Comparative Economic Studies (2014) 56, 200–228. doi:10.1057/ces.2014.4; published online 13 March 2014

Debt Markets in Emerging Economies: Major Trends

Tatiana Didier 1 and Sergio L Schmukler 2

Abstract

G00; G20; G21; G23

INTRODUCTION

The experiences of emerging economies with recurrent currency, debt, and banking crises, particularly during the 1980s and 1990s, highlighted the dangers that poor macroeconomic fundamentals and balance sheets pose for open countries in a context of globalized financial systems. Moreover, the uncertainty across emerging markets resulting from macroeconomic volatility–especially high and unpredictable inflation in a number of countries–was deleterious to debt market development, most of all at the longer maturities. It corroded the role of money as a store of value, leading to a gradual build-up of currency and duration mismatches. The inflexible exchange rate regimes, adopted in part to bring down inflation expectations, often ended up increasing real exchange rates, exacerbating interest rate volatility and currency mismatches, and making countries more prone to self-fulfilling attacks.

By the late 1990s, the prospects for financial sector improvements across emerging economies were somewhat pessimistic given the difficulty of overcoming high systemic risk and volatility, the slow progress in overall financial development, and the large mismatches in currencies and maturities. Most of these developments were the result of inherent deficiencies in emerging economies ( de la Torre and Schmukler 2004. 2006 ; de la Torre et al.. 2007 ). A number of economists shared this pessimism, focusing on the metaphor of ‘original sin’ in emerging economies – that is, the inability to issue long-term debt in their own currencies – as well as on outright dollarization and ‘sudden stops’ that would subject the economies to frequent shutdowns of foreign financing ( Eichengreen and Hausmann, 1999 ; Hausmann et al.. 1999 ; Calvo and Reinhart, 2000 ; Hausmann and Panizza, 2003 ).

These experiences nonetheless shaped the reform agenda of the late 1990s and 2000s. Prudent macroeconomic and financial policies to foster growth, stability, and resilience were implemented. The goal was to adopt well-regarded international standards and to reduce currency and maturity mismatches on the balance sheets of the public and private sectors. At the same time that they withdrew the state from the markets through fiscal reforms aimed at reducing borrowing and thus avoided crowding out, many emerging economies undertook significant efforts to expand the scope and depth of their financial systems. The idea was that financial development is not only linked to faster growth and greater welfare, but deeper financial systems are usually perceived as more resilient to shocks and less prone to volatility and financial crises. 1

New data from the mid- to late-2000s and several anecdotal accounts suggest some reasons for optimism. Emerging economies have improved their macroeconomic performance, lowered inflation, and reduced fiscal deficits ( Gourinchas and Obstfeld, 2011 ). These policy achievements, together with high liquidity in international markets, have allowed many emerging economies to issue long-term bonds in domestic markets, as foreign investors have expected further appreciations of local currency and entered local markets in search of higher yields. In addition, these economies weathered the storms of the global financial crisis relatively well, showing strength and resilience in their macro-financial sectors ( Eichengreen, 2009 ; Didier et al.. 2012 ). Indeed, in a break with history, most countries across the emerging world avoided domestic financial crises even as financial systems in the G-7 spiraled down into near collapse, averted only by large and unprecedented government bailouts.

The main goal of this paper is to document the major trends in the development of debt markets in emerging economies during the 1990s and the 2000s. We cover both bank and bond financing. The primary value of this exercise is to put in perspective the absolute and relative size and the evolution of different components of the financial system using both traditional and new indicators. While we focus on the borrowers’ (firms and government) side, we also provide some evidence on the savers’ (households) side indirectly through the size and behavior of institutional investors that channel their savings. In particular, we document the evolution of the main financial intermediaries aside from banks: pension funds, mutual funds, and insurance companies. We also investigate how the nature of financial activity (currency, maturity, and scope of credit) has developed and to what degree changes in the size of markets have implied greater availability of financing for corporations. Because it is very difficult to evaluate the extent of financial development given the lack of clear benchmarks, we provide comparisons over time and across regions relative to gross domestic product (GDP) and across different measures of market size and activity. For completeness, parts of the paper also include figures for equity markets.

