رشد در اندازه بازار سهام و فعالیت های تجاری: یک مطالعه بین المللی

This paper presents new evidence on the role of macroeconomic and institutional factors in equity market development and on the sources of equity market growth. Using panel data on 33 countries, I find that development of financial intermediaries and trade openness are positively associated with equity market size, and that development of financial intermediaries is also positively associated with the level of activity in equity markets. Government consumption is negatively associated with equity market activity. I construct a direct estimate of the effect of institutional factors on equity market development that compares a country's actual level of development to a hypothetical “best-practice” country having the same macroeconomic fundamentals as the original country. I show that the level of equity market development of an average country is around 30% below its maximum potential. There are wide differences in institutional characteristics across countries and over time, and Canada, the United States, and Singapore possess the most shareholder-friendly institutional frameworks that foster larger and more active equity markets. It appears that institutional improvements and changes in financial technology have provided the major impetus for the phenomenal expansion of global equity markets.

Stock markets around the world have experienced phenomenal expansion over the past 30 years. According to Datastream, the aggregate market capitalization of all national equity markets has grown from less than US$1 trillion in 1974 to over US$17 trillion by the end of 1997, and the corresponding annual equity turnover is US$3.4 billion in 1974 and US$11 trillion in 1997. Some countries appear to possess both larger and more active equity markets than others, such as the United Kingdom and the United States. Others appear to have either large and illiquid stock markets such as Jordan, or active but small stock markets such as Germany. Apparently, none of the latter two cases are ideal from a country developmental point of view, and thus it is important to simultaneously examine what factors determine the extent of a country's equity market size and activity. The process of growth in national equity market size and trading activity is still imperfectly understood. One view points to improved macroeconomic and financial fundamentals as the source of the growth. Others, more skeptical of efficiently functioning markets, suggest that the legal and institutional environment of a country matters for the size and extent of a country's capital markets (La Porta et al., 1997 and La Porta et al., 1998). If this is so, how effective are legal and institutional factors in bringing countries to their “best-practice” potential? And is a country's institutional framework structured optimally? So far, the literature has not come up with an exact metric on the role of laws and institutions in affecting national equity market development. In this paper I adopt the stochastic frontier modeling approach to investigate the sources of equity market growth in size and trading activity and to provide a direct estimate of the effect of laws and institutions on equity market development. The stochastic frontier model captures the very simple intuition, that for a given level of macroeconomic conditions a country with an efficient institutional framework is associated with larger and more active stock markets.1 With fully efficient laws and institutions, a country can be observed to be associated with the frictionless neoclassical level of equity market size (trading activity). Imperfect laws and institutions would prevent a country's equity market from reaching its frictionless maximum capacity: they force the actual level of equity market size (trading activity) to be below, but never above, the frictionless neoclassical level of market size (trading activity). In reality, governments often subsidize activities, especially equity markets, as a point of pride. This suggests that at some point in time a country might have larger (and/or more active) equity markets than would be created under frictionless markets with efficient institutions. This possibility is captured by the standard two-sided error in the stochastic frontier model. In this paper, I show how the “best-practice” potential for each country can be estimated in a panel data setting. Specifically, under the stochastic frontier model, a country's equity market size (trading activity) is specified as a linear function of its macroeconomic fundamentals as in a usual regression framework (i.e., the location of the frontier). However, there is a one-sided error term (in addition to the normal error term) in the regression which captures the gap between the actual level of equity market size (trading activity) and its frictionless counterpart if the country possessed the most efficient institutional framework. This gap (i.e., the shortfall from the frontier) can be viewed as a measure of the effect of a country's institutional characteristics on its equity market development, insofar as the frictionless benchmark represents the maximum level of development that would result if there were no institutional impediment. Another useful insight from the stochastic frontier model is that it allows for the decomposition of the change in equity market size (trading activity) into three types of change: macroeconomic conditions, institutional factors, and financial technologies. Under the stochastic frontier framework, the equity market size (trading activity) of a country is determined by its macroeconomic characteristics (fundamentals), institutional factors (efficiency) that affect the gap between the actual level of market development and the “best-practice” potential, and financial technologies. Over time, due to