Descarga la aplicación para disfrutar aún más
Vista previa del material en texto
CHILEAN FINANCIAL MARKETS AND CORPORATE STRUCTURE* Fernando Lefort and Eduardo Walker** April, 2003 _____________________________________ * We thank Ronald Espinosa and Catherine Tornel, who provided excellent research assistance. ** Both authors, Escuela de Administración, Pontificia Universidad Católica de Chile. 1 I. INTRODUCTION It is well known by now that financial markets are important for economic growth and development. This is probably even more so in the case of emerging economies. In that context, the recent trends in stock trading volumes, market capitalization and new issues in many emerging economies should be a matter of concern. Furthermore, there have been important acquisitions and takeovers of many local firms (previously considered to be the "flagships" of the local economies) by foreign multinationals.1 This may cause local stock markets to shrink even further. At the same time, many important firms in emerging economies have gained direct access to international capital markets, making us question the role of “local” capital markets in this new context. On top of the above, the 1998 Asian Crisis has had negative effects on the future prospects of firms as well as on the overall performance of the economy. Indeed, in the case of Chile for the period 1999-2001 the simple average annual GDP growth rate was 2.1 percent, compared with an average of 6.8 percent for the previous 13-year period. Thus, the question of disentangling the different causes for the recent evolution of local capital markets becomes important. Here we look at this question by analyzing the evolution of Chilean capital markets (a "top-down" perspective) and of Chilean conglomerate structures ("bottom-up"), seeking to understand these recent trends by putting them in a more general, economic, legal and historical context. We ask the question whether in this case it is true that capital markets have become smaller, and in what sense, seeking to determine if there have been permanent structural changes. If capital markets have indeed become smaller or less useful from the perspective of productive firms, we should also find a downward trend in firm outside financing. Finally, we analyze these facts in the context of a discussion regarding the role that capital markets are expected to have in small economies. We organize the paper in accordance to the following working hypothesis: emerging economies are characterized by poor institutions and formal capital markets, and are subject to large economic and political shocks that shape corporate structure and trigger economic and legal reform. Financial market evolution responds to this changing environment. In section II of the paper, we summarize the main economic events that may have had an influence on capital market development in Chile and relate them directly to the Chilean 1 In Chile, this list includes CTC, Enersis, Endesa, Banco Osorno, Banco Santiago, IANSA, and many others. 2 conglomerate structure. In section III, we give a brief description of the Chilean legal framework and corporate governance mechanisms. In section IV, we describe the recent evolution of Chilean capital markets. In section V, we analyze in more detail the recent evolution of new issues and trading volumes and provide explanation for the observed trends. We end, in section VI, with a discussion on the role of local financial markets for the integrated small economy. II. ECONOMIC HISTORY AND CONGLOMERATE STRUCTURE IN CHILE A. FACTS AND EVENTS THAT HAVE SHAPED CAPITAL MARKET EVOLUTION IN CHILE There are several important economic and political events that have to be considered in order to understand the way that Chilean conglomerates and capital markets have evolved through time.2 Until 1973 capital markets as such did not really exist in Chile. Financial repression and credit rationing gave the rationale for the extensive use of internal capital markets and the subsequent appearance of bank-centered groups. In the mid-seventies, the first round of privatization took place at relatively attractive prices, in the context of a recently liberalized economy, a naïve legal environment and primitive capital markets. This gave incentives to the creation of significantly indebted groups or conglomerates. Following this period, the 1982 debt crisis is perhaps one of the most important events that have shaped the way in which Chilean corporations are organized even today. The crisis meant that most bank-based conglomerates became bankrupt. In addition, many important regulations were adopted as a consequence of the crisis. The debt crisis also implied that most productive firms were back in the hands of the state. Partly as a reaction to the crisis, with the purpose of reducing the country's debt levels, and also in preparation for the second privatization round, the 1984 tax law provided investors with incentives to buy equity. This law also eliminated the tax-related motivation for using debt. For example, Hernández and Walker (1993) show that after the debt crisis there was a significant reduction in the overall use of debt by firms, and a shift in its composition against bank debt. 2 Some of these issues are explained in detail in Lefort and Walker (2000a) 3 The second privatization round is another important event. It took place during the mid-eighties, implying some degree of equity market development, fueled by local pension funds and foreign investment funds. Important efforts were made in order to achieve a wide investor base (capitalismo popular). In theory, privatization of state-owned firms is likely to have important effects on the development of capital markets. Firms that before relied on centralized credit allocation may now opt for the bond and stock markets. Also, if the privatization process purposely considers a vast dispersion of property, higher transaction volumes in stocks are expected. Perhaps the most important economic event in terms of shaping financial markets and explaining capital markets evolution in Chile, was the early pension fund reform. Since its inception in the early eighties, significant pension funds have been accumulated in Chile, representing an important source of funds for companies that are channeled through the Chilean financial system. In addition, pension reform has meant that economic authorities have frequently had to modernize the existing regulations and institutions, trying to improve minority shareholder protection and the functioning of capital markets and supervision, among other issues.3 In general, however, over regulation may entail lack of competition in the financial services industry curtailing the growth possibilities of capital markets. For example, competition among security traders and stock exchanges probably imply transaction costs that do not inhibit trading. Also, prudential bank regulation implies fair competition among alternative fund suppliers, including pension funds. If these conditions are not present, a slower capital market development is expected. Any analysis of the Chilean case can not ignore the role of the very significant macroeconomic reforms that have taken place in the country. Walker and Lefort (2001) provide a list of institutional and economic reforms present in the Chilean case that have probably facilitated or conditioned capital market development. Their list includes macroeconomic stability in the context of a market-oriented economy,4 non-distorted 3 See Walker and Lefort (2001). 4 Eyzaguirre y Lefort (2000) also refer to the close relationship between economic growth andasset accumulation in Chile. 4 fundamental prices,5 adequate tax regime, progressive capital control liberalization, clear property right laws, including bankruptcy legislation and investor protection, and institutional reform in addition to pension reform. We therefore acknowledge that macroeconomic stability is a necessary (but probably not sufficient) condition in order to develop a capital market. B. A BRIEF HISTORY OF THE EVOLUTION OF CHILEAN CONGLOMERATES Conglomerates have been the traditional business structure in Chile for a long time. Their origins and evolution importantly respond to the political and economic events described earlier. During the first half of the 20th century a number of large state-owned companies was created in the context of a national plan of industrialization under the supervision of a public entity (CORFO). The privatization of these companies that took place much later gave origin to several of today’s conglomerates. In addition, responding to financial repression and seeking cheap credit, several groups were created around banks during the fifties.6 The socialist period of 1970-73 imposed severe conditions to the development and continuity of groups. However, Dahse (1979) and González (1978) identify for 1978 more or less the same groups as previous studies, although some important changes in property and new associations had taken place. During the late seventies, the first round of privatization provided a new push to the creation of economic groups, mostly around banks.7 Then, the 1982-83 bank crisis implied a large shock for groups. Bank failures and state intervention caused the disappearance of several conglomerates, such as Vial and BHC, and the reduction of others like Cruzat- Larraín. New laws and regulations put in place in response to the debt crisis greatly reduced the importance of banks for future groups. The second privatization round that took place during the mid-to-late eighties produced an upsurge of new groups. The privatization process was implemented partly 5 Based on the Chilean experience with endemic inflation, they recommend the availability of indexed (inflation-protected) fixed income instruments. Diamond and Valdés (1994) noted this. This is also emphasized in Walker (2002), who argues for the need of a consistent legal environment. 