected Derivative Financial Instruments: Notional Principal Amounts Outstanding

(In billions of U.S. dollars, end-year data)

article image
Sources: Bank for International Settlements.

Traded on the Chicago Mercantile Exchange-International Monetary Market (CME-IMM), Singapore International Monetary Exchange (SIMEX), London International Financial Futures Exchange (LIFFE), Tokyo International Financial Futures Exchange (TIFFE), and Sydney Futures Exchange (SFE).

Traded on the TIFFE and SIMEX.

Traded on the Marché à Terme International de France (MATIF) and LIFFE.

Traded on the Chicago Board of Trade (CBOT), LIFFE, Mid-America Commodity Exchange (MIDAM), New York Futures Exchange (NYFE), and Tokyo Stock Exchange (TSE).

Traded on the MATIF.

Traded on the TSE, LIFFE, and CBOT.

Traded on the LIFFE and the Deutsche Terminbörse (DTB).

Calls plus puts.

Table A4

Annual Turnover in Derivative Financial Instruments Traded on Organized Exchanges Worldwide

(In millions of contracts traded)

article image
Sources: Bank for International Settlements.

Traded on the Chicago Mercantile Exchange-International Monetary Market (CME-IMM), Singapore International Monetary Exchange (SIMEX), London International Financial Futures Exchange (LIFFE), Tokyo International Financial Futures Exchange (TIFFE), and Sydney Futures Exchange (SFE).

Traded on the TIFFE and SIMEX.

Traded on the Marché à Terme International de France (MATIF) and LIFFE.

Traded on the Chicago Board of Trade (CBOT), LIFFE, Mid-America Commodity Exchange (MIDAM), New York Futures Exchange (NYFE), and Tokyo Stock Exchange (TSE).

Traded on the MATIF.

Traded on the TSE, LIFFE, and CBOT.

Traded on the LIFFE and the Deutsche Terminbörse (DTB).

Calls plus puts.

Table A5

Notional Principal Value of Outstanding Interest Rate and Currency Swaps1

(In billions of U. S. dollars)

article image
Sources: Bank for International Settlements, International Banking and Financial Market Developments, various issues; and International Swaps and Derivatives Association (ISDA).

End-December.

Including international institutions.

Including others.

Adjusted for double-counting as each currency swap involves two currencies.

Table A6

New Interest Rate Swaps and Currency Swaps1

(In billions of U. S. dollars)

article image
Sources: Bank for International Settlements, International Banking and Financial Market Developments, various issues; and International Swaps and Derivatives Association (ISDA).

During the respective half of the year.

Including international institutions.

Including others.

Adjusted for double-counting as each currency swap involves two currencies.

Table A7

Currency Composition of Notional Principal Value of Outstanding Interest Rate and Currency Swaps

(In billions of U.S. dollars)

article image
Sources: Bank for International Settlements, International Banking and Financial Market Developments, various issues; and International Swaps and Derivatives Association (ISDA).

Adjusted for double-counting as each currency swap involves two currencies.

Table A8

Changes in Claims on Nonbanks and Liabilities to Nonbanks1

(In billions of U. S. dollars)

article image
Sources: Bank for International Settlements, data reported to the IMF on currency distribution of banks’ external accounts; International Monetary Fund, International Financial Statistics; and IMF staff estimates.

Data on changes in claims and liabilities are derived from stock data on the reporting countries’ assets and liabilities and are net of changes due to exchange rate movements.

As measured by changes in the outstanding liabilities of borrowing countries, defined as cross-border bank credits to nonbanks by residence of borrower.

Excluding offshore centers. Data include some accumulation of interest arrears and reduction in bank claims resulting from debt conversions, sales, and write-offs.

The Bahamas, Bahrain, the Cayman Islands, Hong Kong, the Netherlands Antilles, Panama, and Singapore.

The difference between the amount that countries report as their banks’ positions with nonresident nonbanks in their monetary statistics and the amount that all banks in major financial centers report as their positions with nonbanks in each country.

As measured by changes in the outstanding assets of depositing countries, defined as cross-border bank deposits of nonbanks by residence of depositor.

Change in claims minus change in liabilities.

Table A9

Changes in Interbank Claims and Liabilities1

(In billions of U.S. dollars)

article image
Sources: Bank for International Settlements, data reported to the IMF on currency distribution of banks’ external accounts; International Monetary Fund, International Financial Statistics; and IMF staff estimates.

Data on changes in claims and liabilities are derived from stock data on the reporting countries’ assets and liabilities and are net of changes due to exchange rate movements.

As measured by changes in the outstanding liabilities of borrowing countries, defined as cross-border interbank accounts by residence of borrowing bank.

Excluding offshore centers. Data include some accumulation of interest arrears and reduction in bank claims resulting from debt conversions, sales, and write-offs.

The Bahamas, Bahrain, the Cayman Islands, Hong Kong, the Netherlands Antilles, Panama, and Singapore.

As measured by changes in the outstanding assets of depositing countries, defined as cross-border interbank accounts by residence of lending banks.

Change in claims minus change in liabilities.

Table A10

International Bond Issues by Developing Countries and Regions1

(In millions of U. S. dollars)

article image
Sources: IMF staff estimates based on Euroweek; Financial Times; International Financing Review, and Organization for Economic Cooperation and Development, Financial Market Trends and Financial Statistics Monthly.

Including note issues under European medium-term note (EMTN) programs.

Table A11

International Equity Issues by Developing Countries and Regions

(In millions of U.S. dollars)

article image
Sources: Euroweek; Financial Times; International Financing Review (IFR); and IFR Equibase.
Table A12

Comparative Valuations Across Emerging Markets1

(In ratios; end-December 1993 data)

article image
Source: International Finance Corporation.

Relative to the Morgan Stanley Capital International (MSCI) World Index.

Table A13

Restrictions on Outward Portfolio Investment of Institutional Investors in Five Major Industrial Countries

article image
Source: Chuhan (1994).

Restrictions on Postal Life Insurance are more prohibitive.

Does not include investments in Canada.

Investments are rated by the National Association of Insurance Commissioners (NAIC).

The Employee Retirement Income Security Act (ERISA) of 1974, which governs U.S. private pension funds, requires plan fiduciaries to exercise prudence in investment decisions. ERISA also requires plan trustees to diversify investments to minimize risk.

Table A14

Issues of Closed-End Funds Targeting Emerging Markets in Developing Countries and Regions

(In millions of U.S. dollars)

article image
Source: Lipper Analytical Services.
Table A15

Details of Treasury Note Issues in China

article image
Source: Bi(1993).

Where two interest rates are provided for pre-1989 issues, the higher rate was paid to individual investors, while the lower rate was paid to institutional investors.

In 1989 and 1990 the authorities announced their intention to issue Y 5.5 billion, but ended up issuing more.

In 1991 the authorities announced their intention to issue Y 10 billion.

Refers to announced plans.

The lower rate was paid to holders of the three-year bond.