improvements in the institutional framework, a country's equity market can become larger (more active) and catch up to the frictionless neoclassical level. In addition, equity markets grow if a country improves its fundamentals. Finally, general improvements in financial technology that may weaken participation constraints, such as capital market liberalization (Bekaert and Harvey, 2000, Bekaert et al., 2002 and Trzcinka and Ukhov, 2005) and privatization (Perotti and van Oijen, 2001), easy access by (foreign) institutional investors, and stock trading via the Internet, lead to larger equity markets and more activity for a given level of country characteristics. I interpret the increase in equity market size (trading activity), unrelated to macroeconomic fundamentals or institutional efficiency, as the change in financial technology.2 In sum, the stochastic frontier modeling approach allows me to view the growth in equity market size (trading activity) in terms of three different components: reductions in institutional inefficiency, improvements in macro-fundamentals, and changes in financial technology. This framework allows, among other things, for the investigation of (1) the relative roles of the three components of equity market growth, and (2) the manners in which institutional efficiency changes over time and differs across countries. These are important policy questions that the standard regression modeling framework cannot address. Using panel data on 33 countries, I find that development of financial intermediaries and trade openness are positively associated with equity market size, and that development of financial intermediaries is also positively associated with the level of activity in equity markets, while government consumption is negatively associated with activity and liquidity in equity markets. I relate the shortfall from the full efficiency benchmark to measures of the legal and institutional framework, and find that an average country attains a level of equity market development that is around 30% below its maximum potential. Further investigation reveals that countries with French and German legal origins tend to be associated with smaller but more active equity markets. Laws that protect shareholder rights, transparent accounting standards, solid country credit ratings, and greater economic freedom tend to foster a country's equity markets to become larger and more liquid. Overall, there are wide differences in institutional efficiency across countries and over time, and Canada, the United States, and Singapore clearly possess the most shareholder-friendly institutional frameworks. It appears that improvements in institutional efficiency and changes in financial technology have been major factors behind the phenomenal expansion of global equity markets over the past two decades. Many developed countries and some developing countries experience little to moderate improvements in their macro-fundamentals during the sample period. Improvements in institutional efficiency vary across countries, particularly in developing countries. The evidence suggests that it is the countries that have significantly improved the quality of their institutional frameworks experience the fastest growth in equity market size and activity. This paper is motivated by a growing body of research in finance that examines the relationship between laws and institutions and capital markets. La Porta et al., 1997 and La Porta et al., 1998 show that the legal environment matters for corporate governance and the size and extent of a country's capital markets. Demirgüç-Kunt and Maksimovic (1998) find that differences in legal and financial systems affect firms' use of external financing. Dyck and Zingales (2004) show that insiders in French civil law countries possess systematically higher private benefits of control than those in countries of other legal origins, and Leuz et al. (2003) find that companies in Anglo-American countries exhibit less earnings management than their Continental-European counterparts. Lombardo and Pagano (2006) show a positive cross-country correlation between the quality of legal systems and the expected return on equity. Bhattacharya and Daouk (2002) find that the enforcement of insider trading laws reduces a country's cost of equity. From a theoretical perspective, Castro et al. (2004) introduce investor protection into a standard overlapping generation model of capital accumulation, and their model predicts a positive effect of investor protection on growth. Pagano and Volpin (2005) develop a political economy model where there is mutual feedback between investor protection and stock market development. Better investor protection induces companies to issue more equity and thereby leads to a broader stock market. Building on the prior literature, this paper considers an exhaustive list of institutional factors to assess their respective as well as their joint effects on equity market development. The current paper distinguishes itself from the existing literature in several important aspects. First and foremost, this paper adopts a new econometric technique–the stochastic frontier model which captures cross-sectional as well as temporal variation in equity market size and trading–to obtain a direct estimate of the effect of institutional factors on equity market development and to provide insight on the source of equity market growth. Past research primarily employ cross-sectional regressions (La Porta et al., 1997) and more recently regressions with instrumental variables to control for reverse causality and simultaneity (Beck et al., 2003). In this paper, to deal with the endogeneity issue, the lagged (up to 5 years), instead of contemporaneous, determinants of equity market size and trading activity are considered. The stochastic frontier model estimates the frontier parameters based