6 Paredes and Sánchez (1994) summarize several studies regarding the evolution of groups over the years. Lagos (1961) identifies eleven large groups, all related to banks in 1958. For 1966, Garretón and Cisternas (1970) identify 19 additional groups, most of them presumably small family groups. 7 The most important groups that appeared in that period were Cruzat-Larraín, BHC, and Claro. See Hachette and Lüders (1992). 5 with the purpose of achieving disperse firm ownership. Pension funds were allowed to buy equity for the first time, but eligible firms had to adopt important statutory restrictions, particularly in terms of ownership concentration. Yet economic groups rapidly took control over most newly privatized firms. The large size of the firms being privatized in some cases implied associations of Chilean with foreign companies.8, 9 Only three groups have been present since the sixties: Matte, Angellini and Luksic.10 Eleven of the twelve groups that were present in 1988 were still present in 1998. This is just an indication that this period of high economic growth encouraged the appearance of new conglomerates. The stability in their number from 1996 on, jointly with the wave of acquisitions and the 1999 recession implied a reduction in the number of conglomerates. Since 1996 an increasing number of foreign corporations acquired domestic firms traditionally controlled by Chilean conglomerates. C. CONTROL AND CAPITAL STRUCTURE OF CONGLOMERATES IN CHILE Unlike the U.S. and the U.K., corporate ownership in Chile is characterized by the high degree of ownership concentration. Furthermore, like in most emerging economies the identifying feature of corporate structure in Chile is the generalized presence of complicated structures or conglomerates called “grupos”. A working hypothesis of this paper is that the presence of “grupos” and some of their features are related to our past economic history and specifically to the events described earlier in this section. In this sub- section we summarize recent findings regarding control and capital structures of Chilean conglomerates and relate them, when possible, to some of the events previously discussed. C.1 Control mechanisms Lefort and Walker (2000c) look at several dimensions of control in Chilean conglomerates. Chilean conglomerates use mostly simple pyramid structures to separate control from cash flow rights. Chilean Corporations Law prohibits cross-holdings and, although allowed, dual class shares are relatively rare in Chilean corporations. As of December 2001, only 8 percent of Chilean listed companies had dual shares. Table 1, 8 Like Carter-Holt in the case of the Angelini group. 9 For 1993, Paredes and Sánchez (1994) identify seventeen major groups, 10 of which are new and related with the second privatization round. 10 Paredes and Sánchez (1994) interpret this evidence as significant mobility and no barriers to entry or exit of groups. 6 extracted from Lefort and Walker (2000c) indicates that Chilean groups use relatively simple pyramid structures where it is rare to find 4 layers of public corporations consolidated. However, the table clearly indicates that the number of layers used by groups has increased during the nineties. By 1990, only 13 percent of public corporations affiliated to groups were second or higher level. This figure increased to almost 35 percent by 1998. It is important to keep in mind, that although Chilean conglomerates are formed through relatively simple pyramid schemes of public companies, it is not always easy to ascertain the way the pyramid structures are controlled. The reason is that there are very few people among the largest shareholders. Controllers of these structures hold shares trough private holding companies with fantasy names that participate at all levels of the pyramid structure making very difficult to ascertain ultimate ownership to investors and regulators.11 In spite of these problems, Chilean conglomerates are relatively simple. An interesting hypothesis is that the simplicity of these structures is due to legislation put in place in order to protect pension funds from expropriation. Another consideration regarding the control structures of Chilean conglomerates is that because Chilean banks are forbidden to hold company shares, groups are structured around holding companies instead of banks. That norm is a direct consequence of the debt crisis of 1982. Because of the high degree of ownership concentration in Chilean companies, control is exercised, in practice, through board members elected directly or indirectly by the controlling groups. A survey of board practices in large listed Chilean companies indicates that only 55 percent of all board members are not directly related by family to the controlling shareholders or are not executives in the company or in other companies owned by the same controller.12 Moreover, Lefort and Walker (2000c) show that, on average, each board member holds 1.6 seats where largest groups tend to centralize board positions in fewer people as compared to smaller groups. This evidence suggests that even board members elected with minority shareholder votes are exclusive of group firms. An exception are board members elected by pension funds in large corporations. Iglesias 11 The large differencebetween personal income and corporate tax rates explains the wide use of private holding companies. 12 SpencerStuart and the Business School of the Pontificia Universidad Católica de Chile prepared the 2000 Board Index Report based on board practices used by 55 large listed Chilean companies 7 (2000) shows that 10 percent of board members in firms where pension funds own shares are elected with their votes. External mechanisms of control and corporate governance are rarely important in Chile. For instance, the efficiency increasing role of the market for corporate control in Chile is restricted by the very high levels of company ownership concentration. In the vast majority of companies this concentration eliminates the possibility of hostile takeovers. However, Since 1998 a large number of acquisitions have taken place in Chile. Lefort and Walker (2001) analyze 12 major acquisitions involving transfer of control between 1996 and 1999. They find that the average excess price for these 12 acquisitions was 70 percent, while the average control block purchased amounted to 40 percent of shares. On average, the cumulative abnormal return was approximately 5 percent, indicating that the average acquisition was perceived as value enhancing by the market. C.2. Capital and ownership structures Lefort and Walker (2000c) constructed consolidated balance sheets at the conglomerate level, for all non-financial public companies, in order to describe ownership and capital structure of Chilean economic groups. Some of the results of that study are summarized in Table 2. As we indicated above, groups are the predominant forms of corporate structure in Chile. The table shows that companies affiliated to groups hold 91 percent of total non-financial, listed assets. Conglomerates have increased their use of external finance (measured as the level of debt plus minority interest) reaching 70 percent in 1998. The evidence presented in the table also shows that, in general, controlling shareholders hold more equity than, in principle, is needed for control. The average controlling shareholder held 57 percent of the consolidated equity capital of the conglomerate in 1998. When interpreting this concentration figure, it is important to keep in mind, that a four layers pyramid structure can be controlled with much less than 51 percent of consolidated equity. Some other interesting facts about capital structure in Chilean conglomerates are the following. Minority shareholders own around 40 percent of the equity controlled by Chilean groups with pension funds managers and ADRs representing 25 percent each of minority shareholders interest. Regarding debt composition, Lefort and Walker (2000c) 8 showed that conglomerates are able to get significantly more long-term and bond debt financing than non-affiliated firms. III. LEGAL BACKGROUND Financial markets and corporate structure are also affected by laws and regulations, which themselves many times respond to shocks or transcendental events in the political and economic environment, such as the ones outlined above. A. GENERAL FRAMEWORK Table 3 shows the main laws and supervisory institutions that regulate financial activity in Chile. Among the laws, the most relevant are the Banking Law, the Security Markets Law and the Corporations Law. In addition, a series of other laws and regulations specify the rules of the game for institutional investors such as pension funds, mutual funds and foreign capital investment funds, and rule bankruptcy procedures among other things. Three main supervisory entities overlook different aspects of financial markets in Chile. The “Superintendencia de Valores y Seguros” (SVS) is in charge of supervising capital markets functioning and public company information disclosure practices. The “Superintendencia de Bancos e Instituciones Financieras” (SBIF) supervises the compliance of banking regulations. Finally, the “Superintendencia de Administradoras de Fondos de Pensiones” (SAFP) overlooks pension funds. The Central Bank also participates actively in the financial system regulatory and supervisory process, especially in issues regarding international transactions and foreign market participants. Self regulation is almost inexistent in Chilean capital markets. Regulations are imposed by the appropriate authorities and supervised by governmental entities. Although, the Chilean legal system follows the tradition of French Civil Law, the Banking law, the Securities Market Law and the Corporations Law were written and reformed mimicking their homologues in the US. Since the Chilean Judiciary system does not have the flexibility of a judiciary under Common Law, some tension arises between the spirit of the Law and its application.13 Moreover, there are still obvious differences in ownership concentration, market liquidity and law enforcement between Chile and the US. 13 See Laporta, Lopez-de-Silanes, Shleifer and Vishny (1996). 