Developing bond markets and enhancing the capacity of public and private sector borrowers to issue long-dated, domestic-currency-denominated debt securities is high on the policy agenda in several emerging market countries.[1] Crises in Mexico in 1995 and Argentina in 2001 and in various East Asian countries in 1997–1998 demonstrated the danger of disruptions to the supply of bank credit when other sources of finance are underdeveloped. The short tenor of bank loans, itself a consequence of the tendency for banks to fund themselves with demandable debt, meant that even where financial institutions continued to operate, borrowers finding themselves unable to roll over their maturing obligations might experience a credit crunch. Access to bond markets came to be seen as an essential “spare tire” (Greenspan, 1999).
Yet the development of a well-functioning bond market presupposes extensive infrastructure, including well-developed accounting, legal, and regulatory systems, payments and settlements systems, rating agencies, and networks of brokers to sell bonds. It requires rigorous disclosure standards and effective governance of corporations issuing publicly traded debt securities. It presumes the existence of well-established companies whose operations and credit standing are well known and that are large enough to defray the non-negligible fixed costs of placing a bond issue. These are not preconditions that can be fulfilled overnight. Rather, they are by-products of the larger process of economic and financial development, which is why even in the advanced countries, bond markets historically have been late to develop.[2] As long as some of these developmental preconditions remain unfulfilled, borrowers may prefer to tap the more extensive and efficient bond market infrastructure that exists in the major financial centers. Or they may find it easier to borrow from banks, which rely on long-term relationships with their clients to obtain information and enforce repayment, thereby enabling them to circumvent imperfections in the information and contracting environments.
At the time of the crises that hit several emerging market countries in the 1990s, bond markets of sorts already existed, of course. But even in good times, the ability to issue in local markets was limited to large, well-known entities, and during crises they provided little relief. The yield on new issues skyrocketed. Worse still, market access and liquidity evaporated just when they were needed most. Low secondary market liquidity prevented investors from rebalancing their portfolios, and the illiquidity of the secondary market, which depressed retail demand, in turn limited the ability of potential issuers to place bonds on the primary market. In addition, the appetite of foreign investors seemed to be limited to issues denominated in dollars or other hard currencies, a fact that created further difficulties in bad times when the exchange rate had a tendency to depreciate. This hardly seemed like even a functional spare tire at a time when market conditions seemed to demand a set of high-performance all-weather radials.
Bond markets are not important only in times of crisis. Even during tranquil times, a well-working corporate bond market can help firms to obtain long-term finance and lower their financing costs and can enhance overall microeconomic efficiency. Developing a domestic bond market is likely to be particularly important for meeting the credit needs of small and medium-sized enterprises (SMEs). While global underwriters are interested only in relatively large issues, which are beyond the means of smaller borrowers, local markets are better positioned to acquire and process the information needed to evaluate the creditworthiness of SMEs.
The conclusion drawn from the 1990s crises by policymakers in several emerging market countries was that drastic action was needed to enhance the access of governments and, in particular, private corporations to bond finance. The Financial Stability Forum, the World Bank, the Inter-American Development Bank, the Asian Development Bank, and the Organisation for Economic Co-operation and Development all studied what emerging markets could do to develop local markets. Their key recommendations were for emerging markets to strengthen macroeconomic policies to provide a stable setting for both borrowing and lending, to improve corporate governance to ensure that firms would borrow prudently, to strengthen financial disclosure requirements to enhance the ability of potential bondholders to make prudent investment decisions, to encourage the growth of institutional investors to enhance diversification opportunities and reduce transaction costs, and to solidify bond market infrastructure generally by creating efficient clearing and settlement, credit enhancement, and custodial facilities.
Emerging markets have made important efforts in these directions. Yet the results have been disappointing. Domestic bond markets have grown only slowly. Liquidity is scarce, and turnover rates remain low. Notwithstanding the commentary surrounding the recent surge of funds into emerging markets, foreign participation in local bond markets, corporate bond markets in particular, remains quantitatively limited (with a few prominent exceptions like Mexico).
These disappointing results are rationalized in two (not necessarily incompatible) ways. First, institutions and policies remain weaker in emerging markets than in advanced countries, and there is no quick fix to this problem. Like it or not, eliminating the institutional and policy deficiencies slowing the development of deeper and more liquid bond markets is a difficult and time-consuming process. Second, emerging market countries seeking to develop their bond markets are handicapped by the small size of both their firms and their economies. Market depth and liquidity require a certain minimum efficient scale which is particularly hard to achieve in small countries. This country size effect is amplified by the fact that most emerging market countries tend to have small firms which cannot afford the fixed costs linked to issuing bonds. Related to this is the fact that emerging markets are not first movers in the competition for global market share; there already exist deep and liquid markets in the leading global financial centers. From the point of view of liquidity and transaction costs, it is therefore more attractive for issuers and investors from emerging markets to transact in the major global markets than it is for foreign investors to transact in emerging markets.
