The effect of corporate governance, taxes and ownership structure on dividends and leveragethree essays on the brazilian evidence

Resum

The traditional financial theory proposes that the main objective of a company should be to maximize the wealth of its shareholders, which, in practice, means maximizing the market price of the securities they hold. This objective has generated an abundant literature and a wide controversy about the effects of financial decisions on the value of companies. In this sense, far from the situation of perfect markets analysed by Merton Miller and Franco Modigliani in their seminal works of 1958 and 1961, research in recent decades has shown the relevance of frictions in capital markets such as transaction costs, agency costs, bankruptcy costs or the taxes on the value of firms. In this context, there are many studies that have shown the tax consequences of the company's financial policies. From this perspective, the static trade-off theory establishes that the value of a company will reach its peak when the opposing effects of two forces are balanced: the tax deductibility of interest payments and the increase in bankruptcy costs associated with an increase of the level of debt. Likewise, dividends are also affected by the tax laws of the countries, given the different tax treatment of the distribution of cash compared to its retention. Within the imperfections of the financial markets, deserves special mention the agency theory approach, according to which dividends and debt can be effective instruments to reduce conflicts between the different stakeholders of the company. More specifically, the distribution of dividends or the increase of firm debt are believed to mitigate the possibility of opportunistic behaviour into the firms as dividends and debt reduce the resources under managers’ control and are susceptible to be used in search of their interests and not to maximize the wealth of the investors. Precisely, the agency approach allows to incorporate informational asymmetries between insiders and outsiders, which endows the decisions of dividends and capital structure of a new function. Thus, these financial policies become signals that convey reliable information about the profitable investment opportunities available to the company. A feature of recent research in business finance and corporate governance is the international dimension that these studies have taken, addressing various institutional and legal environments. In this sense, the Law & Finance approach proposed by La Porta et al. (1999, 1997, 1998, 2000), according to which, two types of legal systems can be distinguished: countries whose legal systems come from the Anglo-Saxon or common law tradition tend to offer more adequate protection to investors than countries whose laws they are born in the context of civil or continental law. Taking this approach to the field of dividend policy, there are two hypotheses that explain the relationship between the legal structure of countries and the distribution of profits of companies. The first states that a legal system that offers adequate protection to investors boosts the distribution of dividends as a form of compensation for those shareholders (the outcome hypothesis). The alternative explanation raises that poor legal protection leads companies to distribute higher dividend ratios as a way to make up for the shortcomings of the environment (the substitute hypothesis). Both hypotheses have received considerable empirical support in different places and times. Combining the different theoretical approaches outlined above, it is worth asking whether the incidence of taxes on financial decisions can be modelled by the level of protection enjoyed by shareholders. In other words, the question arises about the relationships between the information transmitted by financial decisions in situations of low legal protection and the effect of taxes on such decisions. All this highlights the convenience of jointly analysing the mechanisms of corporate governance and taxes when investigating financial policies. In parallel with these investigations, there is another line of research that has studied the interaction between the institutional environment and the financial structure of the company, showing how the quality of corporate governance practices influences the relationship between taxes and indebtedness decisions. According to this view, the possible transfers of resources between shareholders and creditors are translated into variations in the flows subject to tax. And, given that the institutional environment modifies the possibility of such transfers, a concatenation is established between the three elements (taxes, legal environment and financial policy) that advise their joint study. Starting from the theoretical foundations previously outlined, this doctoral thesis seeks to perform an analysis of the dividend policy and the capital structure of Brazilian companies. Somehow, paradoxical as it may seem, a doctoral thesis is the result of the doctoral student's individual work and the imprint that his directors, professors and colleagues leave on the Doctorate Program. Therefore, it is not surprising that this doctoral thesis is inscribed in a field of corporate finance broken by professors of the Interuniversity Doctorate Program in Research in Business Economics. When writing these pages, it is inevitable to evoke previous doctoral theses such as that of professors Valentín Azofra Palenzuela on the explanatory factors of the capital structure of Spanish companies, Alberto de Miguel Hidalgo on the interrelation of investment decisions, financing and dividends, Juan Antonio Rodríguez Sanz on the ownership structure or Chabela de la Torre and Ignacio Requejo on the influence of the ownership structure on the company's financial strategy. This means the choice of research topic, with which I have tried to benefit from the knowledge of those who preceded me and, at the same time, take a step forward in the path of knowledge. The choice of Brazil is not capricious but this country is a unique case for the joint study of the tax and institutional effects on corporate decisions since, despite offering relatively little legal protection to minority shareholders and suffering from certain shortcomings in terms of corporate governance, in recent years it has made a considerable effort to improve the quality of its institutional environment while at the same time implementing significant changes in its fiscal order. Both actions have mutual implications and constitute a unique opportunity to advance our knowledge on the subject. Thus, in order to compensate for the deficient protection of minority shareholders the Stock Exchange of the State of Sao Paulo (B3), which is the official Stock Exchange of Brazil, created in 2000 the so-called Novo Mercado (NM). Companies voluntarily joining it are obliged to adopt corporate governance practices that are stricter than those required by national legislation. This process is based on the idea of "functional convergence" proposed by Coffee (1999), according to which, in countries with institutional deficiencies, due to the difficulties of reforming