9 A.1 Banking Sector As we indicated previously, the 1982-83 collapse of the financial system importantly shaped the evolution of the banking sector. As a consequence of the crisis a number of bailout measures were taken. After the crisis (in 1986) a comprehensive new banking law was dictated. In general terms, the new law included partial deposit insurance; requirements that financial investments be valued at market prices; credit risk provision requirements; and restrictions on currency and maturity mismatching. In addition, the new law introduced strict limitations on related lending and prohibited banks to keep shares in their portfolios, with a few minor exceptions.14 The 1986 Banking Law is therefore partly responsible for the reduced importance of banks for corporate structure and governance in Chile. Bank credit was substituted off as a corporate source of funds and replaced partly by equity issues and to a lesser extent by corporate debt. Also, banks stopped being a central unit of economic groups, at least in organizational terms. At the end of 1997 new amendments were introduced to the 1986 Banking Law giving banks more flexibility and widening their business scope.15 We can therefore guess that banks may become more important actors in capital markets development and corporate governance in Chile in the future. A.2 Capital Markets The current institutional arrangements of Chilean capital markets developed starting in 1980 with the creation of the main supervisory entity, the "Superintendencia de Valores y Seguros". The Securities Market Law and the Corporations Law comprise the legal framework governing capital markets and the actions of listed companies in Chile. The main body of both the Corporations Law and the Securities Law was written in 1981. In response to the changing environment and as a consequence of the increasing financial integration and sophistication of Chilean capital markets, both laws were amended in 1989 and more deeply in 1994. Modifications consisted mainly in broadening the investment alternatives to institutional investors, and improving the regulation in matters such as 14 See Eyzaguirre and Lefort (1999). 15 The most important changes were the following: procedures for new bank licenses were established; the Basle recommendations on capital requirements were adopted; regulations on new domestic branches were simplified; international branches and operations were more easily allowed; and banks were allowed to hold shares of companies in related business such as stock brokers, investment and mutual fund managers, factoring and others. 10 conflict of interests and risk rating systems. More recently both laws were amended by the Law Nº 19,705 of year 2000known as the OPA Law.16 In 2001 the Securities Market Law was again amended. B. LEGAL PROTECTION TO INVESTORS AND CORPORATE GOVERNANCE IN CHILE The recent interest for corporate governance practices around the world has also reached Chile. From the local point of view, the large and controversial control premiums, paid in several acquisitions of control stakes of flagship Chilean companies by foreign companies, triggered legal reform and investor awareness of the problem. A standard framework to analyze corporate governance practices is provided by the OECD principles. These principles acknowledge not only the importance of legal protection, but also other mechanisms of corporate governance. The principles are structured in 5 categories that look at shareholders rights, board responsibilities and disclosure of information among others.17 Although it is difficult to ascertain the extent of investor protection and of OECD principles compliance in Chile, we present in Table 4 a tentative summary of the degree of compliance of the main OECD principles. This table was prepared based on the analysis of the legal framework, market participants opinions and the conglomerate structure results discussed previously in this paper. The table shows that a preliminary review to corporate governance practices in Chile indicates that 11 out of the 16 OECD principles reviewed are adequately complied in Chile while 5 are not. These results indicate a 69% of compliance. Among the principles unsatisfactorily complied with it is interesting to note some of the features of Chilean laws and conglomerate structure previously discussed. First, among the shareholder rights, Chilean practices do not assure the correct disclosure of capital and ownership structures. Second, boards do not tend to act in an independent manner from controlling shareholders. However, as we mentioned above, board members elected by institutional investors have played an important role in several cases of alleged violation of minority shareholders rights. Pension fund representatives are prohibited by law to vote for 16 OPA stands as “oferta pública de adquisición de acciones” and refers to the tender offer requirement during takeovers. 17 Lefort (2003) looks at corporate governance in Chile and discusses the compliance of several of these principles. 11 a candidate related to the controlling shareholder and therefore their votes tend to represent the minority interest. They are required to disclose their votes and candidates, and to inform the public the reason behind these decisions. During the last few years, pension and investment fund managers have stood against corporate actions that could hurt the minority interest in the company, alerting the press and the authorities and initiating legal actions. However, the evidence provided here suggests that the professional-independent board member is seldom present in Chilean corporations. With respect to current corporate governance practices, Chilean laws have played an important role. The SVS has taken the lead in recent reforms promoting minority shareholder protection and disclosure. In December of 2001, the Securities Market Law and the Corporations Law were amended. The amendment was known as the Corporate Governance Reform and introduced changes in five areas of the law. First, the market for control was regulated requiring that transactions involving changes of control be performed through a tender offer under a version of the equal opportunity rule. Second, the regulator increased the requirements on information and disclosure to listed corporations, especially in the case of transactions with related parties. Third, large listed corporations were required to form a board committee with a majority of board members unrelated to the controlling shareholder. The role of this committee was specified in the law. Fourth, share repurchases were allowed in order to implement stock option packages as an incentive to executives. Fifth, equal treatment of foreign shareholders was guaranteed by law especially in matters regarding voting procedures. The amendments included a transitory rule that allowed firms to postpone the adoption of the new regulations regarding changes of control for three years. Most large companies adopted the transitory rule. In summary, up to this point in this paper we have argued that economic events have shaped the Chilean corporate structure and legal framework, affecting in the end the corporate governance mechanisms put in place and their efficiency. However, another possibility is that in the decades to come and following the recent legal reforms on Chilean capital markets, we will witness a final outcome that is independent of Chile’s specific history (or path), indicating that, in the end, all governance systems converge to the same optimal pattern.18 Given the strong path-dependence observed, this seems unlikely to us. 18 See Johnson and Shleifer (2000). 