The Evolution of the Domestic Bond Market
There are different ways to describe the evolution of the Latin American domestic bond market. One way is to scale outstanding bonds by GDP (Figure 7.1). By this measure, the advanced economies have the largest bond markets, followed by East Asia, Latin America, and Eastern Europe and Central Asia. While the Latin American bond market is not much smaller than that of East Asia (35 versus 45 percent of GDP), there are large differences in composition. Latin American local markets are heavily skewed toward government bonds, while in East Asia fully half the outstanding stock is made up of issues of financial institutions and corporations.
This situation looks different if bond market capitalization is instead scaled by the size of the domestic financial system (proxied by M2) (Figure 7.2). In Latin America, evidently, it is financial sectors generally and not merely bond markets that are underdeveloped. This suggests that the growth of bond markets should be viewed as an organic part of the process of financial market development, and that countries will develop deep and liquid markets in debt securities only once they have succeeded in reducing the larger obstacles to financial development. Indeed, there are good reasons to think that banking systems and bond markets develop together, as they have 
prerequisites in common. In both cases confidence requires a reasonable level of information disclosure. In turn, mandating disclosure may require regulation by a supervisory agency or securities commission. The development of both a bond market and a sound banking system requires strong creditor rights and an effective system of corporate governance so that small creditors can be assured of being dealt with fairly. In both cases, confidence may require macroeconomic stability, so that depositors and investors do not fear that the value of their claims will be inflated away, and strong creditor rights, so that they are confident that they will be treated fairly in the event of a debt or banking crisis. There are in fact strong complementarities between bond finance and bank finance, and bond market development should not be seen as an alternative to the development of an efficient banking system but rather as part of the same organic process (Box 7.1).
The sequencing involved in this view is rather different from the traditional one in which bank finance develops first because the information and contracting environments are highly imperfect. In such a setting, according to the traditional model, banks in long-term relationships with their clients have a comparative advantage in bridging information gaps, enforcing repayment, and reorganizing problem loans. Bond markets develop only later, once an economy has acquired strong institutions of information disclosure, corporate governance, insolvency reorganization, and so forth. Recent research (e.g., Rajan and Zingales, 2003b) suggests that the actual sequencing of external finance, starting with banks and moving to bond markets and finally equity markets, is not so clear-cut in reality. It differs in different times and places. While not denying the special role of banks in the kind of imperfect information and contracting environment that is characteristic of many emerging markets, the perspective presented here suggests that the development of banking systems does not just precede the development of bond markets; rather, the two are complementary processes.
Clearly, there is also the danger that an imperfectly competitive banking system, in which financial institutions use their incumbency advantage and market power to slow the development of securitization and disintermediation, can retard the growth of the bond market. It may do so by limiting access to the payment system and supporting the maintenance of regulations that increase the cost of underwriting and issuance.[3] The actual situation on the ground appears to vary considerably. The IMF (2002b) observes that banks in Thailand have been able to place barriers in the way of bond issuance in an attempt to limit competition from the bond market. In Chile, the Latin American country with the most active corporate bond market, fully 26 investment banks have been active in underwriting and helping to place domestic debt securities. But it is an exception to the rule. Whereas 20 different commercial and investment banks act as lead underwriters in Brazil, three of them account for 90 percent of issues. The case is similar in Mexico, where three large banks dominate the underwriting and selling side of the market. In a number of East Asian countries, a handful of underwriters similarly dominate the market.
Not only do the Latin American bond markets tend to be small, but they also lag along a number of other dimensions, not just when compared with those in the advanced economies, but even when assessed relative to those in other emerging economies. For instance, the duration of issues on Latin American markets remains relatively short. The 
region has made some progress here, but in terms of, say, the share of bonds with a residual maturity of less than one year, it still compares unfavorably with emerging East Asia, much less with the advanced economies. The majority of issues on Latin American markets have floating rates, and investors demand that interest rates be indexed to inflation or the exchange rate, in contrast to emerging East Asia, where fixed rates are the norm and indexation is virtually nonexistent. About 80 percent of all bonds issued in East Asia between 2000 and 2005 (weighted by value) had a maturity longer than one year and no indexation, whereas the comparable figure for Latin America was less than 10 percent (Figure 7.3). With the exception of a few benchmark issues, turnover rates remain relatively low in Latin America, leaving markets relatively illiquid. And regional markets are still disproportionately dominated by government bonds.
The Development of the Domestic Bond Market:
What Do the Data Say?
One way of systematically comparing the determinants of bond market development is to use aggregate data for a cross-section of countries and series of years to estimate the determinants of bond market capitalization. Borensztein, Eichengreen, and Panizza (2006a) use annual data for 43 countries over the 1995–2004 period and show that a limited number of observable policy variables and country characteristics explain some 70 percent of the difference in bond market capitalization between Latin America and the advanced economies. This same handful of observable variables also explain 90 percent of the difference between the two in the development of the market for the bonds of corporations and financial institutions. This means that if it were possible to take these country characteristics and replace their average values for Latin America with their average values for the advanced economies, the two regions would have private bond markets of similar size.
