Whither Latin American Capital Markets?

Transcription

1 SEPTIMO CONGRESO DE TESORERIA Cartagena de Indias, Colombia October 21-22, 2004 Whither Latin American Capital Markets? Augusto de la Torre The World Bank

2 Structure of the Presentation 1. Evolution of Capital Markets 2. Role of Macro and Institutional Fundamentals 3. Role of Reforms 4. Policy Implications 1

3 Structure of the Presentation 1. Evolution of Capital Markets 2. Role of Macro and Institutional Fundamentals 3. Role of Reforms 4. Policy Implications 2

4 Financial Sector Size Across Countries Percentage of GDP, % 287% 140% 120% 100% 80% 60% 40% 20% 0% n/a U.S. U.K. Japan Thailand Malaysia Hong Kong Peru Mexico Colombia Chile Brazil Argentina Bank Credit to the Private Sector Stock Market Capitalization Amount Outstanding of Domestic Private Sector Bonds 3

5 Evolution of Stock Market Capitalization Percentage of GDP 160% 140% 120% 100% Asian countries 95% 80% 60% G-7 countries 72% 40% 20% Latin American countries 25% 0% Colombia 12%

6 Evolution of Value Traded in Stock Markets Percentage of GDP 120% 160% 99% 86% 95% 140% 100% 120% Asian countries 80% 100% Asian countries 60% 80% 72% 60% 40% G-7 countries 40% 25% 20% 20% 3% 12% 0% 0% G-7 countries Latin American countries Colombia Colombia Latin American countries

7 Evolution of Outstanding Domestic Public Sector Bonds Percentage of GDP 70% 60% G-7 countries 63% 50% 40% 30% 20% Asian countries 23% 17% 10% Latin American countries 0%

8 Evolution of Outstanding Domestic Private Sector Bonds Percentage of GDP 60% 52% 50% G-7 countries 40% 34% 30% Asian countries 20% 10% Latin American countries 11% 0%

9 Trends in the Number of Listed Firms in Stock Markets Across Countries 1, Poland Hungary Czech Rep. Thailand Malaysia Indonesia Peru Mexico Colombia Chile Brazil Argentina

10 Stock Market Internationalization: Ratio of Internationalized Firms to Firms Listed in Domestic Stock Markets Percentage 20% 18% 18% 15% 13% Latin American countries G-7 countries 12% 10% 8% Asian countries 8% 5% 3% Colombia 0%

11 Stock Market Internationalization: Ratio of Value Traded Abroad to Value Traded Domestically 140% 120% 100% 96% 122% 80% 60% 40% 20% 0% 15% 11% 1% 0% 0% 4% 3% 0% 3% 4% Latin American Countries Colombia G-7 Countries East Asian Countries

12 Structure of the Presentation 1. Evolution of Capital Markets 2. Role of Macro and Institutional Fundamentals 3. Role of Reforms 4. Policy Implications 11

13 Why is Latin America Where it is? Is the poor performance of domestic markets in Latin America the result of poor fundamentals? Or is there a shortfall between fundamentals and actual outcomes? Need to control for fundamentals Most analyses of capital market development focus only on domestic market variables, disregarding foreign market activity Not consistent with increasing participation of emerging markets in international financial markets We analyze how fundamentals impact domestic stock market development and internationalization and what is the impact of internationalization on domestic markets We also analyze the impact of fundamentals on the currency composition of government bond markets Equity contracts do not imply exposure to exchange rate risk 12

14 How Do Fundamentals Affect Stock Market Development and Internationalization? Two possible views on how economic fundamentals influence domestic stock market development and internationalization Better fundamentals spur more developed domestic stock markets and reduce the use of international markets Firms want to escape a poor domestic environment (Karolyi 2004) Related to the bonding argument (Coffee 1999; Reese and Weisbach 2002) Better fundamentals increase the attractiveness of assets to foreign investors and lead to more use of international markets Macroeconomic and institutional factors determine the willingness of international investors to supply financing to firms Firms from good institutional environments will be granted access to international markets 13