12 IV. CAPITAL MARKET EVOLUTION IN NUMBERS A EVOLUTION BY INSTRUMENT TYPE Chilean capital markets have grown significantly in the last twenty years. Table 5 shows the decomposition of total financial assets. Panel B indicates that total assets increased from 40 percent of GDP in 1980 to about 2.1 times in 2001. This development is partly explained by the process of capital accumulation experienced by the Chilean economy since the late seventies, in its transition to a new steady-state level of capital, and also by some of the events mentioned earlier, particularly the privatization policy, a drop in discount rates (due to integration), and the implementation of transcendental structural reforms. These have had a noticeable impact in the aggregate value of the stock market. In an international context, Figure 1 shows the result of regressing the average market capitalization to GDP ratio for the period 1990-1997 against per capita income in 1992 and the growth in per capita income during the period 1980-1992 for a sample of 65 countries.19 Both variables turn out to be very significant. The graph shows the relative position of Chile, which tends to be way above the regression line, surpassed only by South Africa, Malaysia and Jordan in terms of the regression error. Panel B of Figure 1 repeats the exercise only that it shows GDP per capita explicitly, illustrating the previous result. The development of the Chilean capital market has not been uniform. It has involved important changes in the composition of financial instruments, the relative importance of different security issuers and investors, the development of new regulatory institutions, and in the trading process itself. Three events help explain the change in the composition of financial assets presented in Table 5.20 First, the 1982 debt crisis explains the jump in government debt to around 12 percent of GDP in 1983 (a shock that persists until today) as the Treasury and later the Central Bank assumed responsibility for the bad loans made by the banking sector. The increase in Central Bank debt is also a consequence of the sterilization of the monetary effects of international reserve accumulation. The crisis explains why as late as 1988 (also see Table 6) market value of equity achieved approximately the same value it had eight years before. 19 Data Source: Demigurc-Kunt and Levine (2002), chapter 2. 20 See Eyzaguirre and Lefort (1999). 13 Table 6 presents stock market indicators. It shows that in 1991 the market value of equity nearly doubled with respect to the previous year. The internationalization of the Chilean economy helps us understand this. Lefort and Walker(2002) estimate a 6 percent equity discount rate drop in 1991, which is presumably attributable to a higher degree of integration of the Chilean capital market with the rest of the world. From then on, there was a significant upward trend in terms of market capitalization until 1998, the year of the Russian crisis. Indeed, in August of that year, the IGPA equity price index had fallen 42.5 percent with respect to the previous year's ending value. Notice also that in 1998, the dividend yield and the book-to-market ratio, interpreted here as a cost of capital indicators, approached the levels they had in the early nineties. The numerator of the dividend yield corresponds to the historical dividends, but even if we substitute the historical dividend for the ex post dividend of the following year (assuming perfect foresight) we obtain similar results. The yield calculated in this way for 1998 is 4.6 percent, similar to its value in 1992 but higher than 1991. Thus, 1998 became a very expensive year for issuing new equity. Equivalently, expected returns were extremely high due to a high risk premium.21 Towards de end of 2001 the dividend yield became even larger. However, dividends that year were exceptionally high,22 and if we look at market-to-book ratios we observe that in 2001 the ratios value was about the same as the previous year’s, which is likely to imply that the risk premium remained relatively high. B EVOLUTION BY INVESTOR TYPE B.1 Pension funds Table 7 shows the evolution of the financial assets held by different investors by investor type. The most notorious increase in relative size comes from the pension funds.23 They have accumulated a sizeable 35 billion dollars in assets (approximately 50 percent of GDP). Table 8, panels A and B respectively show how these funds have been invested and the relative importance of each asset class in the context of the size of its corresponding market. Roughly, in 2001 1/3 was invested in government securities, another 1/3 in the banking sector, 1/5 in stocks, bonds and investment fund shares of local issuers and the rest 21 Dividend level volatility is low when compared with the price volatility. The latter explains most of the volatility of the dividend yield. Thus, it becomes a good predictor of future returns. See Walker (2000). 22 That year COPEC paid out an extraordinarily large cash dividend after selling its participation in GENER. 23 See Walker and Lefort (2001) for an analysis of the effects of pension fund reform on capital market development. 14 abroad. The most salient features of the recent evolution of the pension fund investments are the reduction in local equity accompanied by higher foreign investment. The former is mostly explained by the tender offers placed by foreign companies that took over control of certain large local firms. The latter is probably not independent of - and in part was a reaction to - the Asian crisis, during which local managers bet against the Chilean peso. Panel B of Table 8 indicates that approximately 60 percent of all existing government bonds, 50 percent of mortgage bonds and 35 percent of corporate bonds are held by pension funds. In addition, they hold 12 percent of all time deposits and 6.6 percent of the equity stock. This helps understand the importance of such investors. Looking at information from Chile, Argentina and Perú, Walker and Lefort (2001) find evidence that supports several positive effects associated with Pension Fund Reform. It facilitates the accumulation of "institutional capital" through an adaptive legal framework, increased specialization in the investment decision-making process, it promotes transparency and integrity (particularly via the mandatory risk-rating process), and also through a new corporate governance equilibrium, particularly in Chile, which implemented the reform first. Regulations have pursued the development of a capital market in which pension funds could efficiently channel retirement savings, and the government has developed appropriate supervisory institutions that control the compliance with the new regulatory framework. For instance, limits on types of securities and issuers have been imposed based on risk ratings, portfolio diversification, ownership concentration, and other specific criteria, forcing pension funds to discriminate in favor of less concentrated corporations. Also, pension funds usually contribute to the boards of large corporations with at least one board member that tends to represent the minority interest.24 Walker and Lefort (2001) claim that one of the main channels for pension fund reform to affect economic performance is through a reduction in firms' effective cost of capital. The econometric analyses, with both time-series and panel data estimation techniques, indicate that dividend yields are lower (stock prices and price-to-book ratios larger) after the reform. The evidence also favors the hypothesis of lower security-price volatility after reform and of positive incentives to the creation of new financial instruments (in Chile, local investment funds are an example; see B.3 below). In this context, an 24 See Iglesias (2000). 15 adaptive institutional environment has played a central role. One interesting result however is that there does not seem to be bank "disintermediation" although after reform the role of banks changes. Altogether, perhaps one of the most important consequences of implementing the pension reform is an improvement in the allocation of funds for investment purposes, which should translate into a better resource allocation. This may have permanent positive effects on growth and welfare, even if total savings are not affected. B.2 Insurance companies The second salient feature of Table 7 is the fast growth in the size of the funds managed by insurance companies. This is also associated with pension reform since these companies get part of the pension funds when people retire, insuring pensions for life. Insurance companies invest primarily in long-term fixed-income instruments. Given the asset-liability matching requirements, the growth in funds managed by insurance companies implies a larger supply of long-term funding. B.3 Investment funds Local mutual funds and investment funds both have increased in size. In the case of the former, it is the fixed income funds that have become larger and recently, international investment funds. Local equity mutual funds are rather small.25 Local investment funds correspond to closed-end funds that have been created for the pension funds. The largest inflow into these funds occurred in 1995, during which pension funds were constrained by their investment limits and had few investment choices, which partly explains the phenomenon. Finally, foreign capital investment funds were created in the late eighties, as a consequence of the debt-for-equity swap program implemented by the government. There were significant benefits (including tax credits) associated with their creation. These investors seem to have completed their cycle and currently are in the process of partial liquidation, which explains the steady decline in their value. 25 See Maturana and Walker (2000). 