In the real world, of course, growing bond markets is not so easy. Improvements in policies and institutions take time to work their effects. In addition, the statistical analysis shows that one-quarter of the difference in bond market capitalization between advanced economies and Latin America is due to small country size and the general level of economic development. Another 15 percent of the difference is related to geography and historical factors like the origin of the legal code. Another 15 percent is attributable to the underdevelopment of the financial system, something that again takes time to correct. More easily manipulated policy variables like the exchange rate regime, the presence or lack of capital controls, the level of public debt, bank concentration, and banking spreads are all statistically significant in the empirical analysis but play a smaller role in explaining the difference between the development of the bond markets of advanced economies and that of the bond markets of Latin America. This should not be taken to mean that policies and institutions do not matter (in fact about 20 percent of the difference in the size of bond markets in Latin America and the advanced economies is due to macroeconomic stability, investor protection, and lower cost of enforcing contracts), but it is a reminder that there are no shortcuts. The same policies that are necessary for economic development in general are also necessary for the development of domestic bond markets.
Size Does Matter
As there are significant fixed costs associated with underwriting, publicizing, and distributing a bond issue, and as secondary market liquidity will be greater where there exist a large number of bonds and bondholders, potential market size is an important factor limiting the issue of bonds. This provides a rationale for the traditional Latin American strategy of also relying on international markets, where the fixed costs of issuance are lower and liquidity is higher. At the same time, it provides an obvious rationale for Asian efforts to integrate national markets into something resembling a single regional market, so that the size of the regional economy rather than that of the national economies is the constraint on market growth (Boxes 7.2 and 7.3).
If perceived lack of demand (perhaps because of low saving rates) is what is holding Latin America back from developing bond markets, one possible strategy is to encourage the participation of investors from outside the region. However, foreign investors are most inclined to take positions in countries with larger bond markets (Brazil, for example)—where the costs of closing out positions are least—that is to say, where liquidity is already the greatest. Brazilian authorities have sought to capitalize on this interest by retiring foreign debt from the market and replacing it with domestic currency (interest-rate- and inflation-indexed) issues. Mexico, where foreign participants are reported to hold more than 50 percent of the government’s 10-year bonds and more than 80 percent of its 20-year bonds, has sought to take advantage of foreign participation by issuing exclusively on the domestic market.
To be sure, there is also a foreign demand for “exotics,” or the less-liquid bonds of smaller countries, but this phenomenon is quantitatively limited; for most investors, the limited liquidity of exotics, together with the lack of hedging instruments and the fixed costs of obtaining information about issue quality, currency risk, withholding tax regimes, and so on in smaller markets, limits foreign demand.
This raises the possibility that the globalization of bond markets, and the growing participation of foreign investors in Latin America’s local markets in particular, may be encouraging a bifurcation between the region’s larger and smaller markets by further enhancing the already greater liquidity of the larger markets while having little discernible impact on their smaller counterparts. Similarly, it may be enhancing the liquidity of government bond markets relative to corporate bond markets. This may encourage smaller countries in the region to borrow by issuing global bonds in extraregional financial centers as an alternative to developing their domestic markets. But that in turn may further limit the development and liquidity of local markets and further discourage foreign participation.[4] On the other hand, one can argue that international issues are useful for familiarizing foreign investors with a country’s situation and its debt instruments and that domestic and international issues are complements rather than substitutes.
The Market for Domestic Government Bonds
Simple averages based on a sample of 18 Latin American and Caribbean countries show that over the 1990–1994 period, domestic government bonds were around 16 percent of GDP, ranging between 2 percent of GDP in Bolivia and 74 percent of GDP in Chile (Table 7.1). By 2000–2004, the share of domestic bonds in GDP had grown to 24 percent, ranging from 6 percent in Guatemala to 78 percent in Jamaica. While there are 14 countries in which the share of domestic debt increased by more than five percentage points between 1990–1994 and 2000–2004, there are only two countries with a substantially decreasing share of domestic government bonds in GDP: Chile and Trinidad and Tobago.[5] Note that this trend is not unique to Latin America; the share of domestic bonds in GDP also increased in emerging economies elsewhere over the same time period and in the advanced economies.
It is not clear, however, whether the trends documented above are due to the behavior of total public debt or to its composition. Simple averages for all countries for which data are available show that the share of domestic bonds in total public debt has increased substantially (going from 26 percent in 1990–1994 to 36 percent in 2000–2004), suggesting that the increase in the amount of domestic financing is due to a composition effect (i.e., the amount of domestic government bonds has grown more than total public debt), an observation consistent with the experience of East Asia, the other emerging markets, and the advanced economies. Here, however, Latin American countries can be divided into two groups. In the first group (which includes Argentina, Belize, Paraguay, and Uruguay), the share of government bonds in total debt has decreased, indicating that the increase in domestic government bonds (or the constant level, in the case of Belize) is due to higher levels of debt and not to a shift towards increased domestic debt. In the second group of countries, indeed, there is an increase in the share of domestic government bonds in total public debt, a sign that these countries have adopted a strategy aimed at developing the domestic market and reducing their reliance on international capital markets.