national legislation, it must be the capital markets that take the initiative in adopting government improvement measures. Aiming at granting companies a certain margin of time to gradually improve their operation until they were able to meet the requirements of the NM, the B3 launched the Level 1 and Level 2, which are intermediary categories with stricter requirements of corporate governance than those required by Brazilian legislation and, at the same time, less demanding than those of the NM. In this way, the Brazilian setting constitutes a unique environment for studying the effects of corporate governance on the financial decisions of companies by bringing together companies that, despite being subject to the same institutional framework, are exposed to different requirements of corporate governance. Therefore, the present doctoral thesis prevents the results and conclusions from being biased by institutional and cultural differences between countries. Likewise, the Brazilian case allows us to assess the extent to which a market-level reform that did not have the support of national legislative changes, can replace or amend the institutional deficiencies of the country as a whole. Another institutional aspect in which we deepen in this doctoral thesis is related to the influence of the Brazilian tax legislation in the financial decisions of companies. Specifically, Brazil is the only emerging country that has implemented a system of Allowance for Corporate Equity (ACE), called "interest on equity (IOE)", whose purpose is to allow companies to deduct a notional interest rate for the use of own resources and not only for the use of the debt. A peculiar feature of the IOE is that companies can only enjoy tax deductions by its use when they distribute part of their equity to shareholders. Thus, the IOE is a clear alternative to payout and, in spite of presenting features and functions quite similar to dividends, such as its usefulness to reduce agency conflicts, to signal investment opportunities, to attract shareholders who value the feeling of security, and to provide shareholders liquidity, its tax treatment presents great differences, since, when considered a financial expense, it can be deducted in the balance of the company before the calculation of taxes, thus generating a tax deduction. On the other hand, in the case of dividends, the tax base is determined by the adjusted net profit, without the company being able to obtain tax deductions for the distribution of dividends. The tax treatment of personal income is also different for dividends and for IOE. While shareholders are tax exempt when they receive dividends, they taxed on incomes from IOE and this tax varies according to the type of beneficiary shareholder. Therefore, the tax advantages derived from the use of IOE depend not only on the capital structure of the company but also on its ownership structure. This feature opens the door to consider whether the importance of fiscal considerations in the payout policy may be conditioned by the identity of the company's shareholders. In the same way, the existence of the IOE allows companies to have an alternative non-debt tax shield, with lower bankruptcy costs and that also retains the non-fiscal characteristics of the debt, such as, its signalling effect or its capacity to mitigate agency conflicts. Based on the framework described above, the main objective of this doctoral thesis is to analyse, using the idiosyncrasies of the Brazilian institutional framework, the influence of corporate governance, taxes and the ownership structure on dividends and capital structure decisions of companies. This general objective is specified in two more specific objectives: 1- To analyze if and how the adoption of the NM affects the dividend policy and the capital structure of Brazilian companies. 2- Based on the relevance of taxes in payout decisions and the capital structure of companies, to analyze the moderating role of the ownership structure in said relationship. This sequence of objectives structures this doctoral thesis. Thus, after this introduction, in Chapter 2 we consider the effect of the incorporation into the NM on financing and dividend decisions. This chapter describes the characteristics of the Brazilian institutional environment and the reform undertaken by B3 to improve the protection of investors through the creation of different market levels. Then, we verify if and how the incorporation to the NM and the intermediate levels affects the decisions of dividends and corporate financing, both as mechanisms used by companies to mitigate their agency conflicts. Next, we incorporate the differences in taxation between dividends and IOE to verify the moderating effect of firms’ ownership structure. This analysis is carried out in Chapter 3, in which we detail the tax consequences of the use of dividends and IOE for companies and shareholders and, subsequently, we verify whether and how the identity of the dominant shareholders affects the relationship between taxes and payout decisions. A similar study is undertaken in Chapter 4, in which the effect of the IOE on the capital structure is extended. In this chapter, we expose the similarities between leverage and the IOE and also the differences in their tax treatment, which depend on the ownership structure of firms. Subsequently, in this chapter we analyze the effect of taxes on financing decisions, incorporating the moderating role of the identity of the shareholders in this relationship. This doctoral thesis is closed with a chapter that contains the main conclusions, the results are discussed from a critical perspective and future lines of research are suggested. To facilitate the overview of our research, in Figure 1 we propose a simplified scheme. By way of advancement of our results, we find that Brazilian companies voluntarily incorporated into the NM distribute less dividends and are less leveraged than their counterparts operating in the other segments of B3. This suggests that the NM plays an important role and can be an element that replaces other mechanisms aimed at improving trust between companies and shareholders, namely, dividends and debt. The results also indicate that the incidence of taxes on payout and capital structure decisions depends on the ownership structure of the companies. We have detected that in companies with a predominant presence of individual (or family) shareholders or with pyramidal structures in their chain of owners, taxes seem to be a determining factor in financing or payout decisions. Contrariwise, in companies owned by institutional shareholders, fiscal considerations appear to be irrelevant in the formulation of financial policy. Given the above, this doctoral thesis posits that reforms undertaken in the capital markets stand as an alternative mechanism to overcome the legal deficiencies that prevent countries from improving the quality of the corporate governance. We also conclude that the importance of taxes in corporate decisions is affected by the ownership structure. This indicates that the effectiveness of the efforts undertaken to reduce the debt tax bias is conditioned by the corporate shareholding structure.