16 B.4 Foreign investors The entrance of large foreign investors has increased competition and demanded more sophisticated financial markets. Competition has taken the form of ADRs. In the recent years, ADRs became more important equity investors than pension funds. As of December 1996, total ADR holdings were 6.4 billion US dollars, 0.5 billion more than pension funds. The number of firms issuing ADRs increased steadily and more than 60 percentof firms included in the selective index IPSA were cross-listed as ADRs in the US. These numbers may be indicating that most foreign investors prefer to buy American depository receipts rather than Chilean stock directly. Among possible reasons for this preference we highlight more liquidity and better information and disclosure practices in the US. Firms cross-listing in the US also have to go through the standard process of creating a prospect, standardizing accounting practices and disclosing additional relevant information to investors. Disclosing practices are not only important from a foreign investor’s point of view, but also for local investors.26 Other disadvantages of Chilean capital markets have been the capital gains tax and capital controls, although the one-year “minimum residence” requirement and the “encaje” are currently suspended and capital gains taxes eliminated for foreigners.27 These explanations point to permanent effects, but the recent evolution in ADR holdings shows a dramatic fall. The level in 2001 was 37 percent of the maximum value achieved in 1996. There is a price effect, but even if we use December 1997 prices, ADR holding haved fallen 34 percent since then. Pension fund holdings of local equity were 3.75 billion dollars in 2001, becoming again more important than ADRs. B.5 Banks and others Although the relative size of this "residual" has decreased, its absolute size is still very significant, since it represents more than once GDP (see Table 7). 26Perhaps this is why Saens (1999) finds a 5% and 9% cumulative average-abnormal return for firms announcing ADRs issues in windows of 3 and 21 days respectively. 27 See Eyzaguirre and Lefort (1999) for a discussion of the macroeconomic implications of capital controls in Chile. 17 B.6 Cross-country comparisons Figure 2 shows a Kernel fit of the average asset size of institutional investors to GDP for the period 1980-1995 against per capita income in 1992.28 The figure shows a nonlinear increasing relationship between both variables. In such a context, the figure for Chile in 2001, with private sector assets of about 1.8 times GDP and per capita income of less than 5,000 represents a very notorious outlier. While this requires further explanation, it is likely to be related with the “concurrent conditions” pointed out in section II.. Levine and Carkovic (2002) arrive to a different conclusion. They use traded volumes and the size of the banking sector relative to GDP as measures of financial development, and relate them with growth. Instead, here we see that if we consider all institutional holdings Chile is indeed an outlier in terms of financial development relative to GDP per capita. The most important explanation for this is what we can denominate as the “securitization” of pension-related liabilities. Indeed, pension funds and annuity insurance companies represent about 80 percent of GDP. V. RECENT EVOLUTION OF NEW ISSUES AND TRADED VOLUMES IN CHILE A central question to this paper is whether domestic capital markets in small emerging economies have been drying up. We would also like to know the causes and consequences of such a decrease in the relative importance of local capital markets. In this section we look at two market indicators that in many economies and especially in Chile show clear signs of stagnation: new issues and traded volumes. We find that while traded volumes show an important reduction, in the case of new issues there appears to be a substitution of bond issues for equity financing. We also propose explanations for these phenomena. In the case of new issues we look for market timing evidence which would be related to changing financing costs after the Asian crises. In the case of traded volumes we look at the relationship between ownership structure and trading of securities in the Chilean market. A. EVOLUTION OF NEW ISSUES 28 Data Source: Demigurc-Kunt and Levine (2002), chapter 2. The variables chosen were bank assets, pension fund assets, mutual fund assets and insurance company assets as a fraction of GDP. 18 Figure 3 and tables 9 and 10 present evidence regarding new issues. Panel A of Figure 3 shows bond and stock issues registered with the SVS during the period 1987-2001, and the fraction they represent of total gross capital formation. This proportion has ranged between 10 and 30 percent in recent years, which represents a significant fraction. Bond issues have usually not been important. This may partly be due to the absence of tax advantages to issuing debt and to the fact that for a long time the Central Bank kept local interest rates above their (free-capital-flow) equilibrium levels, proof of which was the need to sterilize large capital inflows. The high interest rate hypothesis has some support based on the evidence presented in Panel B of the same figure, where a negative elasticity with respect to the level of the interest rate seems to exist. We also see in 1999, 2000 and 2001 increasing bond issues associated with decreasing local real (UF-based) interest rates. So we observe that even though until 1997 the bond market was only exceptionally used by firms, the existing institutional settings allowed for a rapid expansion of this market. The evidence regarding bond issues also seems to be consistent with the hypothesis that firms attempt to “time the market” with their financing sources.29 However, we do notice an exception during 1998, with a relatively large bond issue, despite the fact that interest rates were high. Perhaps the explanation can be found in the alternative source of funding, stock issues. That same year the book-to-market ratio (which is assumed to represent the equity cost of capital) was the highest after 1990, and there were very few issues. Panels A and B of Figure 4 considered jointly support the idea of a negative elasticity of new share issues to the book-to-market ratio. The negative relationship is apparent if we exclude the privatization years, particularly 1987. Table 9 shows complementary evidence of Initial Public Offerings (IPOs), which are a subset of the new stock issues. We also find a positive relation with the stock market price level. Not surprisingly, after 1997 the have been no new IPOs. Finally, there also seems to be substitution between debt and equity (with a correlation of –0.4). For example, in 2001 the costs of issuing equity seemed relatively high, judging by the book-to-market ratio relative to its history, and the costs of issuing long-term bonds in local currency (UF) seemed unusually low. Thus the large bond issues, both in relative and absolute terms. 29 See Baker an Wurgler (2002). 19 This idea of substitution can probably be generalized to alternative debt sources. This is also consistent with the market timing hypothesis. Given the exceptionally high bond issues after the 1998 crisis, it is interesting to study the declared or intended uses of this funding source. Table 10 summarizes our findings. In 1998, the crisis year, which seemed expensive to raise outside financing, most of the proceeds were supposedly used to finance investment. In the post-crisis years, most of the external financing was used to repay existing debt. We conjecture that in 1998 and 1999 firms heavily used bank debt to finance their way through the crisis (see the evidence regarding conglomerate financing sources), and during the subsequent years bonds were issued in order to replace this source of funding, attempting to take advantage of the prevailing lower interest rates. The above results illustrate a feature that may be desirable in local financialmarkets, which is to provide a degree of flexibility in the funding sources. C. TRADED VOLUMES Table 11, panels A and B show the evolution through time of traded volumes. If we only look at the totals, the numbers indicate that a peak occurred precisely during the Russian crisis year. Two separate effects explain the large increase with respect to the previous year: foreign exchange transactions more than doubled and "financial intermediation" increased in nearly 50 billion US dollars. Both are precisely a consequence of the crisis: firms urgently needed liquidity (this explains the latter) and there were significant bets against the local currency (which explains the former). These are important services provided by the capital market. A more detailed analysis of Table 11 shows other interesting results. The year 1998 was a clear low in terms of the traded volume in stocks while 1999 seems to show a recovery. However, this figure is significantly affected by the tender offers that took place during that year. Discounting that effect, 1999's volume is below 5 billion US dollars, very similar to the previous year. Confirming this reasoning, in the subsequent years traded volumes dropped further, to near 4 billion. Thus, the evidence indeed may indicate that the stock market has dried up. However, it is important to keep in mind that traded volumes tend to be low when returns are low and when overall economic activity slows down. To 20 illustrate this, Figure 4 correlates annual changes in traded volume as a fraction of GDP and subsequent GDP growth. There is a significant positive linear relationship between these variables. Thus, part of the lower trading volumes may be explained by the overall slow- down of the Chilean economy. Panel B of Table 11 also shows that during the last three years the importance of PRC trading (indexed bonds issued by the Central Bank) was reduced and partly substituted by PRD (dollar denominated bonds issued by the Central