15 How Does Internationalization Affect Domestic Stock Markets? Two possible views on the impact of internationalization on the trading and liquidity of the firms that remain in the domestic market Internationalization hurts the liquidity of domestic firms Trading of international firms migrates to international markets, reducing their domestic trading ( migration ) Given the fixed costs of operating a market, the drop in the trading of international firms increases the transaction costs of domestic firms ( spillovers ) Internationalization increases the attractiveness of firms, leading traders in the domestic market to shift their trading out of domestic firms and into international firms ( domestic trade diversion ) Internationalization improves domestic market liquidity Internationalization transforms a segmented equity market with low liquidity into an integrated market with high trading activity and liquidity Internationalization induces a compositional shift in domestic trading towards domestic firms as the trading of international firms migrates abroad 14

16 Role of Fundamentals: Stock Markets Regressions Results Improved macro policies and institutional quality foster domestic stock market development as well as internationalization Better fundamentals increase international activity relative to domestic activity These outcomes are not independent, as the migration of securities trading abroad has had an adverse effect on the liquidity of local markets Domestic stock markets in Latin America are less developed than predicted by the institutional and macroeconomic variables, and the degree of internationalization in the region is higher than predicted 15

17 Regression Results: Impact of Fundamentals on Domestic Stock Markets Dependent Variable Market Capitalization / GDP Log of GDP per capita Law and Order Shareholders Rights Government Deficit / GDP Capital Account Liberalization Value Traded Domestically / GDP Total Equity Flows / GDP No. of Observations No. of Countries R-squared Least square regressions estimated using random effects models for a panel of 82 countries between 1975 and A constant is estimated but not reported. Red bold numbers mean significance at least at the 10 percent level. 16

18 Regression Results: Impact of Fundamentals on Stock Market Internationalization Dependent Variable Market Capitalization of International Firms / GDP Log of GDP per capita Law and Order Shareholders Rights Government Deficit / GDP Capital Account Liberalization Value Traded Abroad / GDP Total Equity Flows / GDP No. of Observations No. of Countries No. of Uncensored Obs No. of Left-Censored Obs Tobit regressions estimated using random effects models for a panel of 82 countries between 1975 and A constant is estimated but not reported. Red bold numbers mean significance at least at the 10 percent level. 17

19 Regression Results: Impact of Fundamentals on Internationalization Relative to Domestic Activity Dependent Variable Market Capitalization of International Firms / Total Market Capitalization Log of GDP per capita Law and Order Shareholders Rights Government Deficit / GDP Capital Account Liberalization Value Traded Abroad / Value Traded Domestically Total Equity Flows / GDP No. of Observations No. of Countries No. of Uncensored Obs No. of Left-Censored Obs Tobit regressions estimated using random effects models for a panel of 82 countries between 1975 and A constant is estimated but not reported. Red bold numbers mean significance at least at the 10 percent level. 18

20 Regression Results: Effects of Internationalization on Domestic Firms Dependent Variable: Log of One Plus the Turnover Ratio of Domestic Firms Share of international companies Log of one plus the aggregate turnover ratio in international markets Log of GDP per capita Law and order Capital account liberalization Log of total assets Number of firms 2,400 1,531 1,530 1,509 1,517 1,494 Number of observations 12,762 7,312 7,298 7,233 7,138 7,045 Country dummies Yes Yes Yes Yes Yes Yes Year dummies Yes Yes Yes Yes Yes Yes R-squared Least square regressions with robust standard errors for a panel of data covering over 3,000 firms across 55 emerging market countries during the years 1989 to Red bold numbers mean significance at least at the 10 percent level. 19

21 Regression Results: Latin America is Different Dependent Variable: Residuals of Regressions on Stock Market Development Market Capitalization / GDP Value Traded Domestically / GDP Dummy Latin America Dependent Variable: Residuals of Regressions on Stock Market Internationalization Market Capitalization of Value Traded Abroad / GDP International Firms / GDP Dummy Latin America Dependent Variable: Residuals of Regressions on Stock Market Internationalization Relative to Domestic Activity Market Capitalization of International Firms / Total Market Capitalization Value Traded Abroad / Value Traded Domestically Dummy Latin America Least square regressions on the residuals of the estimations on stock market development and internationalization. Each column corresponds to a different specification. A constant is estimated but not reported. Red bold numbers mean significance at least at the 10 percent level. 20