The lower rate was paid to holders of the three-year bond. These interest rates were increased to 12.52 and 14.06 percent, respectively, in May 1993 and to 13.96 and 15.86 percent, respectively, in July 1993.

Table A16

Bonds Issued by China

article image
Sources: Euroweek; International Financing Review; and IMF staff estimates.

World Economic and Financial Surveys

This series (ISSN 0258-7440) contains biannual, annual, and periodic studies covering monetary and financial issues of importance to the global economy. The core elements of the series are the World Economic Outlook report, usually published in May and October, and the annual report on International Capital Markets. Other studies assess international trade policy, private market and official financing for developing countries, exchange and payments systems, export credit policies, and issues raised in the World Economic Outlook.

World Economic Outlook: A Survey by the Staff of the International Monetary Fund

The World Economic Outlook, published twice a year in English, French, Spanish, and Arabic, presents IMF staff economists’ analyses of global economic developments during the near and medium term. Chapters give an overview of the world economy; consider issues affecting industrial countries, developing countries, and economies in transition to the market; and address topics of pressing current interest. Annexes, boxes, charts, and an extensive statistical appendix augment the text.

ISSN 0256-6877.

$34.00 (academic rate: $23.00; paper).

1994 (May). ISBN 1-55775-381-4. Stock #WEO-194.

1994 (Oct.). ISBN 1-55775-385-7. Stock #WEO-294.

1993 (May). ISBN 1-55775-286-9. Stock #WEO-193.

1993 (Oct.). ISBN 1-55775-340-7. Stock #WEO-293.

International Capital Markets: Developments, Prospects, and Policy Issues

by an IMF Staff Team led by Morris Goldstein and David Folkerts-Landau

This annual report reviews developments in international capital markets, including recent bond market turbulence and the role of hedge funds, supervision of banks and nonbanks and the regulation of derivatives, structural changes in government securities markets, recent developments in private market financing for developing countries, and the role of capital markets in financing Chinese enterprises.

$20.00 (academic rate: $12.00; paper).

1994. ISBN 1-55775-426-8. Stock #WEO-694.

1993. Part I: Exchange Rate Management and International Capital Flows, by Morris Goldstein, David Folkerts-Landau, Peter Garber, Liliana Rojas-Suarez, and Michael Spencer.

ISBN 1-55775-290-7. Stock #WEO-693.

1993. Part II: Systemic Issues in International Finance, by an IMF Staff Team led by Morris Goldstein and David Folkerts-Landau.

ISBN 1-55775-335-0. Stock #WEO-1293.

Staff Studies for the World Economic Outlook

by the IMF’s Research Department

These studies, supporting analyses and scenarios of the World Economic Outlook, provide a detailed examination of theory and evidence on major issues currently affecting the global economy.

$20.00 (academic rate: $12.00; paper).

1993. ISBN 1-55775-337-7. Stock #WEO-393.

Developments in International Exchange and Payments Systems

by a Staff Team from the IMF’s Exchange and Trade Relations Department

The global trend toward liberalization in countries’ international payments and transfer systems has been most dramatic in central and Eastern Europe. But developing countries in general have brought their exchange systems more in line with market principles and moved toward more flexible exchange rate arrangements, while industrial countries have moved toward more pegged arrangements.

$20.00 (academic rate: $12.00; paper).

1992. ISBN 1-55775-233-8. Stock #WEO-892.

Private Market Financing for Developing Countries

by a Staff Team from the IMF’s Policy Development and Review Department led by Charles Collyns

This study surveys recent trends in private market financing for developing countries, including flows to developing countries through banking and securities markets; the restoration of access to voluntary market financing for some developing countries; and the status of commercial bank debt in low-income countries. $20.00 (academic rate: $12.00; paper).

1993. ISBN 1-55775-361-X. Stock #WE0993.

1992. ISBN 1-55775-318-0. Stock #WEO-992.

Issues and Developments in International Trade Policy

by an IMF Staff Team led by Margaret Kelly and Anne Kenny McGuirk

Since the mid-1980s, most developing countries have moved toward outward-looking, market-oriented policies and have liberalized their trade regimes. At the same time, industrial countries have acted to liberalize financial markets and foreign direct investment, deregulate services, and privatize public enterprises. This study discusses these and other developments in industrial, developing, and transition economies.

$20.00 (academic rate: $12.00; paper).

1992. ISBN 1-55775-311-3. Stock #WEO-1092.

Official Financing for Developing Countries

by a Staff Team from the IMF’s Policy Development and Review Department led by Michael Kuhn

This study provides information on official financing for developing countries, with the focus on low- and lower-middle-income countries. It updates and replaces Multilateral Official Debt Rescheduling: Recent Experience and reviews developments in direct financing by official and multilateral sources.

$20.00 (academic rate: $12.00; paper)

1994. ISBN 1-55775-378-4. Stock #WE0-1394.

Officially Supported Export Credits: Developments and Prospects

This study examines export credit and cover policies in the ten major industrial countries.

$15.00 (academic rate: $12.00; paper).

1990. By G.G. Johnson, Matthew Fisher, and Elliot Harris.

ISBN 1-55775-139-0. Stock #WEO-588.

Available by series subscription or single title (including back issues); academic rate available only to full-time university faculty and students.

Please send orders and inquiries to:

International Monetary Fund, Publication Services, 700 19th Street, N.W.

Washington, D.C. 20431, U.S.A.

Tel.: (202) 623-7430 Telefax: (202) 623-7201

References

  • “Administrative Arrangements Regarding the Auction of Government of Canada Securities,” Bank of Canada Review (Summer 1993), pp. 71–76.

    • Search Google Scholar
    • Export Citation

  • Altman, Edward I., “Research Update: Mortality Rates and Losses, Bond Rating Drift,” Merrill Lynch Merchant Banking Group (1989).

  • Bailey, Warren, “Risk and Return on China’s New Stock Markets: Some Preliminary Evidence,” Pacific Basin Finance Journal, Vol. 2 (1994), pp. 243–60.

    • Search Google Scholar
    • Export Citation

  • Bank for International Settlements, International Convergence of Capital Measurement and Capital Standards, Part II (Basle: Bank for International Settlements, July 1988).

    • Search Google Scholar
    • Export Citation

  • Bank for International Settlements, International Banking and Financial Market Developments (Basle: Bank for International Settlements, February 1994).

    • Search Google Scholar
    • Export Citation

  • Bank for International Settlements,, and Group of Ten, Report of the Committee on Interbank Netting Schemes of the Central Banks of the Group of Ten Countries (Basle: Bank for International Settlements, 1990).

    • Search Google Scholar
    • Export Citation

  • Bank of Japan, Economic Statistics Monthly (November 1993).

  • Bank of New York, The Complete Depositary Receipt Directory (New York: Bank of New York, 1993).

  • Basle Committee on Banking Supervision, “The Supervisory Treatment of Market Risks” (Basle: Basle Committee on Banking Supervision, April 1993).