The focus of this paper is on emerging economies, which experienced several crises in the past and made efforts to transform their financial systems. Moreover, these markets attracted significant attention from investors, practitioners, and academics over the last two decades because of their rapid growth, high rates of return, and continuous changes. We analyze seven of the largest countries within three geographical regions: Asia (the East Asian countries Indonesia, the Republic of Korea, Malaysia, the Philippines, and Thailand; plus China and India, which are shown separately because of their distinct natures), Eastern Europe (Croatia, the Czech Republic, Hungary, Lithuania, Poland, the Russian Federation, and Turkey), and Latin America (Argentina, Brazil, Chile, Colombia, Mexico, Peru, and Uruguay). 2 We compare the patterns observed in these countries with those in developed regions. Among developed countries. we consider the G-7 countries (Canada, France, Germany, Italy, Japan, the United Kingdom, and the United States) as well as other advanced economies that are typically regarded as being somewhat more similar to emerging markets (Australia, Finland, Israel, New Zealand, Norway, Spain, and Sweden).

We assemble a comprehensive dataset covering a number of dimensions of debt markets, such as, depth, currency, maturity, borrower composition, concentration, and measures related to the breadth of markets. For the depth measures, we use widely available data since the early 1980s from the World Bank’s World Development Indicators, the IMF’s International Financial Statistics, and the Bank for International Settlements. For the other dimensions of debt markets, we also use a number of other sources widely explored in the literature but with a more limited time span. For example, the analysis of banking systems explores data from local central banks and Bankscope, whereas for bond markets for the public sector we use data from local central banks. For bond markets for the private sector, we analyze capital raising activity available from the Thomson Reuters SDC Platinum database, which provides transaction-level information on new issues of publicly and privately placed bonds with an original maturity of more than 1 year. Given that this database does not collect data on debt issues with maturities shorter than 1 year, the dataset does not cover commercial paper. The coverage of the data varies and increases over time. Much more data, especially detailed, are available for the 2000s than for the 1990s, or even the 1980s. 3

The main findings in this paper provide a mixed, nuanced picture of the main trends in the development of debt markets and can be summarized as follows. Since the wave of financial crises that swept through the emerging world in the late-1990s and early 2000s, financial systems in emerging economies have effectively developed, becoming in many respects and by several standard measures deeper and more complex. Capital markets have increased in absolute and relative sizes, suggesting a mild but steady transition from a mostly bank-based model to one that is more complete and interconnected. This entailed the growth of not only bond markets but also equity markets. Although the size of banking systems has increased in most emerging economies (albeit from typically low bases), capital markets and nonbank institutional investors now play a more central role. Moreover, the number and sophistication of the participants have been expanding, even without taking into account the additional increasing participation of cross-border investors.

Importantly, the nature of financing is also changing, in general toward reducing mismatches, but at a slow pace. For example, the private sector has seen an expansion in local currency bond financing, the extent of dollarization of loans and bonds has declined, and the maturity of public and private sector bonds has typically increased. However, not all regions have moved in the same direction. Eastern Europe for instance increased its foreign currency debt before the global financial crisis, which was associated with the higher transmission of the crisis to the countries in that region. Despite these new developments, emerging countries’ financial systems still remain in many aspects underdeveloped in comparison to those in developed countries. For instance, except in some cases mostly in Asia, liquidity in secondary bond markets has been declining. And the public sector still captures a significant share of bond markets.

The rest of the paper is organized as follows. The next section documents and gives a broad overview of where emerging economies stand on commonly used and simple measures of depth of debt markets. The section after that analyzes whether and how the nature of financing has changed over time. The subsequent section examines the nonbank institutional investors. The last section concludes by discussing some of the challenges ahead for the development of debt markets in emerging economies.

