Comparing Financial Market Trends in Latin America and East Asia

As a measure of the growing interest in capital markets in emerging market economies, especially bond markets, a number of empirical studies have recently been carried out that help to identify trends.¹⁰ Their message is that domestic capital markets have been expanding, though in a fairly uneven way across regions and individual countries. In general, this expansion has been accompanied by growth of bank credit and increased use of international financial markets. Thus, the fear that domestic capital markets would displace other sources of finance does not seem to have occurred, although a better understanding of the relationship among the various markets would be helpful.¹¹

This paper looks at an important sub-set of emerging market financial systems, focusing on Latin America and East Asia. Table 1 begins with an overview, comparing the structure of the financial markets as a whole, including banks, bonds, and equity, over the period between 1990 and 2001 (the last year for which full data are available).¹² These data provide the opportunity to contrast both the overall depth of the financial markets in the two regions and the relative weight of the different components. The principal measure used in this table (and most of the others in the paper) is outstanding amounts of finance as a share of GDP to show the importance of the volume of finance relative to the size of an economy. In addition, dollar figures are shown to compare the absolute size of the markets, both between the two regions and with the international capital markets; the latter will be an element of the discussion about the ability of local markets to substitute for international integration.

Several important points emerge from the first panel of the table. First, in 2001, domestic financial markets as a whole in East Asia were twice as deep as those in Latin America (216% of GDP as opposed to 108%). Each individual component echoed the gap between the two regions, although it was especially prominent in the banking sector. Second, the relative strength of East Asia has been present at least since 1990, and it has been growing since then. The growth rate of total domestic finance as a share of GDP in Latin America was 42% compared to 59% for East Asia. It is interesting to note, however, that the Asian figures did not increase between 1995 and 2001. As will be discussed later in the paper, this stagnation was due to the Asian financial crisis and the response to it. We can see from the table that, despite the overall stagnation, there was a change in composition between 1995 and 2001 as stock market capitalization fell substantially, but both bank credit and bonds continued their upward trend. Third, with respect to capital markets (bonds and stock markets), Latin America saw a bigger increase over the period as a whole. In that region, bank credit declined as a share of GDP, while the capital market share more than doubled. Nonetheless, despite the recent growth, the combined bond and stock market share of GDP in 2001 in Latin America was less than half that in East Asia.

The second panel of Table 1 shows the relative importance within each region of the three financial sector components. In 2001, the three were relatively evenly distributed in Latin America, although bonds were slightly more important than bank credit or the stock market. In East Asia, bank credit was substantially more important than either of the other components. Some shifts have occurred over time. For Latin America, the most obvious is the reduced importance of bank credit and the increased role of capital markets, especially the stock market. For East Asia, bank credit remained about half the total, while bonds rose somewhat and the stock market shrank.

While Table 1 focuses on amounts outstanding of each component, another way to look at the process is to focus on recent trends in financial flows. Flow data for the period 1997-2001 significantly reinforce some of the differences observed in the data on stocks. In terms of flows, bank credit was much more significant for Asia and less so for Latin America, bonds were much more important for Latin America (especially for the public sector), while the stock market faded in both regions but especially Latin America.¹³ If these trends continue, it can be expected that the financial systems in the two regions will become even more dissimilar.

Finally, the third panel shows the absolute size of these markets. In Latin America, the three together grew from $770 billion in 1990 to $1.8 trillion in 2001. Subtracting bank claims, the capital markets share went from around $300 billion to a little less than $1.2 trillion in the same period. East Asia’s overall financial sector was substantially larger, but since the banking sector was also larger, the capital markets were about the same size as those in Latin America (around $1.3 trillion). These regional figures need to be broken down by countries, since the markets are currently organized on a national basis. Combining bonds outstanding and stock market capitalization, there were six markets over $200 billion in 2001: Argentina ($231bn), Brazil ($498), and Mexico ($215) in Latin America, and Korea ($525), Malaysia ($202), and Taiwan ($416) in East Asia. This compares with an average of around $5 trillion for the five largest OECD markets in bonds alone.¹⁴

Moving to a closer examination of the three components, including the differences among countries within regions, we begin with bank claims as shown in Table 2. In addition to the overall trends that we have already seen, the table compares credit to the private sector with total credit in each economy. The dominant message is that the private sector got more credit in East Asia than in Latin America. Private-sector credit in Latin America remained a little less than 30% of GDP throughout the period studied, while it increased from 58% to 83% in East Asia. In addition, the private sector share of total credit was much higher in East Asia: over 90% in 1990, falling to 80% in 2001. For Latin America, the figures are 59% and 73%, respectively. In other words, the trend in Latin America was toward a greater share for the private sector, while the opposite occurred in East Asia. By the end of our period, the difference between the two regions on this indicator had been reduced, although a huge gap remained in terms of GDP share.

Table 2 also shows data for the seven Latin American and six East Asian countries for which data are most readily available. With one exception, they are also the largest emerging market economies in each region.¹⁵ A pattern emerges here that will be repeated with all our data: some countries in each region have much deeper financial markets than others. Indeed, the differences within regions are sometimes as important as those across regions. In terms of total bank credit, Brazil and Chile have much more credit available than other Latin American economies. In East Asia, the situation is more even across countries, though Indonesia and the Philippines lag behind. The same general pattern also holds for private-sector credit, although it is important to note that in several countries in both regions, private-sector credit as a share of GDP actually fell between 1990 and 2001 (Brazil, Colombia, Mexico, Venezuela, and Indonesia).