    • Search Google Scholar
    • Export Citation

  • Battley, Nick, ed., The World’s Futures Options Markets (Chicago: Probus Publishing Company, 1993).

  • Bell, Michael W., and others, China at the Threshold of a Market Economy, IMF Occasional Paper No. 107 (Washington: International Monetary Fund, September 1993).

    • Search Google Scholar
    • Export Citation

  • Bi, Keqian, “Credit Markets in China,” Columbia Journal of World Business, Vol. 28 (Fall 1993), pp. 76–95.

  • Bikhchandani, Sushil, and Chi-fu Huang, “The Economics of Treasury Securities Markets,” Journal of Economic Perspectives, Vol. 7, No. 3 (Summer 1993), pp. 117–34.

    • Search Google Scholar
    • Export Citation

  • Blommestein, Hans J., and Michael G. Spencer, “The Role of Financial Institutions in the Transition to a Market Economy,” IMF Working Paper No. 93/75 (Washington: International Monetary Fund, October 1993).

    • Search Google Scholar
    • Export Citation

  • Board of Governors of the Federal Reserve System, Flow of Funds Accounts, Third Quarter 1993 (Washington: Board of Governors of the Federal Reserve System, December 1993).

    • Search Google Scholar
    • Export Citation

  • Board of Governors of the Federal Reserve System(1994a), Federal Reserve Bulletin, Vol. 80, No. 1 (January 1994).

  • Board of Governors of the Federal Reserve System(1994b), Federal Reserve Bulletin, Vol. 80, No. 3 (March 1994).

    Board of Governors of the Federal Reserve System Federal Deposit Insurance Corporation, and Office of the Comptroller of Currency, “Derivative Product Activities of Commercial Banks: Joint Study Conducted in Response to Questions Posed by Senator Riegle on Derivative Products,” January 27, 1993, Exhibit 1.

    • Search Google Scholar
    • Export Citation

  • Bröker, Günther, Government Securities and Debt Management in the 1990s (Paris: Organization for Economic Cooperation and Development, 1993).

    • Search Google Scholar
    • Export Citation

  • Campbell, John Y., and Kenneth A. Froot, “International Experiences with Securities Transactions Taxes,” NBER Working Paper No. 4587 (Cambridge, Massachusetts: National Bureau of Economic Research, December 1993).

    • Search Google Scholar
    • Export Citation

  • Chen, Yung-Sheng, “New Developments in Mainland China’s Securities Markets,” Issues and Studies, Vol. 27 (August 1991), pp. 82–103.

    • Search Google Scholar
    • Export Citation

  • Chuhan, Punam, “Are Institutional Investors an Important Source of Portfolio Investment in Emerging Markets?” World Bank Policy Research Paper No. 1243 (Washington: World Bank, January 1994).

    • Search Google Scholar
    • Export Citation

  • Collyns, Charles, and others, Private Market Financing to Developing Countries, World Economic and Financial Surveys (Washington: International Monetary Fund, December 1993).

    • Search Google Scholar
    • Export Citation

  • European Community, Council Directive 93/6/EEC, “On the Capital Adequacy of Investments Firms and Credit Institutions,” Official Journal of the European Communities (Luxembourg, June 11, 1993).

    • Search Google Scholar
    • Export Citation

  • Evans, Nicolas, “The Global Bond Era Dawns,” Euroweek (September 24, 1993), pp. 10–17.

  • Federal Register, Vol. 58, No. 176 (September 14, 1993), p. 482–06.

  • Federal Reserve Bank of New York, “Administration of Relationships with Primary Dealers,” Appendix E in U.S. Department of the Treasury, U.S. Securities and Exchange Commission, and Board of Governors of the Federal Reserve System, Joint Report on the Government Securities Market (Washington: U.S. Government Printing Office, January 1992).

    • Search Google Scholar
    • Export Citation

  • Feldman, Robert, and Rajnish Mehra, “Auctions: Theory and Applications,” Staff Papers, International Monetary Fund, Vol. 43 (September 1993), pp. 485–511.

    • Search Google Scholar
    • Export Citation

  • “Financial Centre Germany: Underlying Conditions and Recent Development,” Monthly Report of the Deutsche Bundesbank, Vol. 44, No. 3, March 1992, pp. 23–31.

    • Search Google Scholar
    • Export Citation

  • Folkerts-Landau, David, and Peter M. Garber, “The European Central Bank: A Bank or a Monetary Policy Rule?” NBER Working Paper No. 4016 (Cambridge, Massachusetts: National Bureau of Economic Research, March 1992).

    • Search Google Scholar
    • Export Citation

  • Giavazzi, Francesco, and Marco Pagano, “Confidence Crises and Public Debt Management,” in Public Debt Management: Theory and History, ed. by Rudiger Dornbusch and Mario Draghi (Cambridge: Cambridge University Press, 1990).

    • Search Google Scholar
    • Export Citation

  • Global Derivatives Study Group, Derivatives: Practices and Principles (Washington: Group of Thirty, 1993).

  • Goldstein, Morris, David Folkerts-Landau, and others, International Capital Markets: Developments, Prospects, and Policy Issues, World Economic and Financial Surveys (Washington: International Monetary Fund, September 1992).

    • Search Google Scholar
    • Export Citation

  • Goldstein, Morris, David Folkerts-Landau, (1993a), International Capital Markets: Part I Exchange Rate Management and International Capital Flows, World Economic and Financial Surveys (Washington: International Monetary Fund, April 1993).

    • Search Google Scholar
    • Export Citation

  • Goldstein, Morris, David Folkerts-Landau, (1993b), International Capital Markets: Part II. Systemic Issues in International Finance, World Economic and Financial Surveys (Washington: International Monetary Fund, August 1993).

    • Search Google Scholar
    • Export Citation

  • Goulder, Lawrence H., “Implications of Introducing U.S. Withholding Taxes on Foreigners’ Interest Income,” in Tax Policy and the Economy, ed. by Lawrence H. Summers (Cambridge, Massachusetts: MIT Press, 1990).

    • Search Google Scholar
    • Export Citation

  • Greenspan, Alan, Testimony Before the Committee on Banking, Housing, and Urban Affairs, U.S. Senate (Washington, March 2, 1994).

  • Howell, Michael J., Angela Cozzini, and Luci Greenwood, Cross-Border Capital Flows: A Study of Foreign Equity Investment—1991/92 Review (London: Baring Securities, 1993).

    • Search Google Scholar
    • Export Citation

  • Howell, Michael J., Angela Cozzini, and Luci Greenwood, North America, Europe, Africa, Pacific Rim, and Latin America: Global and Emerging Markets Strategy (London: Baring Securities, April 1994).

    • Search Google Scholar
    • Export Citation

  • Hu, Yebi, China’s Capital Market (Hong Kong: Chinese University Press, 1993).

  • Huizinga, Harry, “Withholding Taxes and the Cost of Public Debt,” IMF Working Paper No. 94/18 (Washington: International Monetary Fund, February 1994).