DEPTH OF DEBT MARKETS

We start by providing some basic stylized facts showing where emerging economies stand on commonly used broad indicators of debt market development, making comparisons among themselves and with developed countries over the past three decades. More specifically, we focus on the depth of debt markets, analyzing the size of the banking system and bond markets, and contrasting it to the depth of equity markets as measured by market capitalization.

Overall, we observe that financial systems in emerging countries have developed significantly over the past two decades, indicating a mild but steady transition from an ‘old’, mostly bank-based model to a ‘new’, more complex and interconnected model in which capital markets play a more central role. Despite these new developments, emerging countries’ financial systems still remain in many aspects underdeveloped in comparison to those in developed countries.

Regarding the banking system, total banking claims as a share of GDP has expanded in most emerging economies, albeit from typically low bases. Asian economies (our sample of East Asian economies plus China and India) stand ahead of other emerging countries not only in absolute size but also in growth over the past 30 years, whereas Latin American countries stand at the other end of the spectrum lagging behind other developing countries. For example, the banking systems expanded on average across countries in Asia by 67 % between 1980–1989 and 2000–2009. Total bank assets across Eastern European countries also increased significantly, almost 50 % on average. In contrast, those in Latin America increased by 14 % over the same period. These changes are computed by calculating the change across decades for each country and then averaging across countries within each region. Figure 1 shows the patterns at the country level for the different regions.

Total bank assets. (a) G7 and other advanced economies; (b) Asia; (c) Eastern Europe and Latin America.

Notes. This figure shows the average annual ratios of banking claims to GDP between 1980 and 2009 for selected countries. The statistics for China in the 1980–1989 period include only banking claims to the private sector. The data sources are the IMF’s International Financial Statistics (IFS) and the World Bank’s World Development Indicators (WDI).

Full figure and legend (123K )

The banking sector in developed countries was deeper to start with and has typically expanded faster than the banking sectors in many emerging economies over the past three decades. For example, bank size increased by almost 70 %. growing from 79 % to 107 % of GDP on average across developed countries between 1980–1989 and 2000–2009. Nonetheless, there is some heterogeneity across the developed world, where the banking system in some countries remained stagnated.

The patterns of bond financial development differ across countries over the past two decades, among both developed and developing countries. On average and in percentage change terms, bond markets have grown significantly in developing economies, but far less in developed countries. For example, bond market capitalization in Asia and Latin America grew about 140 % on average in the 2000s relative to the 1990s and in Eastern Europe markets expanded 60 % on average, whereas advanced countries experienced almost no growth, on average. Figure 2 shows the heterogeneity present at the country level.

Bond market capitalization. (a) G7 and other advanced economies; (b) Asia; (c) Eastern Europe and Latin America.

Notes. This figure shows the average annual ratios of market capitalization of outstanding bonds in domestic markets to GDP between 1990 and 2009 for selected countries. Domestic bond securities are defined as those issued by residents in domestic currency and targeted at resident investors. The data sources are the Bank for International Settlements (BIS) and the World Bank’s World Development Indicators (WDI).

Full figure and legend (107K )

Although emerging economies are closing the gap in the development of bond markets, they still lag behind in comparison with developed countries. For example, bond markets in Asia, although they are the most developed ones (in terms of overall depth) among the developing world, remain small compared to those in the G-7, at 56 % of GDP on average for East Asian economies, 35 % for China, and 33 % for India during 2000–2009, compared to about 86 % for developed countries.

Last but not least, there has been some convergence in the structure of financial systems as well. A mild but steady transition from a mostly bank-based model to one that is more complete and interconnected has been the broad trend in emerging countries ( Figure 3 ). For example, capital markets (bond and equity) in Latin American countries accounted for 64 % of their total financial systems on average in the 2000s in contrast with 54 % observed in the 1990s ( Figure 3. Panel B). Similarly, these markets have grown from 45 % to 55 % of the size of the financial system in Eastern European countries and from 18 % to 45 % of the financial system in China. In developed countries, these markets typically account for about 60 % –65 % of the financial system.