Interestingly, in the early 1990s several Latin American countries had small and shrinking domestic government bond markets, but the Tequila crisis that hit the region in early 1995 played a key role in the development of domestic bond markets (Cowan and Panizza, 2006).[6] The financial turmoil that followed the postcrisis Mexican devaluation deprived several countries of access to the international capital market and forced them to rely on the domestic market.[7] Under normal circumstances, countries probably would have quickly reverted to international borrowing, but the series of financial crises that followed the one originating in Mexico led to discontinuous access to the international capital market and convinced policymakers of the importance of developing a reliable domestic source of financing.
In fact, a cursory look at the development of Latin American government bond markets illustrates that a crisis, either domestic or international, is often the event that kick-starts a country’s domestic bond market.[8] The most striking case is Chile, where the large amount of outstanding government bonds (mostly issued by the central bank) is the legacy of the banking crisis that hit the country in the early 1980s.[9] Mexico and Uruguay started issuing domestic government bonds after the debt crisis of early 1982 prevented them from accessing the international capital market. In the case of Argentina, the government bond market started in 1990–1991, when the government issued several bonds to consolidate central bank debt to the commercial banks and to consolidate existing liabilities in regard to pensioners, suppliers, and victims of the country’s military regime. Most of these earlier issues were thus compulsory. Debt and domestic bonds became less important as soon as the Argentine government gained access to the international capital market after successfully restructuring its defaulted international bank loans (Fernández et al., 2006).
Besides changes in levels, there have also been changes in the composition of domestic debt in Latin America and the Caribbean. Focusing on the 1997–2004 period, there are five countries (Argentina, The Bahamas, Brazil, Mexico, and Venezuela) that substantially reduced the share of their domestic debt indexed to foreign currency by substituting for it either domestic currency debt or debt indexed to prices (Figure 7.4).[10] There are also six countries (Barbados, Bolivia, Costa Rica, Nicaragua, Peru, and Uruguay) with a more or less unchanged share of foreign currency debt over the period, and four countries (Chile, Colombia, Honduras, and Jamaica) with an increasing (but still small) share of foreign currency debt.
Focusing on maturity, there are four countries in the region (Argentina, Brazil, Colombia, and Mexico) that lengthened the maturity of their domestic debt over the same period (Figure 7.5 plots the share of domestic debt with maturity less than one year), four countries (Barbados, Chile, Nicaragua, and Uruguay) with a more or less unchanged maturity over the period, and one country (Peru) with an increasing share of short-term government bonds (and thus a shortening maturity).
Two problems with the measures of debt structure highlighted above is that they do not capture the share of debt indexed to the short-term interest rate and that they do not highlight possible trade-offs between currency and maturity risk, as countries that issue more debt in nonindexed domestic currency may have debt with shorter maturities (see Chapter 13 for a discussion of this trade-off).
Private Issuers
A survey covering six Latin American countries (Argentina, Brazil, Chile, Colombia, Mexico, and Uruguay) showed that five of these countries had no private domestic bond market whatsoever at the beginning of the 1990s (Cowan and Panizza, 2006).[11] This is despite the fact that regulatory reforms allowing or fostering bond issuance had been carried out in several of these countries during the 1980s.[12] Macroeconomic instability is partly to blame for this lack of development. Argentina and Brazil had hyperinflation episodes in the early 1990s, while in Mexico and Uruguay, inflation fell below 100 percent per year only as late as 1989 and 1992, respectively.[13] Chile, on the other hand, has enjoyed relative macroeconomic stability since the mid-1980s. In 1990, outstanding corporate bonds in Chile amounted to close to 5 percent of GDP. Colombia is something of an outlier along this dimension: despite broad macroeconomic stability in the 1980s, the private bond market failed to develop in the country.
Despite their similar starting points, private bond markets in these countries followed different paths during the 1990s (Figure 7.6). Issuance in Argentina started in earnest in 1991, following the reduction of inflation brought about by the convertibility plan and a tax reform that leveled the playing field between bank and bond finance, and continued until the recession of 1998. The market for Mexican bonds has grown more or less continuously since 1990 following stabilization of inflation in the late 1980s. New regulation introduced in 2001 has led to renewed growth in recent years. The Uruguayan bond market had a brief period of growth after 1994, until a series of corporate scandals led to reduced issuance and stagnation. Brazil experienced the fastest bond market growth among these countries, but this was concentrated in the years immediately following the reduction in inflation brought about by the Real Plan. Starting from basically nothing in 1990, the stock of private bonds in Brazil reached more than 10 percent of GDP by 1994 and fluctuated between 8 and 13 percent of GDP over the 1994–2004 period, then rose markedly, with the increase sparked by a new wave of issuance in 2005 (four times the level of 2004), with leasing companies being the most active market participants.
Once again Colombia is an outlier in the sample. Private bond market growth there has been slow and erratic, so that even by the end of the decade, outstanding private bonds remained below 1 percent of GDP. The evolution of the private bond market in Colombia stands in stark contrast to high levels of growth in the country’s public debt sector, leading to concerns that government debt may be crowding out the private market (Aguilar et al., 2006).