    • Search Google Scholar
    • Export Citation

  • International Finance Corporation, Quarterly Review of Emerging Markets, Fourth Quarter 1993 (Washington: International Finance Corporation, 1994).

    • Search Google Scholar
    • Export Citation

  • International Monetary Fund, World Economic Outlook, October 1993: A Survey by the Staff of the International Monetary Fund, World Economic and Financial Surveys (Washington: International Monetary Fund, October 1993).

    • Search Google Scholar
    • Export Citation

  • International Monetary Fund, World Economic Outlook, April 1994: A Survey by the Staff of the International Monetary Fund, World Economic and Financial Surveys (Washington: International Monetary Fund, April 1994).

    • Search Google Scholar
    • Export Citation

  • Investment Company Institute, Mutual Fund Factbook (Washington: Investment Company Institute, 33rd ed., 1993).

  • Kleiman International Consultants, Fifth Annual Emerging Stock Markets Survey (November 1993).

  • Kroszner, Randall S., “An International Comparison of Primary Government Securities Markets,” in U.S. Securities Market: The Scholars’ Assessment, Vol. 1 (Chicago: MidAmerica Institute, April 1993).

    • Search Google Scholar
    • Export Citation

  • Loser, Claudio, and Eliot Kalter, eds., Mexico: The Strategy to Achieve Sustained Economic Growth, IMF Occasional Paper No. 99 (Washington: International Monetary Fund, September 1992).

    • Search Google Scholar
    • Export Citation

  • McDonough, William J., Remarks Before the Sixty-Sixth Annual Mid-Winter Meeting of the New York State Bankers Association (March 1994).

  • Miller, Merton, “Functional Regulation,” Address at the Fifth Annual Pacific-Basin Capital Markets Finance Conference (Kuala Lumpur, June 1993).

    • Search Google Scholar
    • Export Citation

  • Moody’s Bond Survey, March 7, 1994.

  • Mullins, David W., Jr., Statement Before the Subcommittee on Securities of the Committee on Banking, Housing, and Urban Affairs, U.S. Senate (Washington, January 23, 1992); reproduced in Board of Governors of the Federal Reserve System, Federal Reserve Bulletin, Vol 78, No. 3 (March 1992), pp. 195–98.

    • Search Google Scholar
    • Export Citation

  • Mussa, Michael, and Morris Goldstein, “The Integration of World Capital Markets,” in Changing Capital Markets: Implications for Monetary Policy (Kansas City: Federal Reserve Bank of Kansas City, 1993).

    • Search Google Scholar
    • Export Citation

  • Pensions and Investments, January 25, 1993.

  • Pensions and Investments, January 24, 1994.

  • Pirrong, S. Craig, “The Market for Treasury Derivative Securities: Microstructure and Market Power,” in U.S. Treasury Securities Market: The Scholars’ Assessment, Vol. 1 (Chicago: MidAmerica Institute, April 1993).

    • Search Google Scholar
    • Export Citation

  • Quale, Andrew, “Legal Primer for Non-U.S. Issuers Raising Financing in the United States,” paper presented at the Second Annual Conference on Emerging Market High-Yield Sovereign and Corporate Bonds (New York, January 22, 1993).

    • Search Google Scholar
    • Export Citation

  • Republic New York Securities, Year End Hedge Fund Review 1993 (1994).

  • Rodriguez, Anthony P., “Government Securities Investments of Commercial Banks,” Quarterly Review, Federal Reserve Bank of New York (Summer 1993), pp. 39–53.

    • Search Google Scholar
    • Export Citation

  • Sargent, Thomas J., “Fact or Fiction: Shortening Debt Maturity Lowers Interest Costs” (Chicago: Catalyst Institute, December 1993).

1

As noted in the introduction, the figures discussed here pertain only to central government debt. In some countries, the debt of other levels of government also is significant.

2

The recent increase in bank holdings in the United States is analyzed by Rodriguez (1993).

3

The data for Sweden are problematic, owing to a change in the basis for reporting. The reported figure for foreign holdings for 1992 includes all debt in foreign currency plus foreign holdings of domestic currency debt, while data for earlier years are not broken down in this way.

4

Data were not available for Italy, and French data were only available for 1992. The maturity of government debt is a flawed measure of the term of the debt, but better measures such as duration are generally unavailable.

5

In addition to the countries discussed in this section, other countries, including Italy, Canada, Spain, and Finland, have primary dealer systems in place.

6

See section on Liquidity in the Secondary Market for a discussion of interdealer brokers.

10

Since the Japanese Government pays a commission on both the auctioned and the syndicated share, the commission does not result in the price on the syndicated tranche being higher than on the auctioned tranche. See Kroszner (1993) for details of the syndication process.

11

For the ten-year Bund, about 30 percent each is sold through the syndicate and the tap and 40 percent is sold at auction. Kroszner (1993) contains details on the German bond consortium and the issuance of government bonds in Germany.

12

BTFs denote Bons du Trésor à taux fixe; BTANs, Bons du Trésor à taux annuel; and OATs, Obligations Assimilables du Trésor.

13

On the New York Stock Exchange, for example, designated specialists in particular stocks are charged with maintaining continuous and orderly markets. In return for this, these specialists have privileged access to the order flow in the stocks.

14

In addition to formal tap sales, in many cases central banks purchase part of the debt issued by the government and subsequently sell debt in the secondary market. Such operations bear some resemblance to tap sales, although central banks often segregate securities used for open market operations from those held for investment.

15

Uniform-price and discriminatory systems have a variet) of other names. The terminology of Bikhchandani and Huang (1993) is followed here.

16

Canada, however, uses syndicates for its inflation-indexed Real Return Bonds, which it began to issue in 1991.

17

These issues are discussed in Bikhchandani and Huang (1993) and in Feldman and Mehra (1993).

18

See Pirrong (1993) for some background on the U.S. when-issued market.

20

The table involves a number of distortions. The U.S. data exclude transactions not involving primary dealers. The German data exclude OTC and offshore transactions, which are substantially more than half—perhaps 80–90 percent—of trading volume. The French data include repurchase agreements.

21

See Kroszner (1993) for more information on the JGB benchmark.

22

In some cases, the new tranche of the reopened security is not identical to the existing security until after the first coupon has been paid.

23

A reverse repo is the other side of a repo. It entails purchasing a security and agreeing to resell it in the future.

24

Continuing repo contracts are contracts that can be terminated at any time by either party.

25

Data from the Board of Governors of the Federal Reserve System (1994a), Tables 1.41 and 1.43.

26

The tax rate paid by the seller of the bond is 0.01 percent of the transfer price for financial institutions and 0.03 percent for other sellers. The transaction tax is applied at both ends of the Gensaki, both on the sale and on the subsequent repurchase.

27

Data from Bank of Japan (1993), pp. 83, 171. Government debt outstanding excludes debt held by the Government and the Bank of Japan.

28

Bonds issued by foreign issuers in the U.S. domestic market are known as “Yankee bonds.”

29

See Box 3 for a discussion of withholding taxes and their effect on borrowing costs.