22 Role of Fundamentals: Government Bond Markets The focus of the analysis is the currency denomination of government bonds Foreign currency debt can be less expensive than domestic currency debt but exposes governments to exchange rate risk, increasing the likelihood of financial crises The size of the economy and the currency regime are two important macro factors that might impact the development of government bond markets and their currency composition 21

23 Role of Fundamentals: Government Bond Markets Regressions Results Larger size of the economy and better fundamentals foster domestic currency government bond markets development Fixed exchange rate regimes and a smaller size of the economy increase the share of foreign currency bonds Latin American countries tend to have smaller domestic currency government bond markets than predicted by fundamentals 22

24 Regression Results: Impact of Fundamentals on Domestic Currency Government Bond Markets Dependent Variable: Log of Domestic Currency Government Bonds Outstanding / GDP Lagged variables Initial values Log of GDP Log of total deposits / GDP Inflation index Fiscal burden Institutionalized democracy Actual exchange rate regime Observations Number of Countries Feasible generalized least squares regressions with heteroskedastic error structures and autocorrelation within countries for a panel of 36 countries between 1993 and The actual exchange rate regimes increases with the degree of flexibility of the exchange rate. Red bold numbers mean significance at least at the 10 percent level. 23

25 Regression Results: Impact of Fundamentals on Foreign Currency Government Bond Markets Dependent Variable: Log of Foreign Currency Government Bonds Outstanding / GDP Lagged variables Initial values Log of GDP Log of total deposits / GDP Inflation index Fiscal burden Institutionalized democracy Actual exchange rate regime Observations Number of Countries Feasible generalized least squares regressions with heteroskedastic error structures and autocorrelation within countries for a panel of 36 countries between 1993 and The actual exchange rate regimes increases with the degree of flexibility of the exchange rate. Red bold numbers mean significance at least at the 10 percent level. 24

26 Regression Results: Impact of Fundamentals on the Currency Composition of Government Bond Markets Dependent Variable: Log of Foreign Currency Government Bonds Outstanding / Total Government Bonds Outstanding Lagged variables Initial values Log of GDP Log of total deposits / GDP Inflation index Fiscal burden Institutionalized democracy Actual exchange rate regime Observations Number of Countries Feasible generalized least squares regressions with heteroskedastic error structures and autocorrelation within countries for a panel of 36 countries between 1993 and The actual exchange rate regimes increases with the degree of flexibility of the exchange rate. Red bold numbers mean significance at least at the 10 percent level. 25

27 Regression Results: Domestic Currency Government Bond Markets are Less Developed in Latin America Dependent Variable: Residuals of Regressions on Domestic Currency Government Bonds Lagged variables Initial values Dummy Latin America Number of observations Number of countries R-squared Least square regressions on the residuals of the estimations on government bond market development. Each column corresponds to a different specification. A constant is estimated but not reported. Red bold numbers mean significance at least at the 10 percent level. 26

28 Structure of the Presentation 1. Evolution of Capital Markets 2. Role of Macro and Institutional Fundamentals 3. Role of Reforms 4. Policy Implications 27

29 Role of Reforms Panel data results might be driven by-cross country differences and not by the evolution of variables over time A relevant question is whether the poor state of development of securities markets in Latin America is the consequence of lack of s relative to other regions The evidence shows no shortage of s Did these s have the expected positive impact on capital markets? Was the impact of s different in Latin America? 28

30 No Shortage of Reforms: Financial Liberalization 3.0 More liberalization G-7 Rest of Western Europe Southeast Asia 1.5 Latin America Less liberalization The liberalization index is calculated as the simple average of three indices (liberalization of the capital account, domestic financial sector, and stock market) that range between 1 and 3, where 1 means no liberalization and 3 means full liberalization. These data are then aggregated as the simple average between countries of each region. 29