Relative size of debt markets. (a) Size of domestic financial systems; (b) Composition of domestic financial systems.

Notes. This figure shows the average size and structure of domestic financial systems between 1990 and 2009. Panel A shows total banking claims, the market capitalization of outstanding bonds, and the equity market capitalization as a percentage of GDP. Panel B shows the same figures expressed as percentage of the total domestic financial system. Numbers in parentheses show the number of countries included in the statistics for each region. The data sources are the IMF’s International Financial Statistics (IFS), the Bank for International Settlements (BIS), and World Bank’s World Development Indicators (WDI).

Full figure and legend (149K )

BEYOND FINANCIAL DEPTH

The increased depth of debt markets in emerging economies has come along with changes in the nature of financing. For example, the private sector has seen an expansion in local currency bond financing, the extent of dollarization of loans and bonds has declined, and the maturity of public and private sector bonds has typically increased. However, plenty of room remains for future development of the scope and depth of debt markets: bank credit has stagnated in various countries; firm financing has declined in relative terms; and private bond markets remain typically small and illiquid. We now review more systematically these qualitative developments in domestic debt markets in emerging economies and compare them with the trends in developed countries.

The composition of bank credit between the public and the private sector vary significantly across countries and has experienced some changes over the past two decades, not only in emerging countries but also in developed countries. The large expansion of banking systems in developed economies has been concentrated mostly in an increase of their claims on the private sector, which rose from 77 (50) % of GDP in the 1980s to 113 (98) % in the 2000s in G-7 economies (other advanced economies), accounting for 90 (97) % of total bank lending ( Figure 4. Panel A). In East Asian countries, lending to the private sector also expanded considerably, from 44 % to 72 % of GDP, or from 79 % to 87 % of total bank lending. In contrast, governments increased their borrowing not only in absolute but also in relative terms in many emerging markets, particularly in Eastern Europe, India, and Latin America over the same period, where the public sector represented a large fraction of total bank lending during the 2000s, at about 30 %. 34 %. and 27 % of the total claims by the banking sector, respectively. In the G-7 and East Asian countries that number was around 10 % and 13 %. respectively, over the same period.

Composition of bank credit. (a) Lending to the private and public sectors; (b) Composition of bank credit to the private sector.

Notes. This figure shows in Panel A the average share of public sector and private sector claims on total banking claims between 1980 and 2009. The percentages shown within the bars represent the size of both public and private claims as a percentage of GDP. For China, the data on claims on the public sector are not available for the 1980–1989 period, hence no data are shown for this period. Panel B shows the average share of commercial, mortgage and personal credit as share of total banking credit to the private sector. Numbers in parentheses show the number of countries in each region. The data sources are the IMF’s International Financial Statistics (IFS) and local sources.

Full figure and legend (168K )

Although not greatly expanding, credit to the private sector in emerging economies has undergone significant qualitative changes in its composition ( Figure 4. Panel B). For instance, credit to the private sector in China and Eastern Europe has shifted away from commercial lending and household financing toward mortgage credit. In Latin American countries, qualitative changes in the composition of private sector credit have also occurred. Consumer credit has grown significantly to the detriment of firm financing and mortgage lending. In contrast, the composition of bank credit has remained relatively stable in developed countries.

These patterns across emerging economies broadly suggest an unbalanced expansion of credit in a particular segment at the expense of the underdevelopment of others. For example, mortgages appear comparatively small across Latin American countries and commercial lending seems relatively small in China. Mortgages have increased significantly in both China and Eastern Europe, capturing in 2008–2009 58 % and 52 % of the loans, respectively. These patterns in the development of banking systems suggest that as emerging countries have grown over the past two decades, banks have expanded, in relative terms, in areas where it has been easy for them to grant credit at low risk, such as consumer credit through credit cards and collateralized loans, such as car loans and housing (not to mention the expansion of credit to the government). The increased use of capital markets by some corporations, which has lessened demand for bank finance, would also be consistent with these patterns.