31

These bills are issued by the Bundesbank for monetary policy purposes and not to finance the federal deficit.

34

“Financial Centre Germany: Underlying Conditions and Recent Development,” Monthly Report of the Deutsche Bundesbank, March 1992, pp. 23–31.

35

See “Administrative Arrangements Regarding the Auction of Government of Canada Securities,” Bank of Canada Review (Summer 1993), pp. 71–76.

1

There are also concerns about the involvement of nonbanks in the derivative markets; see section on Strengthening Capital Requirements, below.

3

The notional principal value refers to the face value of the instruments underlying the derivative contract. This is used to calculate payments under the contract and is not a measure of the exposure of the institutions holding these contracts. The latter is provided by the replacement cost of the contract, which is estimated at approximately 2 percent of the notional value for interest rate swaps and 5 percent for currency swaps.

4

The measure of turnover used in Table A4 is the number of contracts traded. Comparisons over time are complicated, since contracts representing different amounts of the underlying instruments are traded on different exchanges and new contracts are frequently introduced.

5

See Goldstein, Folkerts-Landau, and others (1993a) for a discussion of the impact of the exchange rate crisis in September 1992 on derivatives trading.

6

These complex products have names such as swaptions, forward swaps, knockouts, step-up recovery floaters, index-amortizing notes, and lookback options.

7

See Board of Governors of the Federal Reserve System and others (1993).

8

While many mutual funds are barred by their investment guidelines from participating in the swaps markets, some are bypassing the restriction by investing in structured notes.

9

With this increased use of derivatives have come more frequent reports of losses as in the cases of, among others, Metallgesellschaft, Codelco, Rockefeller Center Properties, Procter and Gamble, and Gibson Greetings (see Box 1).

12

Different risk weights are proposed for qualifying issues with different residual maturity because uncertainty about creditworthiness increases with the life of the security.

14

For precise definitions of the different categories, see Bank for International Settlements (1988), Part II.

15

Delta measures the sensitivity of the price of a derivative security with respect to a change in the price of the underlying security.

16

The representative duration measure for a particular maturity band is computed as the modified duration of a bond with a maturity equal to the mid-point of the time band, assuming an 8 percent interest rate and an 8 percent coupon. Furthermore, the assumed interest rate change is designed to cover about two standard deviations of one month’s yield volatility in most major markets.

17

See Section III for a description of how the shorthand method is applied to foreign exchange exposures.

19

See Federal Register (1993).

21

The importance of managerial oversight of risk management operations is exemplified by the recent losses incurred by Codelco and Metallgesellschaft discussed in Box 1.

1

For further discussion on global bonds, see Annex I.

2

The Euronote category comprises Euro-commercial paper and EMTN in addition to note-issuance facilities and other short- and medium-term borrowing facilities.

3

For further discussion on emerging markets, see Annex IV.

4

The following centers report activity for the international banking statistics: Australia, Austria, The Bahamas, Bahrain, Belgium, Canada, the Cayman Islands, Chile, Denmark, Finland, France, Germany, Hong Kong, Ireland, Italy, Japan, Korea, Luxembourg, the Netherlands, the Netherlands Antilles, Norway, Panama, Portugal, Saudi Arabia, Singapore, Spain, Sweden, Switzerland, the United Arab Emirates, the United Kingdom, and the United States.

6

The Danish banking industry, while also suffering from high loan losses, has not required government support on anything like the scale of the other Nordic countries. The four major Danish banks—Den Danske Bank, Unibank, Sparekassen Bikuben, and Jyske Bank—reported operating results in 1992 not unlike those in Finland, Norway, or Sweden. High loan-loss provisions and large valuation losses on securities holdings contributed to a combined pretax loss of DKr 8.2 billion in 1992 (compared with DKr 338 million in 1991). Despite continued high loan-loss provisions, the banks’ pretax profits turned sharply upward in 1993, owing in large part to gains on bond holdings (which are carried at market value). At year-end, the banks’ combined pretax profit was DKr 5.5 billion.

While all of the Danish banks have incurred substantial loan losses in recent years, problems at the eighth largest bank, Varde Bank, had grown so serious in November 1992 that the central bank arranged a consortium of the seven larger banks to provide a DKr 750 million guarantee fund. Despite some balance sheet restructuring, the central bank determined in 1993 that the bank was about to fail. Consequently, it was decided in December 1993 to transfer the DKr 7 billion in performing assets to Sydbank Sonderjylland and to retain the DKr 4 billion in nonperforming assets in Varde Bank, which will be wound down over the next few years by the central bank with the support of a government guarantee. This is another example of the “good bank/bad bank” model for resolving banking difficulties.

However, the Danish banking situation differed in at least two important respects from those in the other three Nordic countries. First, Danish banks have always been well capitalized. Even at the end of 1992, the major banks had capital/ asset ratios in excess of 11 percent. Second, banks are required to report the current market value of their securities holdings and must set aside loan-loss provisions at the first hint of problems rather than waiting for arrears to accumulate. As a result, bank profitability will be affected by a downturn in the economy earlier than otherwise, but banks will be better equipped to deal with any problems that emerge.

7

Earlier exchange rate depreciation had actually contributed to problems in the Finnish banking sector. The corporate sector had relied heavily on foreign-currency-denominated bank loans and many of these became nonperforming in 1991–92 when the depreciation of the markka led to significantly higher debt-service costs.

8

The estimated gross amounts committed, with attribution, are SKr 92 billion in Sweden (Bank Support Authority), Fmk 39 billion in Finland (IBCA Limited), and NKr 28.2 billion in Norway (IBCA Limited).

9

The Savings Bank of Finland was itself formed in 1992 by the merger of 41 small savings banks.

10

The BSA provided a SKr 2.5 billion capital-adequacy guarantee for the initial public offering. This guarantee can be utilized if the bank’s capital falls below 9 percent of its risk-weighted assets. If fully utilized, this facility would result in a 54 percent BSA share in Foreningsbanken equity (86 percent of the votes).

12

The official definition of nonperforming loans includes loans to borrowers that have legally been declared bankrupt (¥ 2.3 trillion) and loans on which interest has not been paid for 180 days (¥ 11.3 trillion). These figures on nonperforming loans do not include restructured loans. The Japan Center for International Finance estimates the stock of restructured loans at ¥ 13–14 trillion. The increase in nonperforming loans between March and September 1993 is a net number: nonperforming loans that have been sold to the Cooperative Credit Purchasing Company or charged off are removed from the non-performing category, as well as from the balance sheets of banks.

13

As the data in Table 17 show, most loan sales to the CCPC have been arranged immediately prior to the end-September and end-March reporting dates.

14

This measure differs from the Japanese definition of operating earnings, which includes unrealized gains and losses on the investment bond portfolio. These are removed to approximate banks’ cash flow.

15

In 1993/94, loans were sold to the CCPC at an average discount of 51 percent.

16

Based on data provided by IBCA Limited.

1

For a description of China’s external borrowing strategy, see Annex V of this report.