31 No Shortage of Reforms: Capital Markets Regulation and Infrastructure Percentage of Latin American Countries Having Implemented Reforms 120% 100% 88% 94% 91% 100% 92% 80% 60% 56% 63% 64% 62% 40% 31% 25% 27% 33% 20% 15% 0% Supervisory Agency Creation Establishment of Insider Trading Laws Custody Arrangements Trading Systems Clearing and Settlement Processes 0% Before 1990 By 1995 By

32 Percentage of GDP 70% No Shortage of Reforms: Pension System Reform Assets Held by Mandatory Pension Funds in Latin America 60% 61% 50% 40% 40% 30% 20% 10% 0% 17% 16% 13% 9% 8% 8% 6% 3% 4% 3% 1% 3% 3% 0% 0% 1% Argentina Bolivia Chile Colombia Costa Rica El Salvador Mexico Peru Uruguay Dec-1998 Jun

33 No Shortage of Reforms: Privatization Cumulative Amount Raised by Privatizations in Latin America Billion U.S. dollars

34 Role of Reforms: Regressions Results All the s analyzed have a significant positive impact on domestic stock market development and internationalization Reforms accelerate internationalization, leading to a higher share of international activity relative to domestic activity The impact of s on domestic stock market activity was smaller in Latin America than in other countries Reforms in Latin America had a higher impact on stock market internationalization than in other regions 33

35 Regression Results: Impact of Reforms on Domestic Stock Markets Financial liberalization Privatization Reform No. of observations No. of countries R-squared Financial liberalization Market Capitalization / GDP Infrastructure Value Traded Domestically / GDP Infrastructure Institutional Institutional Enforcement of insider trading laws Enforcement of insider trading laws Pension Pension Privatization Reform No. of observations No. of countries R-squared Fixed effects regressions for countries implementing s between 1975 and A constant is estimated but not reported. Red bold numbers mean significance at least at the 10 percent level. 34

36 Regression Results: Impact of Reforms on Stock Market Internationalization Market Capitalization of International Firms / GDP Financial liberalization Privatization Reform No. of observations No. of countries R-squared Financial liberalization Infrastructure Institutional Value Traded Abroad / GDP Infrastructure Institutional Enforcement of insider trading laws Enforcement of insider trading laws Pension Pension Privatization Reform No. of observations No. of countries R-squared Fixed effects regressions for countries implementing s between 1975 and A constant is estimated but not reported. Red bold numbers mean significance at least at the 10 percent level. 35

37 Regression Results: Impact of Reforms on Internationalization Relative to Domestic Activity Market Capitalization of International Firms / Total Market Capitalization Financial liberalization Infrastructure Institutional Enforcement of insider trading laws Privatization Reform No. of observations No. of countries R-squared Value Traded Abroad / Value Traded Domestically Financial liberalization Infrastructure Institutional Enforcement of insider trading laws Pension Pension Privatization Reform No. of observations No. of countries R-squared Fixed effects regressions for countries implementing s between 1975 and A constant is estimated but not reported. Red bold numbers mean significance at least at the 10 percent level. 36

38 Regression Results: Reforms Had a Lower Impact on Domestic Stock Markets in Latin America Financial liberalization Privatization Reform Dummy Latin America interacted with No. of observations No. of countries Financial liberalization Infrastructure Infrastructure Market Capitalization / GDP Institutional Value Traded Domestically / GDP Institutional Enforcement of insider trading laws Enforcement of insider trading laws Pension Pension Privatization Reform Dummy Latin America interacted with No. of observations No. of countries Fixed effects regressions for countries implementing s between 1975 and A constant is estimated but not reported. Red bold numbers mean significance at least at the 10 percent level. 37

39 Regression Results: Reforms Had a Higher Impact on Stock Market Internationalization in Latin America Financial liberalization Privatization Reform Dummy Latin America interacted with No. of observations No. of countries Financial liberalization Market Capitalization of International Firms / GDP Infrastructure Infrastructure Institutional Institutional Enforcement of insider trading laws Value Traded Abroad / GDP Enforcement of insider trading laws Pension Pension Privatization Reform Dummy Latin America interacted with No. of observations No. of countries Fixed effects regressions for countries implementing s between 1975 and A constant is estimated but not reported. Red bold numbers mean significance at least at the 10 percent level. 38

