Do early institutions exert a long-term impact on subsequent institutional and economic development? In particular, do endowments and legal tradition explain the subsequent development of financial markets and rule of law? This dissertation undertakes a historical study of Brazilian equity and corporate debt markets to explore these questions.

Social scientists agree that institutions are important for economic development, but it is hard to determine which institutions are incidental to growth and which ones generate it. One strategy has been to look for exogenous factors that might explain the variation in development around the world today. A common approach has been to investigate whether the conditions at the time of settlement in former European colonies led to the adoption of institutions that had subsequent long-term effects on development. In fact, research based on current economic indicators has shown that there is a strong correlation between certain early institutions and today's development levels. These correlations have been particularly strong in studies that explain financial market development around the world using legal origin as an explanatory variable.

According to the “law and finance” literature, the variation in financial development around the world can be explained by the differences in the legal tradition countries follow. For this literature, legal origin determines both the extent of investor protections and the level of financial market development.¹

Extensions of this literature also show that early conditions such as settler mortality and the proportion of indigenous population to settlers at the time of colonization are powerful explanatory variables of financial development.

All of these studies are explicitly historical in their theoretical setup and, simultaneously, make history irrelevant for their studies because they take data from only the endpoints of their period of study. On the one hand, these authors argue that early conditions determined subsequent institutional and financial developments. On the other hand, these studies do not do historical research. Historical processes are assumed to happen the way they do because institutional variables such as legal tradition or early conditions are highly correlated with current measures of financial development.

This dissertation tests one main hypothesis that related to the implied persistent effect of legal origin. If specific early conditions or legal systems have persistent effects, determining the level of investor protections and financial development, then we should not observe significant variation in these variables over time. For instance, we should not expect to find a country going from having no legal protection for investors to having most of the protections for investors the “law and finance” literature considers relevant. Moreover, we would not expect to find international conditions or domestic political disputes driving most of the changes in legal protections for investors or the variation in financial outcomes over time.

Using detailed historical institutional analysis of the development of financial markets in Brazil, I challenge the implied persistent (path-dependent) effect of early institutions and provide a detailed explanation of the long-term variation in equity and bond market development, shareholder and creditor protections, and court enforcement of corporate bond contracts.

I use original data from the Rio de Janeiro Stock Exchange Archive to show that the long-term trend of equity- and bond-markets size and trading volume in Brazil roughly follows what we observe in other countries and thus is linked to issues that affected most economies after 1913, such as the decrease in international capital flows. Then, I look at the determinants of investor protections and their enforcement to see if they track the variation in outcomes over time. I find no clear causal relationship between investor protections and financial development. In fact, the explanation for the changes in investor protections over time is simple: The interests of the ruling coalitions at different moments shaped different institutions (laws and enforcement) that determined different financial outcomes. There is no strong effect of early institutions on investor protections if we look in detail at the variations throughout history; politics is more important to understanding the variations at the country level.

I argue that Brazil is a good natural laboratory to test the hypothesis that in history there is a great deal of variation in the institutional and market outcomes within countries for three reasons. First, it is a developing civil law country that today has a relatively small financial market. Second, Brazil today has one of the slowest systems of bankruptcy law and has relatively weak investor protections. In the 1990s, Brazil ranked among the worst countries in terms of creditor rights and of shareholder protections. Finally, all the current indices of rule of law today show that in Brazil the enforcement of property rights is weak.

In order to study the change in investor protections over time, I looked at the rights of creditors and shareholders highlighted in the “law and finance” literature. First, I looked at the evolution of the creditor rights that this literature considers most important to understanding the variation in the size of corporate bond markets in Brazil. Creditor rights in Brazil between 1889 and 1940 were relatively strong. Between 1890 and 1908, Brazil in fact had the four protections for creditors that this literature considers relevant for the development of bond markets. I also looked at bankruptcy court cases at the National Archive of Brazil to understand how the enforcement of creditor rights might have differed in the past. I find that there was a strong enforcement of creditor rights in Brazil between 1850 and the first two decades of the twentieth century. Court cases show that bondholders had priority in case of bankruptcy. Once a judge had declared a company bankrupt, creditors were in control of the firm; they could replace managers immediately and recuperate their claims by reorganizing or liquidating the company.

Second, I looked at the protections for shareholders that the “law and finance” literature considers relevant to explaining equity market development. I compiled indices of shareholder rights from joint-stock company laws between 1882 and 2001. I find that shareholder rights changed rapidly over time, getting increasingly better on paper from 1882 and 1976. Then there was a reversal in the number of protections during the 1990s that was counteracted by the 2001 joint stock company law (which provides strong protections for shareholders). Because there is no perfect method for measuring the enforcement of shareholder rights, I followed the “law and finance” methodology and studied corporate ownership concentration in order to ascertain if small shareholders were protected in the past. According to this literature we should observe ownership concentration when shareholder rights were lacking or their enforcement was poor. Small investors would not participate actively in equity markets if they were not protected against the abuses of managers or other shareholders. I find low concentration of ownership before 1940 and a drastic increase in concentration in the following decades. I provide several explanations based on changes of the regulatory regime and the increasing participation of the federal government in the ownership of corporations in strategic sectors. I argue that instead of thinking that early institutions determine legal protections for investors, we should find the explanation for the variation in these protections in the political economy of company laws. Finally, I discuss the protections to minority shareholders inside the firm as a more relevant variable to study ownership concentration.

The findings of this dissertation have important policy implications. I argue that the interests of the members of the ruling coalition determine the legal environment that investors face: how many protections there will be in the law and how much courts and regulatory agencies will enforce them. If ruling coalitions determine the legal environment, which in turn matters for the growth of financial markets, then financial development will ultimately depend on the interests of whoever is in power. This political economy problem relates to what political scientists call the “commitment problem.” This means that any government strong enough to define and protect property rights is also strong enough to abrogate them when such an act is in its interest. Thus, changes in ruling coalitions can complicate the process through which governments commit to protect investors, which can ultimately hurt financial market growth. If politics are so important for financial market development, then the commitment problem can adversely affect economic welfare. This is particularly important if we consider that there are many causal links between financial market development and economic growth.

I end this dissertation by suggesting that socializing the benefits of financial markets is a good way for countries to commit to financial development. If more citizens have a stake in financial market development, then weakening investor protections is more costly for politicians. Workers and middle-class families are more inclined to defend financial markets if they can reap more benefits from them, for example, in the form of mutual and pension funds or through mortgage and debt securitization.