2

A Matador bond is a bond issued in Spain by a nonresident.

4

See Pensions and Investments, January 25, 1993.

5

Information provided by Morningstar and is based on its data base of about 3,500 U.S. open-end mutual funds. These data exclude closed-end funds. Information for before 1993 is not available.

6

See Pensions and Investments, January 24, 1994, p. 17.

8

Board of Governors of the Federal Reserve System (1993). The information on mutual funds includes open-end and closed-end funds and money market mutual funds.

10

J.P. Morgan, Latin American Eurobond Index.

13

Under this type of rule, a regulator requires a pension fund or an insurance company to exercise prudence—which is not defined precisely—in their investment decisions.

14

Pensions and Investments, January 24, 1994.

15

There are also unsponsored depository receipts, but these are now obsolete. This discussion of depository receipts and private placements is based on Bank of New York (1993) and Quale (1993).

16

A U.S. dollar interest rate is presented here for illustrative purposes. For bonds denominated in other currencies, the appropriate comparison would be with a government instrument denominated in the same currency.

17

Moody’s Bond Survey, March 7, 1994, p. 6838.

19

This strategy is explained in more detail in Loser and Kalter (1992) and Collyns and others (1993), Section V.

2

The ten-year bonds issued prior to 1985 actually had an effective maturity of seven and one half years, since the maturity date of a certificate was determined by lottery, with 20 percent of each issue maturing in each year after the fifth year following issue.

3

Although in some cases, the purchase of a certain large number of shares gave the investor the right to nominate a member of the board of directors.

5

Data on trading activity in Shanghai and Shenzhen prior to 1991 are from Hu (1993), Chapter 3.

6

Data from International Finance Corporation (1994). The distinction between A shares (which are reserved for mainland Chinese investors) and B shares (which are reserved for non-Chinese investors) is discussed below.

7

In September 1990, the PBOC headquarters set up a Quotation Center for government securities, which provided on-line pricing information to securities dealers. However, it never provided a trading facility and has subsequently been overtaken by the STAQS.

8

Renamed the Shanghai Stock Exchange in October 1993.

9

The interest rate was set at 1.5 percentage points above the deposit rate. Interestingly, the same year’s Y 2 billion Ministry of Finance note issue was distributed through forced allocations.

10

Bond futures were listed on the Shenzhen exchange in March 1994.

11

For example, Japanese banks have reduced considerably their lending to developing countries. See Goldstein, Folkerts-Landau, and others (1992) for a discussion of the Basle accord and the treatment of exposures to developing countries.

12

Foreign commercial banks and official (multilateral and bilateral) creditors have been the major source of external borrowing to date. At the end of June 1993, medium- and long-term debt stood at $16.7 billion to banks and $22.3 billion to official creditors, constituting about 25 percent and 34 percent, respectively, of the medium- and long-term debt at mid-1993.

13

External debt is guaranteed by the ten windows and local governments, and not by the Central Government (except for what it borrows directly).

14

At this stage, it was also announced that CITIC would no longer borrow on behalf of the Government.

15

The authorities have viewed official financing to be inflexible as its disbursement is conditional on various criteria—often linked to the use of the resources—despite its concessional terms.

16

For export credit agency-insured or for cofinanced credits from international financial institutions, the average maturity for credits of 12 years is longer than for bonds, with the average for all loans—7.3 years—approximately equal to the average on unenhanced bonds.

17

Only equity placements by companies based in China are considered as international equity placements, thus excluding international issues through special purpose vehicles (e.g., Bermuda-based subsidiaries) and secondary share listings. Notable examples of such exclusions are the New York Stock Exchange listings of Brilliance China Automobile (October 1992), China Tire (June 1993), and Ek Chor (July 1993). Other excluded secondary listings include those in Melbourne and Toronto.

18

IFC Emerging Markets Data Base.

19

An additional push to the B share market was given by the SEC assurance that U.S.-based investment funds could invest in China.

20

Much of this capital is then recycled back to China in the form of direct investment. However, the extent of the financial linkages is hard to disentangle as Hong Kong maintains no balance of payments statistics.

21

All but one H share listing were issued through a fixed-price offering and were usually heavily oversubscribed.

22

Of the remaining $4.3 billion, $0.7 billion came from Japan and $1.1 billion from Taiwan Province of China.

24

See Bell and others (1993) for a discussion of the history of enterprise reforms and especially the concept of ownership in China.

25

The role of banks and securities markets in providing a market-based corporate governance mechanism is discussed in Blommestein and Spencer (1993).

26

Judging by the market’s reaction to the prospects of level one ADRs for Shanghai B shares, this vehicle could prove to be useful in providing liquidity to the B share market; the Shanghai B share market was buoyed for several months at the beginning of 1993 by the prospects of ADR issues.

27

Hong Kong is supposed to retain nominal financial independence for another 50 years following its transfer to China in 1997.

28

The H share index is based on the Tsingtao, Shanghai Petrochemical Company, and Maanshan Iron and Steel issues weighted by their respective market capitalization.

29

Moody’s also upgraded three financial institutions (BOC, CITIC, and PCBC) to A3 in December 1993. Standard and Poor’s rates China a BBB, though the rating was upgraded from stable to improving.

  • Collapse
  • Expand
Cover World Economic and Financial Surveys
World Economic and Financial Surveys
International Capital Markets 1994
ISBN:
9781451942118
ISSN:
0258-7440
Publication Date:
01 Jan 1994
Publisher:
International Monetary Fund
DOI:
https://doi.org/10.5089/9781451942118.083
Pages:
145
  • <sup><xref rid="Rapp01fig02" ref-type="fig">Chart 2.</xref></sup>
    View in gallery
    Chart 2.

    Government Debt Outstanding in Major Industrial Countries

    (In percent of GDP)

  • <sup><xref rid="Rapp01fig03" ref-type="fig">Chart 3.</xref></sup>
    View in gallery
    Chart 3.

    Gross Issues of Medium- and Long-Term Government Securities in Major Industrial Countries

    (In percent of GDP)

  • <sup><xref rid="Rapp01fig04" ref-type="fig">Chart 4.</xref></sup>
    View in gallery
    Chart 4.

    Gross Issues of Short-Term Government Securities in Major Industrial Countries

    (In percent of GDP)

  • <sup><xref rid="Rapp03fig05" ref-type="fig">Chart 5.</xref></sup>
    View in gallery
    Chart 5.

    Long-Term Government Bond Yield for Seven Major Industrial Countries, January 1990–March 19941

    (In percent)

  • <sup><xref rid="Rapp03fig06" ref-type="fig">Chart 6.</xref></sup>
    View in gallery
    Chart 6.

    Major Industrial Countries: Changes in Stock Market Indices, January 1990–May 1994

    (Twelve-month changes, in percent)

  • <sup><xref rid="Rapp03fig07" ref-type="fig">Chart 7.</xref></sup>
    View in gallery
    Chart 7.