Table 3 shows total domestic bonds outstanding and bonds issued by the private sector. As can be seen, data problems are more severe for bonds than for bank loans. Several countries are missing from the table for 1990. It is not clear if the data are unavailable or if there were no bond issues; in any case, they would not have been large. For the private sector, data are only available as of 1993 and again for selected countries. For total bonds outstanding as a share of GDP, we already know that East Asia far outpaces Latin America, but an even greater difference exists for private-sector bonds. While the share going to the private sector increased in both regions, by 2001 the private sector (including both the corporate and financial sectors) in Latin America still accounted for only 20% of total bonds outstanding (representing 6% of GDP). In East Asia, the private sector represented 91% of total bonds (39% of GDP). By definition, then, the public sector accounted for 80% of bonds in Latin America, but less than 10% in East Asia.

Again, we see substantial differences within as well as across regions. Chile and Brazil continue to dominate in Latin America, as do Korea and Malaysia in East Asia. There nonetheless remains a very large gap between the leading economies in the two regions, with the two East Asian leaders having 55% of GDP in private-sector bonds, while the figure is less than 20% in Latin America.

The stock market is potentially an important source of finance for private firms,¹⁶ but the figures for market capitalization give a greatly inflated view of their role. As mentioned earlier, issuance of shares on the markets in both regions (the so-called primary markets) has fallen to a very low level in the last five years (to only 2% of GDP in Latin America and 8% in East Asia) compared to the market capitalization figures shown in Table 4 (39% and 62% of GDP, respectively). In terms of individual countries, there are some small changes in comparison with bank loans and bonds. Chile and Argentina have the two largest stock markets (in terms of GDP share) in Latin America, while Malaysia and Taiwan top the list in East Asia.

Table 4 also shows the turnover ratio, which is an indicator of how active the secondary market is in each country or region. The measure is defined as the total value of shares traded during a given period, divided by the average market capitalization for the period. The ratio is important because the more liquid a market is (the higher the turnover ratio), the more willing investors are to put money into the stock market. It is also argued that more active trading provides more accurate pricing of individual issues and improves the allocation of resources. The data show that turnover in East Asia vastly exceeds that in Latin America (95% to 14%, respectively); indeed, turnover has actually fallen in Latin America during the past decade. There appears to be some relationship between the absolute size of market capitalization and turnover, as opposed to share of GDP, which is our basic measure in this paper. Of the five markets with capitalization over $100 billion, four (Brazil, Mexico, Korea, and Taiwan) had the highest turnover – although Brazil and Mexico were way below their Asian counterparts.¹⁷ Thailand is an exception with high turnover but a relatively small market. Chile is also an interesting case with the largest market in terms of GDP in Latin America but a low turnover; the reasons will become clear a bit later.

Finally, Table 4 shows the number of firms listed on each stock market over the period 1990-2001. The two regions started out in 1990 with very similar numbers: 1,624 in Latin America and 1,664 in East Asia. By 2001, however, the number of listed firms in Latin America had fallen to 1,344, while in Asia they had more than doubled to 3,706. Every country in the Asian sample showed buoyant growth, while only two in Latin America grew at all. The trend in number of listed firms complements the data on market capitalization, so we can better understand the relative role of price and volume increases.

Up till now, we have been focusing on domestic financial markets, but governments and the private sector have another alternative through the international financial markets. International banks can provide loans to borrowers in emerging market economies (either through the head offices or local branches in emerging economies themselves), and actors in the latter can also issue bonds or stocks internationally. Table 5 provides information on international bank loans and bonds outstanding [data on equity will be added later], in 1995 and 2001. Here we find a somewhat different pattern than we have seen in previous tables: international loans and bonds in 2001 constituted a slightly larger share of GDP (29%) in Latin America than they did in East Asia (24%). The difference was more significant for bonds than bank loans. Looking at the data in another way, international finance (loans and bonds) as a share of total finance for Latin America was 21% versus 10% in East Asia. Nonetheless, the addition of the international component does not offset the difference we have already found with domestic finance between the two regions. Three countries had ratios well above the regional averages: Argentina, Korea, and Malaysia; in the case of the two Asian countries, this international dependence developed between 1995 and 2001, presumably related to the financial crisis. Argentina’s share also rose substantially, and that country also was in crisis conditions in 2001.

International loans and bonds are not the only way in which international capital plays a role in the financial sector of emerging economies. The share of foreign ownership of banks and other financial service institutions has increased tremendously in the last few years, especially in Latin America but to a lesser extent in Asia as well. In addition to FDI, portfolio investment in local stock and bond markets has also increased as countries have liberalized their restrictions on foreign participation. It was hoped that a greater foreign presence would strengthen the financial systems through better management, newer technology, and deeper pockets, but the effects are still controversial.¹⁸

A final set of data that is essential for analyzing capital markets concerns the purchasers of bonds and stocks. Table 6 begins to shed some light on this issue, although data are difficult to compile and country-specific categories make them difficult to interpret; the data are also limited to debt securities. Nonetheless, some useful points can be extracted from the table. First, commercial banks are the dominant purchasers in East Asia, holding nearly half of all bonds; in Latin America, the banks' share is less than one third. Second, on the Latin American side, the single largest category of purchaser is institutional investors (private pension funds, insurance companies, and investment funds of various kinds). The third major difference concerns the category “others,” which includes individual investors and corporations. This group is much more significant in Asia than in Latin America. These different types of buyers are important because they have different types of portfolio requirements. To take the two extremes, institutional investors tend to buy and hold, while individual investors may be much more likely to trade frequently. Thus, the profile of investors is probably related to the turnover rate we saw in Table 4 (assuming that owners of stock and bonds behave in similar ways).