Another key qualitative change in the nature of bank lending in emerging countries is a widespread decline in the dollarization of loans – although Eastern Europe is an exception ( Figure 5 ). Moreover, foreign currency deposits have become less dollarized not only in Asia and Latin America, but also in Eastern Europe. These developments are likely a consequence of the emerging market crises of the 1990s, when currency mismatches rendered the private sector vulnerable to currency fluctuations and limited policy options. Despite this decline in the extent of dollarization of the banking systems, the share of foreign currency loans and deposits remain particularly high when compared to developed world, especially in Eastern Europe and Latin America.

Dollarization of the banking system. (a) Share of foreign currency loans; (b) Share of foreign currency deposits.

Notes. This figure shows the extent of dollarization of loans and deposits in the banking system. Panel A shows foreign currency denominated loans as share of total loans averaged between 2000 and 2009. Panel B shows the extent of deposit dollarization as share of total deposits averaged between 1990 and 2009. Numbers in parentheses show the number of countries in each region. The data source is the IMF’s International Financial Statistics (IFS).

Full figure and legend (86K )

The concentration of banking systems may raise concerns about banking competition. When fewer and larger banks (higher concentration) exist, banks might be more likely to engage in anticompetitive behavior ( Berger, 1995 ). The literature has linked bank competition with lower prices for banking products, increased access to finance, and greater bank efficiency. Empirically, banking systems in emerging countries, with the exception of Latin American countries, are also becoming less concentrated, with a decreasing share of loans and deposits in the top five banks ( Figure 6 ). At the same time, foreign banks are increasing their presence in emerging markets more broadly ( Claessens and van Horen, 2014 ). The Eastern Europe and Latin American regions have the highest penetrations, noticeably larger than those in China, East Asia, India, and other advanced economies.

Concentration of banking systems. (a) Share of loans by the top five banks; (b) Share of deposits by the top five banks.

Bond markets

Despite their considerable expansion between 2000 and 2009, private bond markets as a percentage of GDP remain relatively small in emerging countries in comparison to those in more developed countries and to public bond markets. These private bond markets entail bonds issued by the corporate sector and financial institutions. For example, private bond market capitalization typically represented around 40 % of GDP in developed countries during the 2000s, whereas it stood at only 23 % of GDP in East Asian countries, a mere 10 % in Latin America, and at 4 % in Eastern Europe over the same period ( Figure 7. Panel A). A notable development is that private bond markets typically grew faster as a percentage of GDP than public (government) bonds, gaining space in relative terms and hinting at less crowding out by the public sector. East Asian countries are the exception – the capitalization of private bonds as a proportion of the total bond market capitalization declined slightly from the 1990s to the 2000s, from 45 % to 42 % on average over this period. In China, the relative expansion of the public sector bond financing is more striking, rising from about one half to almost two-thirds of the total bond market capitalization over the same period.

Activity in primary and secondary bond markets. (a) Composition of borrowers in bond markets; (b) Bond market turnover.

Notes. This figure shows in Panel A the average size of private and public bonds outstanding in domestic markets as a percentage of GDP between 1990 and 2009. Domestic bonds securities are defined as those issued by residents in domestic currency and targeted at resident investors. Panel B shows the average value of bond market trading as share of total bond market capitalization. Trading data include domestic private, domestic public, and foreign bonds traded in local stock exchanges. Numbers in parentheses show the number of countries in each region. The data sources are the Bank for International Settlements (BIS) and the World Federation of Exchanges (WFE).