In Chile, the initial level of bond market capitalization was higher than in the other five countries, but the country experienced no bond market growth until the late 1990s. Several factors may explain the recent private bond market growth in Chile (Braun and Briones, 2005). The first relates to monetary policy—and in particular to the extreme spike in short-term interest rates brought about by the defense of the peso in 1998. This increase in short-term rates (and the associated credit crunch) made long-term nonbank financing increasingly attractive. A second and related explanation suggests that the closure of the international financial markets, combined with a large stock of institutional assets, jump-started the corporate bond market in Chile (Cifuentes, Desormeaux, and González, 2002). If the effects are permanent, then this would be another case in which a negative shock leads to the development of the domestic bond market. The third explanation suggests that the increase in private bonds was due to the market space vacated by falling Chilean public debt in this period.
Despite this recent growth, the stock of private debt in most of these countries remains low. In four of the six countries considered, private bond markets remained below 5 percent of GDP in 2005. The two exceptions are Brazil and Chile, in which outstanding private bonds were above 10 percent of GDP in 2005. In any case, these values are considerably lower than the averages of East Asia and the advanced economies (28 and 70 percent of GDP, respectively).
Besides differing in trends and levels, bond markets in the six countries covered also differ in the characteristics of the instruments issued (Table 7.2). Nominal debt is still rare in the region, with most countries issuing debt indexed to prices (Argentina and Chile), to the interest rate (Brazil), or to the dollar (Argentina and Uruguay). In this sense, the bond market mimics the maturity and indexation structure of other forms of corporate finance in the region: short and/or dollarized in the cases of Argentina, Brazil, and Uruguay (Kamil, 2004).
Furthermore, there does not appear to be a clear movement in the private sector towards “safer” forms of debt (“safer” from the perspective of the borrower, that is), such as debt indexed to prices and nominal debt. Mexico is the sole exception. In Mexico, the share of private nominal bonds has increased significantly in recent years, in line with developments in the government bond market. However, dollarization of private and government bonds has taken different tracks in Argentina. Despite falling shares of dollar-denominated government bonds, dollarization of private debt has remained high, even though new issuance and extensive rescheduling in 2002 and 2003 allowed ample opportunities for restructuring.
Where the six countries do coincide is in the type of firm that issues debt. Four out of the six case studies carry out a detailed firm-level analysis of the determinants of issuing bonds. Broadly speaking, the case studies find that large firms, with higher than average leverage, more tangible assets, and higher profitability, are more likely to issue bonds. A cursory look at the data shows that bond markets in Latin America are characterized by a small number of large firms issuing sizable bonds. This suggests that there are large fixed costs associated with bond issuance. The fact that many firms are “repeat issuers” (as suggested by the smaller number of firms than issues) points towards the existence of two forms of fixed costs: those related to becoming an issuer (disclosure costs, required accounting changes, etc.) and those related to each specific issuance (underwriting fees, etc.). These high issuance costs may be one explanation for the importance of alternative debt instruments in some of the countries considered, such as the checks of deferred payment that have become an increasingly common form of financing for firms in Argentina.
The importance of issuance costs is confirmed by the results of a set of firm-level surveys which show that a sizable fraction of firms that used to issue bonds but no longer do so identify “high emission costs” and “emission requirements” as the main reasons for no longer issuing bonds. Moreover, firms surveyed usually state that minimum size, information requirements, and lengthy procedures make bonds less attractive as a source of financing vis-à-vis bank financing. At the same time, bonds dominate banks in terms of maturity and interest rates.
The importance of market size in these case studies is consistent with the cross-section results of Borensztein, Eichengreen, and Panizza (2006a) and Eichengreen and Luengnaruemitchai (2004), who find that country size is one of the few variables that has a systematic effect on the size of the private bond market. Interestingly, Borensztein, Eichengreen, and Panizza (2006a) show that size matters when bond stocks are scaled both by GDP and by a measure of broad financial development, indicating that in larger countries bond markets not only are larger but are also relatively more important within the financial system.
The obvious question, therefore, is whether the fixed issuance and disclosure costs that make bonds attractive only to a small group of large firms are particularly high in the region. Domestic issuances are more than twice as expensive in Uruguay as in Mexico. More importantly, in both Brazil and Chile, issuance costs for debt placed offshore are lower than the domestic costs (Table 7.3). Considering the higher relative prices of nontradable goods in the United States, this suggests space for significant reductions in the average costs of issuance in Latin America. It is an open question whether the cost differences are due to the existence of fixed costs in market infrastructure (and should therefore diminish as bond markets expand) or are due to differences in regulation and financial market structure that would lead to higher costs for a given size of total issuance.
Do bond markets in the region provide an alternative to bank financing for firms in the region? Not always. For a start (and as discussed above) bond markets remain small in most countries surveyed in this chapter. Furthermore, in many countries, private bond issuance is dominated by financial sector firms for whom bonds provide an alternative to demand deposits, time deposits, CDs, etc. This is particularly true in Mexico and Uruguay, where more than 80 percent of private sector bonds are issued by the financial sector. The upside of this is that smaller firms can potentially tap longer-term credit through financial intermediaries. In Brazil, for instance, about 70 percent of the volume consists of issues by leasing companies and is associated with long-term financing through leasing of fixed assets.