Bank). However, it is hard to understand why in 1999 less than half of the previous year's PRC volume was traded. Considering zero-coupons also, traded volumes in UF-denominated bonds have been steadily falling since 1997, being only partially replaced by dollar-denominated bonds. Furthermore, the bid-ask market for the 8-year PRC disappeared in 1998. The mortgage bond market, however, has remained relatively active, by recovering its 1997 levels. In any case, it is interesting to notice that in 2001 the trading volume in bonds issued by the private sector surpassed that of government bonds. In an international context, using data from Demigurc-Kunt and Levine (2002) we find that the Chilean stock trading volumes relative to GDP appear to be quite low. In 1997 trading volumes were similar to that of the lower middle income group. This is presumably explained by ownership structure characteristics (explained below). Levine and Carkovic (2002) and others notice this. They argue that trading volume is more highly correlated with growth than other capital market development indicators, so Chile stands out as having grown quickly despite of this. In our discussion on the next section we argue that, despite this empirical regularity, there probably are other indicators that better represent the usefulness of capital markets in order to finance growth. Chilean traded volumes, especially, in equity may be low for other more structural reasons. Here we look at a straightforward hypothesis. That is whether the high degree of ownership concentration explains low traded volumes. Lefort and Walker (2000c) show that the degree of ownership concentration in non financial listed firms increased during the nineties. Therefore, a more structural explanation of the decreasing traded volumes may be related to corporate structure. It is unlikely that ownership structure of Chilean companies will change rapidly over the next few years. Therefore, if a high degree of ownership 21 concentration implies low traded volumes we do not expect this sign of dryness to change in the near future. We use corporate data from 1990 to 2000 of 140 non financial listed companies in Chile. We regressed yearly traded volumes for such companies on ownership concentration, controlling by size, pension fund and ADR presence as investors, and year, industry and group dummies. Results are presented in Table 12. We find that size, pension funds and ADRs positively affect traded volumes. Moreover, firms affiliated to conglomerates also present higher traded volumes. However, as hypothesized, when the share of company equity controlled by the group’s controller increases, traded volumes importantly decrease. Therefore, we find that the drop in traded volumes experienced by Chilean companies is also explained by structural factors that we only expect to change slowly on time. VI. SUMMARY AND DISCUSSION: THE ROLE OF LOCAL FINANCIAL MARKETS IN SMALL OPEN ECONOMIES In this paper we have reviewed the evolution of the Chilean financial markets and conglomerates. In doing so we have looked at the different reforms and events that have accompanied this process. We have argued that the macroeconomic reforms, the debt crisis, the resulting banking, security and corporation laws, the tax laws, the privatization rounds and the pension reform each have had a distinct and important impact on the development of the Chilean capital markets. On the other hand, the same events have been important in shaping conglomerate structures. This has meant a significant development of capital markets and growth in conglomerate sizes, both in absolute terms and relative to GDP. In this context of apparently steady growth and development, two distinct facts appear: the 1998 Asian crisis, with the consequent shrinkage of the traded volumes, and the wave of takeovers. These events probably are related to one another given the relatively low market value (“cheap prices”) achieved by firms during 1998. A. ROLE OF LOCAL CAPITAL MARKETS Capital markets are expected to provide two kinds of services: funding of the “appropriate kind”, including short-term liquidity; and risk sharing. How may this be 22 different for a relatively small economy open to international capital flows? What becomes the role of local capital markets if firms can raise funds abroad, either via equity or debt? We hypothesize that local capital markets can better provide some services because of considerations such as currency matching, scale factors, transaction costs and the information asymmetries involved in providing these services from abroad. Different kinds of funding may be required over the different stages of the business cycle. A first role that we identify for local capital markets is that of an external crises "shock absorber". According to Caballero (2001), crises typically happen after terms-of- trade shocks. Then, flows to emerging markets tend to stop, or even to be reversed, which produces a short-term liquidity crunch. Thus, the liquidity service that can be provided by local capital markets is most valuable during these crises. It is presumably too expensive for foreign capital markets to provide short-term liquidity services, especially in local currency. Even after an emerging economy successfully achieves domestic macroeconomic stability, such as the case of Chile, the main source of business cycle variability will arise from external shocks. In the medium-term after the initial shock, local firms will increase their need of external finance in order to reduce the real consequences of the shocks. However, since these external shocks affect the risk premia required to emerging markets, equity financing will appear to be particularly expensive. Issuing bonds and selling them abroad mayalso seem relatively expensive. Thus, given the evidence that firms engage in market timing of their financing sources, local firms will turn to the domestic capital markets, especially to banks (given less information asymmetry in this case), in order to get the required funding. If domestic capital markets are not well developed, the lack of domestic funding may deepen the downturn in real economic activity. Thus, in this sense, local capital markets still have an important role to play in the context of the global economy, although possibly redefined Second, domestic capital markets are better suited to provide funding denominated in local currency (or linked to the domestic CPI, for example) due to asymmetric risk assessments of the currency unit. The expected bankruptcy costs and risks associated with debt issued in local and foreign currency may be quite different, both from the perspective of investors and firms. For instance, for a local worker, a fixed income, CPI-indexed, 23 government-guaranteed instrument is as close as it gets to a risk-free asset. This is not necessarily true for the average foreign investor. Also, from the perspective of firms that operate in the non-tradable sector of the economy, issuing debt denominated in foreign- currency instead of local currency will increase expected bankruptcy costs. Thus, there will be welfare-enhancing and/or risk-reducing contracts between people from the same country or region that would probably not exist between local and foreign agents under similarly favorable conditions. Third, from the perspective of facilitating firm financing, we distinguish between large and small companies. In the case of large firms, many will have direct access to foreign financing. However, local and foreign funding are not always perfect substitutes. When it comes to raising equity, and after firms have gone through the process of issuing ADRs, local and foreign financing are indeed perfect substitutes. The same is true for debt, but only if currency considerations are unimportant. In the case of smaller firms, by virtue of their size and the fixed costs involved in issuing securities abroad, it is likely that they will remain financed by local capital markets, in terms of equity and debt, using local or foreign currency. Fourth, domestic capital markets play an important role reducing information asymmetry. Young and riskier firms that require external funding will probably have to go through a screening process performed by the domestic market before they have access to international capital markets. In this sense, we can think of an international "pecking-order theory" perspective. Myers' (1984) pecking-order theory establishes that due to information asymmetries, the preferred financing sources will be those that minimize the expected losses due to it, from the perspective of the issuer. Therefore, young and unknown firms will use local debt, primarily short-term (perhaps bank-) debt. As the information asymmetry is reduced, these firms may go on to issuing equity-like instruments. If this theory is valid internationally, and if cross-border information asymmetry is greater, a period of acquaintance with the local market may be necessary before firms go on to issue securities abroad, such as Yankee bonds and ADRs.30 In the latter two cases, local financial markets can be understood as the "nursery" of 30 Also see Guiso, Sapienza an Zingales (2002) who find that find that financial development enhances the probability an individual starts their own business, favours entry, increases competition, and promotes growth of firms. These effects are weaker for larger firms. 