Full figure and legend (105K )

Regarding liquidity in secondary bond markets, it remains a source of concern in a number of emerging countries. For example, bond market turnover, that is, the total number of bonds traded multiplied by their respective matching prices as a share of the total market capitalization, was in 2008 and 2009 around 60 % in the G-7 countries and reached 146 % on average across other advanced economies, whereas it was merely 12 % in Latin American countries and 15 % in India ( Figure 7. Panel B). However, there is significant heterogeneity in turnover levels across emerging economies. In East Asia and Eastern Europe, liquidity in bond markets stood at 45 % and 56 % over the same period, respectively. Furthermore, while bond market turnover declined in China, India, and Latin American countries, it has expanded considerably in East Asian countries, where trading volumes in secondary markets grew from 27 % during 2000–2003 to 45 % in 2008–2009 ( Figure 7. Panel B). Overall, these patterns suggest that primary bond markets seem to have developed substantially more than secondary markets, and they are broadly consistent with the evidence that institutional investors hold bonds to maturity and do little trading ( Raddatz and Schmukler, 2013 ).

The profile of new bond issues across emerging economies has been shifting over the past two decades. The maturity profile of both public and private sector bonds in Latin America has been extended during the 2000s vis—vis the 1990s, though surprisingly it has remained largely unchanged for Asian countries. For example, while the average maturity of newly issued private bonds in East Asia was 5.8 years in the 2000s (up from 5.7 in the 1990s), across Latin American countries it has increased significantly from 6.1 years to 8.1 years ( Figure 8. Panel A). The maturity of private bonds in the G-7 countries was not only longer to start with (at 9 years in the 1990s) but it is also lengthened (to 10.4 years in the 2000s). Due to data availability, we can only compare these trends on the bond market for the private sector with those of the public sector for Latin American countries. The increase in the average maturity of public debt in Latin American is even more striking, though it is not uniform across the region – between the 2000–2003 and the 2008–2009 periods, Brazil, Peru, and Uruguay showed significant increases in the maturity of public bonds, while Argentina’s and Chile’s public debt maturity remained largely the same or declined slightly ( Figure 8. Panel B).

Full figure and legend (81K )

Moreover, private bonds denominated in domestic currency in local markets have also increased as a share of total issued bonds by the private sector across East Asia and Latin America. In particular, foreign currency bonds decreased from 20 % to 19 % and from 33 % to 25 % of total outstanding private sector bonds in East Asian and Latin American countries, respectively, in the 2000s ( Figure 9. Panel A). Similar trends are observed for public sector bonds in Latin America ( Figure 9. Panel B). Despite this increase in the depth of local currency bond markets, emerging economies still lag behind developed countries, where the share of local currency bonds is significantly higher.

Currency composition of bonds at issuance. (a) Share of foreign currency bonds by the private sector; (b) Composition of outstanding bonds by the public sector.

Full figure and legend (122K )

Overall, the nature of bond financing is changing, though at a slow and somewhat uneven pace. As in developments in the composition of bank debt, these trends probably reflect a conscious effort by governments to change the profile of their debt, given the serious rollover difficulties that mismatches generated during earlier periods of global and domestic shocks ( Broner, Lorenzoni, and Schmukler, 2013 ).

INSTITUTIONAL INVESTORS

From the saver’s perspective, debt markets across emerging economies have also become more complex. While in the past typically only banks interacted directly with borrowers and lenders, in recent years there has been a greater diversity of players with a broader set of institutions, such as pension funds, mutual funds, and insurance companies. These nonbank financial institutions are intermediating savings, providing economy-wide credit, and offering a broader variety of products, as shown briefly in this section. The rise of these nonbank intermediaries has been a significant factor in the development of local debt markets across emerging countries to the extent that they provide a stable demand for financial assets. Nevertheless, as argued below, emerging economies still have a long way to go in raising the sophistication of its institutional investors as most of the savings are still channeled to government bonds and bank deposits.

Although banks continue to play a significant and stable role, nonbank financial intermediaries, such as pension funds, mutual funds, and insurance companies, have been gaining considerable space in emerging markets around the world ( Figure 10 ). For instance, pension fund assets represented 15 % of GDP in East Asian countries and 19 % in Latin American countries in the second half of the 2000s. Eastern European countries have typically smaller institutional investors (especially in the Russian Federation), but also fast-growing. As with most other features of the markets examined so far, these intermediaries are still smaller on average in emerging countries than in developed countries, reflecting to some extent the less developed state of their financial systems.