40 Regression Results: Reforms Had a Higher Impact on Stock Market Internationalization in Latin America Market Capitalization of International Firms / Total Market Capitalization Financial liberalization Privatization Reform Dummy Latin America interacted with No. of observations No. of countries Financial liberalization Infrastructure Institutional Enforcement of insider trading laws Pension Value Traded Abroad / Value Traded Domestically Infrastructure Institutional Enforcement of insider trading laws Pension Privatization Reform Dummy Latin America interacted with No. of observations No. of countries Fixed effects regressions for countries implementing s between 1975 and A constant is estimated but not reported. Red bold numbers mean significance at least at the 10 percent level. 39

41 Structure of the Presentation 1. Evolution of Capital Markets 2. Role of Macro and Institutional Fundamentals 3. Role of Reforms 4. Policy Implications 40

42 Policy Implications What Went Wrong and What to Do Next? The same evidence examined from different perspectives can lead to different diagnoses and policy recommendations Three possible views Be patient and redouble the effort Get the sequence right Revisit basic issues and revise expectations 41

43 Policy Implications Be Patient and Redouble the Effort Diagnosis: The salient elements of capital market are well known Past efforts were essentially right Crises are part of growing pains and cleansing process Perceived failures are related to lack of implementation Reform dividends have long gestation periods Policy recommendation: Accelerate and deepen s 42

44 Policy Implications Get the Sequence Right Diagnosis: Financial liberalization went too fast Opening before achieving a minimum threshold of institutional strength leads to instability and crises Capital markets development has a building block nature Policy Recommendation: Get the sequencing right Re-prioritize s, strengthening the regulatory and institutional framework before further liberalization Reforms should be ordered so that preconditions are put in place first 43

45 Policy Implications Revisit Basic Issues and Reshape Expectations Diagnosis: Gaps in knowledge and shortcomings of previous views Reforms spur both domestic securities markets development and internationalization Resistance to might not yield in the absence of outside competition Implicit premise that developed capital markets are the adequate benchmark for developing countries Policy Recommendation: Step back and revisit basic issues 44

46 Financial Globalization Undue emphasis on domestic capital markets is not consistent with financial globalization Liquidity Liquidity is a positive function of market size and economies of scale and network and agglomeration effects Risk Diversification Policy Implications Basic Issues Local capital markets in Latin America have added very little to risk diversification and the scope for diversification might be curbed by the economics of the risk diversification process 45

47 Policy Implications Basic Issues Financial Globalization Undue emphasis on domestic capital markets is not consistent with financial globalization What matters is access to finance, regardless of where or by whom provided However, international financial integration does not eliminate the need for local markets Local financial development still matters for growth (Guiso, Sapienza, and Zingales 2003) Interface of financial globalization, weak currencies, and debt market development 46

48 Policy Implications Basic Issues Liquidity Secondary market liquidity underpins price revelation and is key for the development of other markets Liquidity is a positive function of market size and economies of scale and network and agglomeration effects Size threshold for liquidity in domestic government bond markets: billion U.S. dollars (McCauley and Remolona 2001) Size threshold for minimum liquidity in corporate bond issues: at least 50 million U.S. dollars according to market participants In Latin America the adverse impact of small market size on liquidity is increased by high concentration Improvements to market infrastructure can increase liquidity but will not lead to qualitative jumps 47

49 Policy Implications Basic Issues Risk Diversification Local capital markets in Latin America have added very little to risk diversification Pension funds equity and corporate bond holdings concentrated in a few highly rated issuers Small domestic markets can only sustain a few issuers that meet size, quality, and liquidity thresholds Systemic risk cannot be diversified away through investments in domestic markets Foreign investment managers heavily biased towards large issuers, both sovereigns and corporates Scope for diversification might be curbed by economics of the risk diversification process 48

50 Policy Implications Final Remarks Expectations need to be reshaped to accommodate the implications of financial globalization More analysis is needed to develop a light version of securities markets for small countries It is still an open question whether regional integration offers any advantages compared with better and deeper integration with developed financial centers Further research is needed to understand why Latin America has performed worse in terms of domestic market development but better in terms of internationalization than other regions 49

51 THE END 50