Furthermore, even if expensive and underdeveloped, domestic markets are likely to be the only way in which small and medium-sized enterprises can access the bond market. A sample of 22 emerging market countries shows that the median issue is US$17 million on the domestic market but $100 million on the international market (Table 7.4). Data for a sample of six Latin American countries show an even more striking difference: the median value of domestic issuances is $22 million and that of international issuances is $175 million. A way of focusing on the segment of the market relevant to SMEs is to look at the lowest decile of the distribution. In this case, the mean and median are both roughly $1.5 million for domestic issues but about $5 million for international issues. Again, the difference is even more striking in Latin America, where mean and median values for domestic issuance are about $0.6 million but those for international issuance are well above $40 million. This is consistent with the notion that SMEs in a position to borrow and service only relatively small amounts of debt may be able to place issues on domestic markets even when they are locked out of international markets owing to their small scale (Borensztein, Eichengreen, and Panizza, 2006c).
Asset-backed securities are a fairly new family of financial instruments in Latin America, but their presence has been growing rapidly in the past two or three years, and they show interesting potential. The instruments include mainly securities that enable firms and financial institutions to securitize their receivables, mortgage-backed securities, and commercial paper. Several countries are creating the legal framework that permits the development of this type of instrument. In Brazil, the development of mortgage-backed securities (certificado de recebíveis imobiliários, or CRIs) and receivables investment funds (fundos de investimentos em direitos creditórios, or FIDCs), with impetus from the central bank and the securities and exchange commission, was a significant step in widening the market. Issues of these two instruments grew by 250 percent in 2005 and amounted to the equivalent of US$4.5 billion. Argentina is an interesting example in which asset-backed securities provided something of a spare tire for the ailing banking sector in recent years. The securitization of receivables (fideicomisos financieros), especially consumer credit and agricultural export revenues, and a form of commercial paper (cheques de pago diferido) grew strongly in recent years. The issue of fideicomisos, in particular, more than tripled in 2005, although the volume was still a fairly modest US$1.7 billion.
Asset-backed securities, if successfully implemented, may address two of the issues that hold back private bonds in Latin America. First, the legal and institutional requirements for exercising creditor rights are simpler for asset-backed securities because the assets are backed by collateral—such as property or a vehicle—and foreclosing should be less complicated than going through bankruptcy proceedings, as would be required in the case of unsecured corporate bonds. While effective and speedy recourse to the legal system still imposes requirements for institutions that may not be available everywhere, the bar is set lower for asset-backed securities. Second, the problem of small firm size can be circumvented by pooling a large number of firms in a structured security. Banks can use their superior expertise in selecting credits and can avoid carrying an excessive volume of correlated risk credits on their books. Especially in cases where banks have been very conservative in lending, asset-backed securities can be an effective instrument for increasing financial intermediation and investment in Latin America.
The international financial institutions can play a role in promoting securitization. Just to give one example, the IDB’s Private Sector Department has provided guarantees to the Brazilian Securities Mortgage Instruments Warehouse Facility and is planning to conduct further operations based on future cash flow securitization.
Interactions between the Government and Private Bond Markets
The presence of a liquid market for government bonds can benefit the corporate bond market in terms of providing the necessary infrastructure for trading, producing information about the future path of interest rates, and providing a benchmark curve. However, bigger is not always better, as the benefits related to the creation of pricing and hedging instruments can be eliminated if the government crowds out private borrowers (McCauley and Remolona, 2000).
Empirical analyses of the relative costs and benefits of having a large domestic government bond market yield mixed results. Eichengreen and Luengnaruemitchai (2004) find no significant impact of the size of the government bond market on the development of the private bond market and conclude that this may be because the benefits in terms of liquidity and market infrastructure balance the costs in terms of crowding out. It is not obvious, however, how this crowding out would work. While a higher level of government debt may crowd out private investment through its effect on interest rates, it is less clear why the method of financing should matter. In fact, crowding out would require the assumption that private investors are willing to allocate a fixed share of their portfolio to bonded debt.
Borensztein, Eichengreen, and Panizza (2006c) use bond-level data for a sample of 16 emerging market countries to determine whether the level of public debt versus its composition has a differential effect on the development of the private bond market. They find that, with total public debt controlled for, having a large domestic bond market is associated with longer maturity and lower spreads of corporate bonds and that, with the share of domestic government bonds controlled for, higher levels of total public debt are associated with shorter maturity and higher spreads of corporate bonds.[14] This suggests that crowding out comes from the level of public debt and not from its composition. A simple way to check the crowding out versus market development hypothesis is to compare the evolution of the composition of government debt with that of private debt. Figure 7.7 shows that the share of domestic government bonds in total public debt is positively correlated with the share of corporate bonds in total domestic credit to the private sector.[15] This finding goes in the same direction as the results of Borensztein, Eichengreen, and Panizza (2006c) in suggesting that, for a given level of public debt, the higher the share of domestic financing, the greater the development of the private bond market.[16]
A set of surveys aimed at determining whether institutional investors value the benefits of a larger government bond market found that investors in Chile, Colombia, Mexico, and Uruguay tended to agree with the idea that a large stock of public debt is important for the development of the corporate bond market, while investors in Brazil tended to disagree with this statement. There also seems to be agreement on the fact that having a yield curve is a crucial element for having the ability to price corporate bonds. The same surveys also found that government and corporate bonds seem to be competing in the portfolio of institutional investors located in Uruguay and to some extent Mexico, but this is not the case in Brazil, Chile, and Colombia (Cowan and Panizza, 2006).