24 younger/smaller firms. However it is also true that new information technologies and remote trading systems imply that distribution and information costs will be reduced over time, even for smaller/younger firms. The point is whether a significant degree of cross- border transaction costs and information asymmetries will remain even after this new technology is in place in order to justify this role. Our hypothesis is that at least certain information asymmetry, perhaps related with the idiosyncrasies of the different countries, will remain over time. Finally, regarding risk sharing, simple diversification arguments allow us to conclude that global capital markets may be much more effective for spreading out certain investment risks. However, the benefits associated with sharing these risks internationally have to be balanced against scale economy arguments and currency considerations. In the case of smaller firms, the costs associated with cross-border equity issues may exceed the benefits of international risk sharing. The second point is that "risk" may not have the same meaning for local and foreign investors. As discussed, a long-term CPI-indexed, government-guaranteed instrument may be fairly similar to a risk-free asset from the perspective of a local pensioner, for example, but not from that of a foreign investor. Similar arguments apply to the case of equity. For example, utilities may provide stable and relatively predictable future real cash flows measured in local currency. The risk assessment of such security will depend on the perspective of the investor. As a consequence, we may expect clientele effects for the different kinds of securities. This is a possible explanation for the home-equity bias documented in the literature for institutional stock investments. B. HAS THE CHILEAN FINANCIAL MARKET PLAYED THE ROLE IT IS SUPPOSED TO? The evidence we have reviewed does indicate that the Chilean financial market has provided some of the services it is expected to. During the crisis it satisfied liquidity and speculative requirements. From the perspective of firms that could not postpone the raising of new capital, it gave the opportunity of issuing debt instead of equity. The evidence regarding conglomerate's financial structure also is consistent with this, indicating greater bank financing in the crisis year. In the following years, we observe a notorious increase in local currency denominated (UF) bond issues and a sharp drop in ADR holdings, which 25 again supports the idea of substitution among financing sources. This illustrates that there is not necessarily a one-to-one relationship between liquidity (or market depth) and the ability of financial markets to provide valuable services. It may not be liquidity that is required as much as the option of obtaining appropriate financing, when needed. The lack of market liquidity does imply an extra cost for the issuer that is deducted from the price paid by investors. Nevertheless, the market impact factor due to the lack of liquidity of the Chilean market is estimated to be 23 basis points (see Table 13), which seems relatively small.31 Regarding the role of local markets as a “Nursery” of younger or smaller firms, the evidence does show significant IPO activity until 1997. In only one case, which corresponds to a relatively large firm (DHS), ADRs were issued directly without previously listing the stock in the domestic market. In any case, the evidence does seem to indicate a structural break in 1998. Not only the stock market but also the longer-term fixed income market became less liquid. The reduced liquidity of the stock market may be related with the takeover wave, but even before 1998 liquidity indicators were relatively low. Why did the growth rates in traded volumes decrease after 1995? Can the Chilean capital market attain a faster rate of development? Is the lack of development related with the takeover wave? These are related questions. Trading volumes, stock issues, stock returns and future GDP growth rates all are positivelyrelated variables. They are also negatively associated with the risk premia required to a country. The evidence indicates that after 1998 risk premia have been high and expected growth rates low. So one possibility is that we are observing a transitory phenomenon, that will be reverted once the growth rates and the risk premia revert to more normal levels. Yet an alternative interpretation for the endemic low trading volumes and the high property concentration, that also helps explain the 70 percent excess price paid during the takeover process of local firms by foreign ones, is that there is weak investor protection. The above premia would thus correspond to private benefits of control, which in turn may represent the present value of future expropriations to minority shareholders and other 31 For example, compared with a situation in which there is no market impact and a required rate of return of 15%, considering 23 bp of market impact the latter would become 15.035%. 26 stakeholders.32 Property remains concentrated because of rent protection and investors do not participate more in the capital markets because of a lack of trust. According to Shleifer et.al. this may be enough to explain a lack of capital market development. The above conclusion would be somewhat surprising given that the Chilean legal principles emulate those of the US. Thus, if true, in the Chilean case the problem must be identified to be the interpretation and enforcement of the law. But this is not obvious. The sharp differences in ownership concentration and market liquidity between Chile and other countries may be attributable to factors other than the legal protection of investors, and therefore more closely related to the past Chilean economic history, specifically, to the events described earlier in this paper. There is no convergence, only path dependence. REFERENCES Barclay and Holderness (1989), “Private Benefits from Control of Public Corporations”, Journal of Financial Economics, 25 Baker, Malcolm and Jeffrey Wurgler (2002). Market Timing and Capital Structure. Journal of Finance, fothcoming. Bebchuk, Lucian (1999). “ Separation of Ownership and Control”. Working paper Harvard Law School. Claessens, Stijn, Simeon Djankov and Leora Klapper (1999). “The Role and Functioning of Business groups in East Asia and Chile”. ABANTE, Vol 2, Nº2 Demirgüç-Kunt, Asli and Ross Levine (2002), editors. Financial Structure and Economic Growth, MIT Press. Dietl, Helmut (1998). “Capital Markets and Corporate Governance in Japan, Germany and the United States”, Routledge. Eyzaguirre, Nicolás and Fernando Lefort (1999). “Capital Markets in Chile: A Case of Succesful International Financial Integration”, in Chile Recent Policy Lessons and Emerging Challenges, WBI Development Studies. Ed. G. Perry and D.M. Leipziger. 32 Still, it is possible that foreign investors looked at the “long term” expected returns of their eventual acquisitions, concluding that the 1998 downturn in market values was a propitious opportunity for buying local firms, and were willing to pay back a fraction of the recent price drops. The level of rationality assumed in this argument is debatable, and represents another attempt at timing the market. 27 Guiso, L., P. Sapienza and L. Zingales (2002). Does Local Financial Development Matter? CEPR Discussion Paper No. 3307, March. Hachette, Dominique y Rolf Lüders (1992). La Privatización en Chile. Centro Internacional para el Desarrollo Económico, CINDE, Chile. Hernández, Leonardo and Eduardo Walker (1993). “Estructuras de financiamiento corporativo en Chile (1978-1990)”. Estudios Públicos Nº6, Invierno. Holderness and Sheehan (1988), “The Role of Majority Shareholders in Publicly Held Corporations”, Journal of Financial Economics, 20 Iglesias (2000). ABANTE. Khanna, Tarun and Krishna Palepu (1997). “Why focused strategies may be wrong for emerging markets” HBS, July – August. Khanna, Tarun and Krishna Palepu (1998). “The future of business groups in emerging markets: Long run evidence from Chile” HBS, Research Working Papers. Khanna, Tarun and Krishna Palepu (1999). “ Policy Shocks, Market Intermediaries and Corporate Strategy: The Evolution of Business Groups in Chile and India”, Harvard Business Review. Khanna, Tarun and Krishna Palepu (2000). “ Is group affiliation Profitable in Emerging Markets? An Analysis of Diversified Indian Business Groups” Journal of Finance, forthcoming. La Porta, Rafael, Florencio López-de-Silanes, Andrei Shleifer and Robert Vishny (1996). “Law and Finance”, NBER WP 5661. La Porta, Rafael, Florencio López-de-Silanes, Andrei Shleifer and Robert Vishny (1997). “Legal determinants of External Finance”, The Journal of Finance, July. Larraín, Felipe and Patricio Meller (1990). “La experiencia socialista-populista chilena: la Unidad Popular, 1970-1973.” Cuadernos de Economía 27, December, Nº82, p. 317- 356. Lefort, Fernando and Eduardo Walker (2000a), “The Evolution of Chilean Conglomerates”, ABANTE, Vol 2, Nº2, October. Lefort, Fernando and Eduardo Walker (2000b), “Ownership and Capital Structure of Chilean Conglomerates: Facts and Hypothesis for Governance”, ABANTE, Vol 3, Nº1, April. 