on the cross-sectional information, and their changes (capturing the change in financial technology) are obtained through the temporal variation in equity market size and trading activity. The estimation of efficiency parameters is based on both the temporal and cross-sectional variations. Thus, even if several institutional variables such as legal origin are not time-varying, the cross-sectional variation plus the temporal variation in some other institutional variables (country credit rating for instance) will help identify the model and lead to efficient parameter estimates. In contrast, standard fixed effects models cannot accommodate time-invariant institutional factors in a panel data setting. The new technique allows me to construct an exact estimate of the effect of laws and institutions on equity market development—the shortfall from the frontier. Moreover, the new framework permits me to decompose the total growth in equity markets into three components: improvements in institutions, improvements in macro-fundamentals, and changes in financial technology. Second, my investigation in equity trading from a law and finance perspective is new to the literature, as is my use of some of the macroeconomic variables. Past research, such as La Porta et al. (1997) and Beck et al. (2003), focuses on examining the (historical) determinants of financial development. They employ cross-sectional regressions and adopt measures of financial development covering the banking sector, the stock market and the debt market. In contrast, this paper focuses on the determinants of growth in equity market size and trading activity, and employs panel data sets to explore the impact of macroeconomic shocks on stock prices and trading. In a contemporaneous paper, Claessens et al. (2006) examine how stock market development and internationalization of financial markets are affected by economic fundamentals. Using panel data, they find that higher GDP per capita, law and order, better shareholder protection, sounder macro-policies, and greater openness of capital markets are associated with both more developed local markets and internationalization. Different from Claessens et al. (2006), I consider a comprehensive list of legal and institutional variables, and motivated by the growth literature I also introduce some new macro-variables as fundamentals that could impact equity market development: the level of financial intermediaries in the economy, government consumption, and trade openness. It turns out that these variables are associated with a country's equity market development in an important way. The remainder of the paper proceeds as follows. In Section 2, the modeling framework and decomposition of growth in equity market size and trading activity into its three components are presented. The measures of equity market size and trading activity and list of macroeconomic fundamentals and institutional efficiency factors are given in Section 3. Section 4 presents the empirical results, and Section 5 offers some concluding remarks. Estimation details are provided in the appendix.

This paper contributes to the literature in several major ways. First, it constructs a direct measure of the effect of the institutional framework on a country's equity market development. Second, it finds evidence in support of strong association between some new macroeconomic variables and a country's level of equity market development. Finally, it provides evidence on the sources of growth in equity market size and activity. While considerable effort in the past has been expended on explaining cross-sectional variations in equity market size and activity, the stochastic frontier model adopted in this paper allows me to focus on the temporal component of growth in addition to the cross-sectional relationships. Using panel data on 33 countries, I find that development of financial intermediaries and openness to trade are positively associated with the size of equity markets, and that development of financial intermediaries is also positively associated with the level of activity in equity markets, while government consumption is negatively associated with activity and liquidity in equity markets. Interestingly, openness to trade is negatively associated with equity market liquidity. Also, liquidity in equity markets increases over time. The above results are robust to the use of different lag structures and various control variables. Under the stochastic production frontier model, legal and institutional characteristics affect a country's equity market size and activity through the channel of institutional efficiency. I show that the level of equity market development of an average country is around 30% below its maximum potential. Specifically, French and German legal origins are associated with smaller but more liquid equity markets. Laws that protect shareholders rights, accounting standards that produce high-quality, comprehensive and comparable corporate financial statements, good country credit ratings, and greater economic freedom of a country tend to foster larger and more active equity markets. Overall, there are wide differences in institutional efficiency across countries and over time. It appears that improvements in institutional efficiency and changes in financial technology have provided the major impetus for the phenomenal expansion of global equity markets over the past two decades. Many developed countries and some developing countries experience little to moderate improvements in their market fundamentals during the sample period. Improvements in institutional efficiency vary across countries, particularly in developing countries. The evidence suggests that it is the countries that have significantly improved the quality of their institutional frameworks experience the fastest growth in equity market size and activity.