What about spillover effects from the composition of public debt to that of private debt? As discussed above, there is no clear pattern emerging from the countries surveyed by Cowan and Panizza (2006). Whereas in Mexico, “safer” public bond debt has been accompanied by “safer” private bonds, this has not been the case in other countries. This is not surprising, as the relationship between private and public debt structure is complex. On the one hand, having a CPI-indexed or a nominal yield curve for public debt should make pricing of similar types of private debt easier. On the other hand, debt currency and maturity is the outcome of insurance transactions against various idiosyncratic risks (inflation, real exchange rate, real interest rate, etc.) between suppliers and demanders of credit. Changes in risk aversion or in the expectations regarding these shocks may be different in the public and private sectors. Take, for instance, the case of Argentina and assume that the reduction of foreign currency public debt came from an increased awareness in the public sector of the risks of this form of debt. This being the case, the government would then be willing to pay the higher interest charged on peso debt—which in the Argentine case specifically took the form of larger present value of payment in the negotiations in regard to defaulted debt. Nothing guarantees, however, that the private sector experiences the same shift in demand for safer debt. Indeed, seeing a government balance sheet with larger nominal debt may actually make investors more concerned about opportunistic government behavior and therefore demand more dollar debt contracts rather than fewer.
The Importance of Having the Right Plumbing
While there is an extensive literature that focuses on the relationship between market development and macroeconomic stability, much less is known about the role of market microstructure in emerging market countries (Madhavan, 2000). Economic theory often takes a view of financial markets as a black box that absorbs information and produces an optimal allocation of capital and pricing of risk. Microstructure theory tries to move away from this black box description and emphasizes that institutional features and trading mechanisms (often referred to as the plumbing of the market) are important determinants of market efficiency. According to this view, microstructure plays a key role in determining liquidity, efficiency, trading costs, and volatility (Glen, 1994). Politics also plays a role in determining microstructure, and if primary dealers manage to “capture” the state treasury, microstructure can become a source of rent and inefficiencies (Kroszner, 1998). This section briefly reviews some issues related to market microstructure, but it does not focus on issues related to the development of market infrastructure (such as depository and settlement facilities).[17] Furthermore, the section concentrates on the microstructure of the market for government bonds and hence does not discuss issues like corporate governance and the establishment of local credit-rating agencies, which are more relevant for the development of corporate bond markets.[18]
One way in which microstructure can affect market liquidity (i.e., the capacity for buying or selling without delay and with a minimum effect on prices) is through the decision on whether to trade through continuous or periodic auctions. The former system has the benefit of giving participants continuous access to the market, but the latter system has the benefit of bunching transactions together and hence increasing the depth of the market. Execution risk is often high in newly developed markets, which are often thin and illiquid, a situation that makes periodic markets preferable. Continuous auctions are preferable, on the other hand, in well-developed markets with a large number of participants. Debt managers can also improve liquidity in secondary markets by issuing a smaller number of security types and hence reducing the fragmentation of their own debt stock. Here as well, however, there are trade-offs. While having a small number of securities may be good for volatility purposes, it does reduce authorities’ ability to issue instruments with different types of indexation and maturity.
Microstructure can have an effect on market efficiency (i.e., the idea that prices incorporate all available information) in that it affects the type of information available to investors. Take, for instance, the decision on whether a given security should be traded on an organized exchange or intermediated by dealers in decentralized over-the-counter (OTC) markets. Organized exchanges have advantages in terms of transparency and price discovery and hence tend to be positively associated with market efficiency. However, OTC markets are less institutionally complex, can provide immediate liquidity under uncertain market conditions, are cheaper to manage (because they require less-demanding clearing and settlement systems), and, having lower fixed costs, may dominate organized exchanges when markets are small.19 While most countries trade government bonds on the OTC market, several Latin American countries also use organized exchanges (Table 7.5).
Price discovery on the primary market may also depend on the type of auction used to allocate government bonds. There are essentially two types of auctions used for this purpose: discriminatory price auctions (also called American or Yankee auctions) and uniform price auctions (also called Dutch auctions). In the first type of auction, securities are sold to each primary dealer at the bidding price quoted by that dealer. In a uniform price auction, in contrast, all successful bidders pay the same price (usually the lowest among those offered by the successful bidders). In theory, Yankee auctions should lead each dealer to pay its reservation price; in practice, though, dealers know that they might be subject to the “winner’s curse,” and this may induce them to underbid. In fact, there is some evidence that Dutch auctions may increase revenues (Bartolini and Cottarelli, 1997), and the U.S. Treasury has now moved to uniform price auctions. The problem is that Dutch auctions may also lead to collusion, especially in markets with a small number of primary dealers. In practice, Argentina, Chile, Colombia, and Peru use Dutch auctions, Brazil and Venezuela use Yankee auctions (Venezuela exclusively, Brazil predominantly), and Mexico uses both Dutch and Yankee auctions depending on the type of security (Table 7.5). Kroszner (1997) suggests that Dutch auctions are superior to Yankee auctions but that the latter increase the value of primary dealers’ privileged information. As a consequence, discriminatory price auctions often result in a hidden transfer to the primary dealers.