28 Lefort, Fernando and Eduardo Walker (2001), “Gobierno Corporativo, Protección a Accionistas Minoritarios y Tomas de Control”, Documentos de Discusión 1, SVS, May. Lefort, F. and E. Walker (2002). Cambios estructurales e integración. Discusión y análisis del mercado accionario chileno. Cuadernos de Economía (Abril), Año 39, No. 116, p. 95-122. Leland, H.E. and David Pyle (1977), “Informational Asymmetries, Financial Structure, and Financial Intermediation”, Journal of Finance, 32(3). Levine, Ross and María Carkovic (2002). “Finance and Growth: New Evidence and Policy Analysis for Chile”, Central Bank of Chile Working Paper 157. Majluf, Nicolás, Nureya Abarca, Darío Rodriguez and Luis Arturo Fuentes (1998). “Governance and Ownership Structure in Chilean economic groups”. ABANTE, Vol 1, Nº1, April Maturana, Gustavo and Eduardo Walker (2000). “¿Crean Valor los Fondos Mutuos Chilenos?” Mimeo. Megginson (1990), “Restricted Voting Stock, Acquisition Premiums, and the Market Value of Corporate Control”, The Financial Review, 25, for England. Paredes, Ricardo and José Miguel Sánchez (1994). “Grupos Económicos y Desarrollo: El Caso de Chile”. Ilades-Gerogetown. Mimeo. Rajan, Raghuram, Henri Servaes and Luigi Zingales (2000) “The Cost of Diversity: The Diversification Discount and Inefficient Investment” Journal of Finance, forthcoming Scharfstein, David and Jeremy Stein (1997) “The Dark Side of Internal Capital Markets: Divisional Rent Seeking and Inefficient Investment” NBER working paper Nº5969. Scharfstein, David (1998) “The Dark Side of Internal Capital Markets II: Evidence from Diversified Conglomerates” Working Paper, MIT Sloan School of Managment. Shleifer, Andrei and Robert W. Vishny (1997). “A Survey of Corporate Governance”. Journal of Finance, Vol. 52, Nº2, June. Shleifer, Andrei and Simon Johnson (1998) “Coase and Corporate Governance in Latin America” ABANTE, Vol 2, Nº2. Superintendencia de Valores y Seguros (1999). Legislación Mercado de Valores. 29 Tarziján, Jorge (1999). “Internal Capital Markets and Multimarket contact as explanations for conglomerates in emerging markets”. ABANTE, Vol 2, Nº1, April. Valadares, Sílvia Mourthé and Ricardo Pereira Câmara Leal (1999), “Ownership and Control Structure of Brazilian Companies”. ABANTE, Vol 2, Nº2, October. Valdés, Salvador (1988). “Ajuste Estructural en el Mercado de Capitales: La Evidencia Chilena”. Mimeo. Valdés, Salvador (1992). “Financial Liberalization and the Capital Account: Chile, 1974- 1984”. Mimeo.Walker, E. (2000). "Predictibilidad de rendimientos en el mercado accionario chileno". Forthcoming in Trimestre Económico. Julio-Septiembre. Walker, E. and F. Lefort (2001). Pension reform and capital markets: Are there any hard links? Social Protection Discussion Paper 0201; Pension Reform Primer Paper, World Bank. Walker, E. (2002). The Chilean Experience in Completing Markets with Financial Indexation, in Indexation, Inflation and Monetary Policy, F. Lefort and K. Schmidt- Hebbel, eds., Central Bank of Chile. Zingales, Luigi (1997). “Corporate Governance”, NBER WP 6309 Zingales (1995), “What determines the Value of Corporate Votes?, Quarterly Journal of Economics, 110 Zingales (1994), “The Value of the Voting Right: A Study of the Milan Stock Exchange Experience”, Review of Financial Studies, 7 30 -5 -4 -3 -2 -1 0 1 -3.0 -2.5 -2.0 -1.5 -1.0 -0.5 0.0 FITTED A C TU A L Figure 1 A. Log Market Cap to GDP (Average 1990-1997) versus Per Capita Income (1992) and Growth CHL -5 -4 -3 -2 -1 0 1 0 5000 10000 15000 20000 GDP_CAP92 LO G (M K TC A P _G D P ) B. Log Market Cap to GDP (Average 1990-1997) versus Per Capita Income (1992) Kernel Fit (Epanechnikov, h= 2551.8) CHL 31 0.0 0.5 1.0 1.5 2.0 2.5 3.0 0 5000 10000 15000 20000 GDP_CAP92 TO TA L A SS ET S / G D P Figure 2 Total Asset Size to GDP (Average 1980-1995) versus Per Capita Income (1992) Kernel Fit (Epanechnikov, h= 2551.8) 32 Figure 3 New Issues and Financing Cost Proxies Panel A Panel B 0 500 1.000 1.500 2.000 2.500 3.000 3.500 4.000 4.500 19 87 19 88 19 89 19 90 19 91 19 92 19 93 19 94 19 95 19 96 19 97 19 98 19 99 20 00 20 01 Is su es (U S$ M n) 0 5 10 15 20 25 30 35 40 45 To ta l I ss ue s / G ro ss C ap F or m at io n BONDS (a) STOCK ISSUES TOTAL ISSUES / GROSS CAPITAL FORMATION (%) 0,4 0,6 0,8 1,0 1,2 1,4 1,6 19 87 19 88 19 89 19 90 19 91 19 92 19 93 19 94 19 95 19 96 19 97 19 98 19 99 20 00 20 01 B O O K / M K T 4,0 4,5 5,0 5,5 6,0 6,5 7,0 7,5 8,0 8,5 9,0 IN T R A TE BOOK/MKT AVG CENTRAL BANK RATE 10 Yrs (b) 33 Figure 4 GROWTH RATES VERSUS CHGES IN STOCK TRADING VOLUMES/GDP GDP GWTH(t) = 0,0513xChg Trd Vol(t-1) + 5,23 R2 = 0,4176 -2 0 2 4 6 8 10 12 14 -50 0 50 100 Chges Trd Vol/GDP t,t-1 (%) G w th t+ 1 (% ) 34 TABLE 1 PYRAMID SCHEMES* Corporations Corporations Corporations Corporations Level 1 Level 2 Level 3 Level 4 1990 93 13 0 0 87,70% 12,20% 1994 124 45 2 1 72,00% 26,60% 1,10% 0,50% 1998 96 40 5 2 67,10% 27,90% 3,50% 1,40% * Source: Lefort and Walker (2000). 35 Conglomerates US$ millions Relative size(%) US$ millions % of assets US$ millions %of assets US$ millions %of assets Largest 16.220 23,00 8.629 53,20 6.180 38,10 1.395 8,60 5 largest 37.704 54,00 17.344 46,00 9.577 25,40 10.783 28,60 10 largest 49.357 70,00 22.161 44,90 11.944 24,20 15.202 30,80 20 largest 57.570 82,00 26.309 45,70 13.759 23,90 17.559 30,50 All conglomerates 63.957 91,00 29.868 46,70 14.646 22,90 19.443 30,40 Non affiliated 6.059 9,00 2.587 42,70 224 3,70 3.248 53,60 Total 70.017 100,00 32.488 46,40 14.844 21,20 22.686 32,40 * Source: Lefort and Walker (2000c). Minority shareholders Controlling shareholders TABLE 2 OWNERSHIP AND CAPITAL STRUCTURE OF CHILEAN CONGLOMERATES* Total assets Debt Equity 36 TABLE 3 LEGAL FRAMEWORK OF CHILEAN CAPITAL MARKETS Pension fund laws Insurance laws Corporations law Security markets law Banking law • Matching requirements • Strict reserve requirements • Otherwise quite flexible • Information requirements and financial statements audited under GAAP • Shareholder meetings with cumulative and proxy voting • Board is governing body • Directors represent all shareholders • Limits on related party transactions • Limits by instrument type and issuer • Important role of Risk Rating Committee • Cannot buy underpriviledge d shares • Rights of shareholders stated and protected • Dual shares allowed with restrictions • Tender offer requirement when large premiums offered • Restrictions on Related lending • Unable to hold shares • Matching requirements • Credit risk provisions • Partial deposit insurance • Valuation at market prices Super. de AFP Super. de Valores y Seguros Super. de Bancos 37 TABLE 4 CHILEAN COMPLIANCE UIT OECD PRINCIPLES OF CORPORATE GOVERNANCE 38 I. The Rights of Shareholders 1 Governance framework should protect shareholders’ rights. S 2 Right to participate and vote in shareholder meetings. S 3 Capital and ownership structures should be disclosed. U 4 Markets for corporate control should be allowed to function in an efficient and transparent manner. S 5 Investors should consider the costs and benefits of exercising their voting rights S II. The Equitable Treatment of Shareholders 1 Equitable treatment of all shareholders, including minority and foreign shareholders. S 2 Insider trading and abusive self-dealing should be prohibited. S III. Role of Stakeholders in Corporate Governance 1 The corporate governance framework should recognize the rights of stakeholders. S 2 Stakeholders should have the opportunity to obtain effective redress for violation of their rights. U 3 Performance-enhancement mechanisms for stakeholder participation. U IV: Disclosure and Transparency 1 Timely and accurate disclosure is made on all material matters regarding the corporation. U 2 Financial and non-financial information must be prepared, audited, and disclosed with high quality standards. S 3 Independent auditors must provide an external and objective assurance about financial statements. S V. Responsibilities of the Board 1 Strategic guidance of the company, the effective monitoring of management, and board’s accountability to shareholders. S 2 The board should treat all shareholders fairly. S 3 The board should act independently from management and controlling shareholders. U Satisfactory: 11 (69%) Unsatisfactory: 5 (31%) 39 Date TOTAL Value % Value % Value % Value % 1980 8.071 52,8 806 5,3 1.178 7,7 5.244 34,3 15.300 1981 5.526 35,2 1.559 9,9 836 5,3 7.797 49,6 15.718 1982 5.373 34,8 1.907 12,4 540 3,5 7.608 49,3 15.428 1983 3.076 24,1 1.842 14,4 2.474 19,4 5.374 42,1 12.766 1984 2.969 22,2 2.060 15,4 3.021 22,5 5.352 39,9 13.402 1985 3.217 20,8 2.150 13,9 3.353 21,7 6.724 43,5 15.444 1986 6.161 31,6 1.713 8,8 4.229 21,7 7.405 38,0 19.508 1987 7.574 33,3 1.785 7,8 4.854 21,3 8.530 37,5 22.743 1988 9.327 34,2 2.301 8,4 5.856 21,5 9.769 35,8 27.253 1989 12.574 37,2 2.916 8,6 6.676 19,7 11.672 34,5 33.838 1990 16.098 40,2 3.346 8,4 9.518 23,8 11.041 27,6 40.003 1991 30.962 52,8 3.974 6,8 12.255 20,9 11.442 19,5 58.634 1992 29.683 48,6 4.307 7,1 13.852 22,7 13.229 21,7 61.071 1993 44.883 55,8 4.927 6,1 14.261 17,7 16.370 20,4 80.442 1994 58.581 59,6 6.147 6,2 16.002 16,3 17.634 17,9 98.364 1995 57.367 56,0 7.109 6,9 16.953 16,6 20.956 20,5 102.385 1996 51.948 50,5 8.202 8,0 18.273 17,8 24.468 23,8 102.892 1997 55.308 48,9 9.195 8,1 20.352 18,0 28.330 25,0 113.184 1998 41.057 40,7 9.588 9,5 17.670 17,5 32.658 32,3 100.972 1999 59.096 47,5 10.925 8,8 18.463 14,9 35.815 28,8 124.300 2000 54.204 43,9 11.781 9,5 19.593 15,9 38.000 30,7 123.579 2001 56.734 43,6 15.068 11,6 17.443 13,4 40.971 31,5 130.216 Gross Capital Formation / GDP (6) 1980 1981 1982 1983 1984 1985 17,7 1986 17,1 1987 19,6 1988 20,8 1989 24,5 1990 24,2 1991 22,4 1992 24,7 1993 27,2 1994 27,4 1995 30,6 1996 31,0 26,4 1997 32,2 27,1 1998 32,2 26,1 1999 21,1 2000 21,0 2001 21,4 (1) Numbers are first expressed in ch$ of December 2001 and then transformed into dollars using the exchange rate of the last trading day of December 2001 ch$656.2/US$ (2) Includes both Public and Private sector Stocks,
Compartir