    Major Industrial Countries: Yield Differentials, January 1990–April 19941

    (In percent)

  • <sup><xref rid="Rapp04fig08" ref-type="fig">Chart 8.</xref></sup>
    View in gallery
    Chart 8.

    Share Price Indices for Selected Emerging Markets1

    (In U.S. dollar terms, December 1988 = 100)

  • <sup><xref rid="Rapp04fig09" ref-type="fig">Chart 9.</xref></sup>
    View in gallery
    Chart 9.

    Yield Spreads at Launch for Selected Developing Countries1

    (In basis points)

  • <sup><xref rid="Rapp05fig10" ref-type="fig">Chart 10.</xref></sup>
    View in gallery
    Chart 10.

    Selected Stock Exchange Indices in China and Hong Kong, 1993–February 1994

    (January 2, 1993 = 100)

Chart 2.

Government Debt Outstanding in Major Industrial Countries

(In percent of GDP)
Chart 3.

Gross Issues of Medium- and Long-Term Government Securities in Major Industrial Countries

(In percent of GDP)
Chart 4.

Gross Issues of Short-Term Government Securities in Major Industrial Countries

(In percent of GDP)
Chart 5.

Long-Term Government Bond Yield for Seven Major Industrial Countries, January 1990–March 19941

(In percent)
Chart 6.

Major Industrial Countries: Changes in Stock Market Indices, January 1990–May 1994

(Twelve-month changes, in percent)
Chart 7.

Major Industrial Countries: Yield Differentials, January 1990–April 19941

(In percent)
Chart 8.

Share Price Indices for Selected Emerging Markets1

(In U.S. dollar terms, December 1988 = 100)
Chart 9.

Yield Spreads at Launch for Selected Developing Countries1

(In basis points)
Chart 10.

Selected Stock Exchange Indices in China and Hong Kong, 1993–February 1994

(January 2, 1993 = 100)

  • “Administrative Arrangements Regarding the Auction of Government of Canada Securities,” Bank of Canada Review (Summer 1993), pp. 71–76.

    • Search Google Scholar
    • Export Citation

  • Altman, Edward I., “Research Update: Mortality Rates and Losses, Bond Rating Drift,” Merrill Lynch Merchant Banking Group (1989).

  • Bailey, Warren, “Risk and Return on China’s New Stock Markets: Some Preliminary Evidence,” Pacific Basin Finance Journal, Vol. 2 (1994), pp. 243–60.

    • Search Google Scholar
    • Export Citation

  • Bank for International Settlements, International Convergence of Capital Measurement and Capital Standards, Part II (Basle: Bank for International Settlements, July 1988).

    • Search Google Scholar
    • Export Citation

  • Bank for International Settlements, International Banking and Financial Market Developments (Basle: Bank for International Settlements, February 1994).

    • Search Google Scholar
    • Export Citation

  • Bank for International Settlements,, and Group of Ten, Report of the Committee on Interbank Netting Schemes of the Central Banks of the Group of Ten Countries (Basle: Bank for International Settlements, 1990).

    • Search Google Scholar
    • Export Citation

  • Bank of Japan, Economic Statistics Monthly (November 1993).

  • Bank of New York, The Complete Depositary Receipt Directory (New York: Bank of New York, 1993).

  • Basle Committee on Banking Supervision, “The Supervisory Treatment of Market Risks” (Basle: Basle Committee on Banking Supervision, April 1993).

    • Search Google Scholar
    • Export Citation

  • Battley, Nick, ed., The World’s Futures Options Markets (Chicago: Probus Publishing Company, 1993).

  • Bell, Michael W., and others, China at the Threshold of a Market Economy, IMF Occasional Paper No. 107 (Washington: International Monetary Fund, September 1993).

    • Search Google Scholar
    • Export Citation

  • Bi, Keqian, “Credit Markets in China,” Columbia Journal of World Business, Vol. 28 (Fall 1993), pp. 76–95.

  • Bikhchandani, Sushil, and Chi-fu Huang, “The Economics of Treasury Securities Markets,” Journal of Economic Perspectives, Vol. 7, No. 3 (Summer 1993), pp. 117–34.

    • Search Google Scholar
    • Export Citation

  • Blommestein, Hans J., and Michael G. Spencer, “The Role of Financial Institutions in the Transition to a Market Economy,” IMF Working Paper No. 93/75 (Washington: International Monetary Fund, October 1993).

    • Search Google Scholar
    • Export Citation

  • Board of Governors of the Federal Reserve System, Flow of Funds Accounts, Third Quarter 1993 (Washington: Board of Governors of the Federal Reserve System, December 1993).

    • Search Google Scholar
    • Export Citation

  • Board of Governors of the Federal Reserve System(1994a), Federal Reserve Bulletin, Vol. 80, No. 1 (January 1994).

  • Board of Governors of the Federal Reserve System(1994b), Federal Reserve Bulletin, Vol. 80, No. 3 (March 1994).

    Board of Governors of the Federal Reserve System Federal Deposit Insurance Corporation, and Office of the Comptroller of Currency, “Derivative Product Activities of Commercial Banks: Joint Study Conducted in Response to Questions Posed by Senator Riegle on Derivative Products,” January 27, 1993, Exhibit 1.

    • Search Google Scholar
    • Export Citation

  • Bröker, Günther, Government Securities and Debt Management in the 1990s (Paris: Organization for Economic Cooperation and Development, 1993).

    • Search Google Scholar
    • Export Citation

  • Campbell, John Y., and Kenneth A. Froot, “International Experiences with Securities Transactions Taxes,” NBER Working Paper No. 4587 (Cambridge, Massachusetts: National Bureau of Economic Research, December 1993).

    • Search Google Scholar
    • Export Citation

  • Chen, Yung-Sheng, “New Developments in Mainland China’s Securities Markets,” Issues and Studies, Vol. 27 (August 1991), pp. 82–103.

    • Search Google Scholar
    • Export Citation

  • Chuhan, Punam, “Are Institutional Investors an Important Source of Portfolio Investment in Emerging Markets?” World Bank Policy Research Paper No. 1243 (Washington: World Bank, January 1994).

    • Search Google Scholar
    • Export Citation

  • Collyns, Charles, and others, Private Market Financing to Developing Countries, World Economic and Financial Surveys (Washington: International Monetary Fund, December 1993).

    • Search Google Scholar
    • Export Citation

  • European Community, Council Directive 93/6/EEC, “On the Capital Adequacy of Investments Firms and Credit Institutions,” Official Journal of the European Communities (Luxembourg, June 11, 1993).

    • Search Google Scholar
    • Export Citation

  • Evans, Nicolas, “The Global Bond Era Dawns,” Euroweek (September 24, 1993), pp. 10–17.

  • Federal Register, Vol. 58, No. 176 (September 14, 1993), p. 482–06.

  • Federal Reserve Bank of New York, “Administration of Relationships with Primary Dealers,” Appendix E in U.S. Department of the Treasury, U.S. Securities and Exchange Commission, and Board of Governors of the Federal Reserve System, Joint Report on the Government Securities Market (Washington: U.S. Government Printing Office, January 1992).