Governments can also increase the efficiency of the market for government bonds by disseminating pre-trade and post-trade information. One way to improve the transparency of the primary market is to publish an issue calendar and make information about the auction outcome public.20 Mohanty (2002) surveys 17 emerging market countries and shows that 14 of them (all in Mohanty’s survey except Colombia, the Philippines, and Peru) have an issue calendar.
One matter on which there is no agreement concerns the role of primary dealers. These are dealers designated by the authorities to be the intermediary between debt managers and investors. Primary dealers are usually granted bidding privileges and special credit lines but are required to guarantee that auctions are fully subscribed and to act as market makers in the secondary markets. While the presence of primary dealers can help in providing a regular source of liquidity and information for debt managers, some markets may be too small to warrant the participation of this type of agent (IMF and World Bank, 2003), and these agents may end up capturing the regulator and extracting excessive rents (Kroszner, 1997). There is less disagreement, on the other hand, on the importance of having a set of market makers and on the fact that these agents need to have instruments (such as forwards, futures, and swaps) that allow them to hedge against interest rate risk (IMF and World Bank, 2003). Another important element for market development is the presence of related markets. The existence of a repo (repurchase agreement through which dealers “lend out” securities that they have in their inventories) market, for instance, is key for the development of the market for government bonds because it allows dealers to maintain the high level of inventories that is necessary to make two-way quotes (IMF and World Bank, 2003).
Microstructure can also affect trading costs. There are essentially two types of costs related to trading a security: fixed costs (commissions and taxes) and the difference between the price received for a sale and that paid for a purchase (the bid-ask spread). In markets with a small number of market makers, bid-ask spreads will tend to be high for at least two reasons. The first relates to the substantial risk absorbed by this small number of agents taking positions for the whole market. The second relates to the fact that the low level of competition is likely to generate monopoly rents. While one might think that a system with a large number of market makers (or an organized exchange with no market makers) should be preferable in terms of having lower spreads, there is a possible trade-off, because once they are deprived of their monopoly rents, some agents may decide to stop acting as market makers, and this could lead to liquidity problems (especially in small markets). Hence, market makers should not extract too much rent from their role, but authorities should recognize that market making entails both liquidity and interest rate risks, and hence they need to make sure that this activity is profitable for the intermediaries. Brazil, Chile, and Mexico have tight bid-ask spreads (close to those prevailing in the United States and in other emerging market countries with low spreads), but the bid-ask spreads in Argentina, Colombia, and Venezuela are wider (substantially so in the case of the latter two countries), as shown in Table 7.5. It is important to note that, besides microstructure, one of the key determinants of bid-ask spreads is market size (Figure 7.8).
Finally, microstructure is related to market volatility (i.e., the size and frequency of price changes that do not reflect changes in the fundamental value of an asset). Circuit breakers and price limits that interrupt trading whenever prices exceed a given threshold are the most common mechanisms for limiting volatility. The benefit of such mechanisms is that they force investors to have a “cooling-off” period and allow them to absorb new information and price the security at its fundamental value. The problem is that these mechanisms reduce liquidity and may slow the convergence of an asset to its fundamental value.
While the above discussion has clarified the importance of market microstructure, it has also highlighted that there are no one-size-fits-all solutions and has indicated that systems that may work well in the presence of large markets may not work as well in the small markets that characterize a number of Latin American and Caribbean countries. Furthermore, the ideal microstructure of markets characterized by a large number of retail investors is likely to be different from that of markets characterized by relatively small numbers of sophisticated institutional investors.
Summing Up
This chapter has shown that Latin America tends to have relatively well-developed markets for government bonds but extremely small private bond markets. The question is whether this contrast is likely to be short lived or enduring. If the problem leading to small private bond markets in Latin America is that years of budget deficits have led to excessive government bond issuance that has crowded out private bond issuance, then many years of primary fiscal surpluses may have to pass before the “overhang” of government bonds is worked down. If it is that Latin America’s history of macroeconomic and financial instability limits investors’ demand to debt securities with interest rates indexed to inflation or the exchange rate, then many years may have to pass before stronger policies leading to reduced volatility will produce a demand for longer-term issues. If perceptions of imperfect corporate governance and unreliable contract enforcement currently render investors reluctant to hold corporate bonds at any price, then some time may have to pass before the relevant reforms begin to create a significant demand. If in smaller Latin American and Caribbean countries the local market’s lack of scale is the obstacle to spreading the fixed cost of an issue and enhancing secondary market liquidity, then reasonable questions can be raised as to whether this obstacle can ever be overcome. Or maybe these estimates are overstated; maybe the relevant reforms will succeed in producing deeper and more liquid bond markets in short order. In sum, the question is: how long will it take for Latin America and the Caribbean to develop deep and liquid bond markets?