    • Search Google Scholar
    • Export Citation

  • Feldman, Robert, and Rajnish Mehra, “Auctions: Theory and Applications,” Staff Papers, International Monetary Fund, Vol. 43 (September 1993), pp. 485–511.

    • Search Google Scholar
    • Export Citation

  • “Financial Centre Germany: Underlying Conditions and Recent Development,” Monthly Report of the Deutsche Bundesbank, Vol. 44, No. 3, March 1992, pp. 23–31.

    • Search Google Scholar
    • Export Citation

  • Folkerts-Landau, David, and Peter M. Garber, “The European Central Bank: A Bank or a Monetary Policy Rule?” NBER Working Paper No. 4016 (Cambridge, Massachusetts: National Bureau of Economic Research, March 1992).

    • Search Google Scholar
    • Export Citation

  • Giavazzi, Francesco, and Marco Pagano, “Confidence Crises and Public Debt Management,” in Public Debt Management: Theory and History, ed. by Rudiger Dornbusch and Mario Draghi (Cambridge: Cambridge University Press, 1990).

    • Search Google Scholar
    • Export Citation

  • Global Derivatives Study Group, Derivatives: Practices and Principles (Washington: Group of Thirty, 1993).

  • Goldstein, Morris, David Folkerts-Landau, and others, International Capital Markets: Developments, Prospects, and Policy Issues, World Economic and Financial Surveys (Washington: International Monetary Fund, September 1992).

    • Search Google Scholar
    • Export Citation

  • Goldstein, Morris, David Folkerts-Landau, (1993a), International Capital Markets: Part I Exchange Rate Management and International Capital Flows, World Economic and Financial Surveys (Washington: International Monetary Fund, April 1993).

    • Search Google Scholar
    • Export Citation

  • Goldstein, Morris, David Folkerts-Landau, (1993b), International Capital Markets: Part II. Systemic Issues in International Finance, World Economic and Financial Surveys (Washington: International Monetary Fund, August 1993).

    • Search Google Scholar
    • Export Citation

  • Goulder, Lawrence H., “Implications of Introducing U.S. Withholding Taxes on Foreigners’ Interest Income,” in Tax Policy and the Economy, ed. by Lawrence H. Summers (Cambridge, Massachusetts: MIT Press, 1990).

    • Search Google Scholar
    • Export Citation

  • Greenspan, Alan, Testimony Before the Committee on Banking, Housing, and Urban Affairs, U.S. Senate (Washington, March 2, 1994).

  • Howell, Michael J., Angela Cozzini, and Luci Greenwood, Cross-Border Capital Flows: A Study of Foreign Equity Investment—1991/92 Review (London: Baring Securities, 1993).

    • Search Google Scholar
    • Export Citation

  • Howell, Michael J., Angela Cozzini, and Luci Greenwood, North America, Europe, Africa, Pacific Rim, and Latin America: Global and Emerging Markets Strategy (London: Baring Securities, April 1994).

    • Search Google Scholar
    • Export Citation

  • Hu, Yebi, China’s Capital Market (Hong Kong: Chinese University Press, 1993).

  • Huizinga, Harry, “Withholding Taxes and the Cost of Public Debt,” IMF Working Paper No. 94/18 (Washington: International Monetary Fund, February 1994).

    • Search Google Scholar
    • Export Citation

  • International Finance Corporation, Quarterly Review of Emerging Markets, Fourth Quarter 1993 (Washington: International Finance Corporation, 1994).

    • Search Google Scholar
    • Export Citation

  • International Monetary Fund, World Economic Outlook, October 1993: A Survey by the Staff of the International Monetary Fund, World Economic and Financial Surveys (Washington: International Monetary Fund, October 1993).

    • Search Google Scholar
    • Export Citation

  • International Monetary Fund, World Economic Outlook, April 1994: A Survey by the Staff of the International Monetary Fund, World Economic and Financial Surveys (Washington: International Monetary Fund, April 1994).

    • Search Google Scholar
    • Export Citation

  • Investment Company Institute, Mutual Fund Factbook (Washington: Investment Company Institute, 33rd ed., 1993).

  • Kleiman International Consultants, Fifth Annual Emerging Stock Markets Survey (November 1993).

  • Kroszner, Randall S., “An International Comparison of Primary Government Securities Markets,” in U.S. Securities Market: The Scholars’ Assessment, Vol. 1 (Chicago: MidAmerica Institute, April 1993).

    • Search Google Scholar
    • Export Citation

  • Loser, Claudio, and Eliot Kalter, eds., Mexico: The Strategy to Achieve Sustained Economic Growth, IMF Occasional Paper No. 99 (Washington: International Monetary Fund, September 1992).

    • Search Google Scholar
    • Export Citation

  • McDonough, William J., Remarks Before the Sixty-Sixth Annual Mid-Winter Meeting of the New York State Bankers Association (March 1994).

  • Miller, Merton, “Functional Regulation,” Address at the Fifth Annual Pacific-Basin Capital Markets Finance Conference (Kuala Lumpur, June 1993).

    • Search Google Scholar
    • Export Citation

  • Moody’s Bond Survey, March 7, 1994.

  • Mullins, David W., Jr., Statement Before the Subcommittee on Securities of the Committee on Banking, Housing, and Urban Affairs, U.S. Senate (Washington, January 23, 1992); reproduced in Board of Governors of the Federal Reserve System, Federal Reserve Bulletin, Vol 78, No. 3 (March 1992), pp. 195–98.

    • Search Google Scholar
    • Export Citation

  • Mussa, Michael, and Morris Goldstein, “The Integration of World Capital Markets,” in Changing Capital Markets: Implications for Monetary Policy (Kansas City: Federal Reserve Bank of Kansas City, 1993).

    • Search Google Scholar
    • Export Citation

  • Pensions and Investments, January 25, 1993.

  • Pensions and Investments, January 24, 1994.

  • Pirrong, S. Craig, “The Market for Treasury Derivative Securities: Microstructure and Market Power,” in U.S. Treasury Securities Market: The Scholars’ Assessment, Vol. 1 (Chicago: MidAmerica Institute, April 1993).

    • Search Google Scholar
    • Export Citation

  • Quale, Andrew, “Legal Primer for Non-U.S. Issuers Raising Financing in the United States,” paper presented at the Second Annual Conference on Emerging Market High-Yield Sovereign and Corporate Bonds (New York, January 22, 1993).

    • Search Google Scholar
    • Export Citation

  • Republic New York Securities, Year End Hedge Fund Review 1993 (1994).

  • Rodriguez, Anthony P., “Government Securities Investments of Commercial Banks,” Quarterly Review, Federal Reserve Bank of New York (Summer 1993), pp. 39–53.

    • Search Google Scholar
    • Export Citation

  • Sargent, Thomas J., “Fact or Fiction: Shortening Debt Maturity Lowers Interest Costs” (Chicago: Catalyst Institute, December 1993).