O presente estudo investiga os impactos da redução nas restrições financeiras para a captação de dívidas no mercado de capitais brasileiro sobre o financiamento e os investimentos de companhias brasileiras de capital aberto e fechado, não financeiras. Empiricamente, o estudo interpreta que a Instrução CVM nº 476, vigente a partir do ano de 2009, que trata das ofertas públicas de valores mobiliários distribuídas com esforços restritos, pode ter reduzido as restrições financeiras enfrentadas por sociedades anônimas brasileiras, dado que tal mudança regulatória, comparativamente ao mecanismo tradicional para o acesso ao mercado de capitais nacional, reduziu os custos de emissão de debêntures (valor mobiliário em foco) no mercado de capitais nacional, tornou o acesso ao capital mais rápido e reduziu uma barreira regulatória para as sociedades anônimas fechadas, impedidas de realizar emissões públicas de debêntures no mercado de capitais nacional antes da CVM 476. Sobre o financiamento, foram propostas as hipóteses de que a CVM 476, isto é, a redução nas restrições financeiras, possibilitou que as companhias aumentassem sua alavancagem total, devido ao aumento da alavancagem de longo prazo. Sobre os investimentos, foram propostas hipóteses relacionadas ao caixa e aos investimentos em ativos de longo prazo. Em relação ao caixa, a hipótese propõe que a CVM 476 possibilitou que as companhias mantivessem menores saldos em caixa, devido à redução no componente de restrição financeira da política de caixa. Sobre os investimentos em ativos de longo prazo, a hipótese propõe que a CVM 476 possibilitou que companhias aumentassem tais investimentos, pois poderiam tornar-se menos dependentes da geração interna de recursos e de outras fontes de capitais. É esperado que tais efeitos sejam maiores para as sociedades anônimas de capital fechado, pois, a priori, espera-se que tais companhias sejam mais restritas financeiramente do que as abertas, especialmente no que se refere ao acesso ao mercado de capitais nacional. Para investigar as hipóteses, foram realizadas duas abordagens empíricas. Na primeira, as emissões de debêntures pela CVM 476 foram adicionadas aos modelos como a variável explicativa de interesse, sendo representada por uma variável binária que indica a emissão das debêntures, no ano da emissão. Na segunda abordagem, a CVM 476 é tratada como um experimento natural que pode ter provocado um choque exógeno nas restrições financeiras, ou seja, uma diminuição nas restrições financeiras enfrentadas pelas sociedades anônimas brasileiras de capital aberto e fechado (as companhias tratadas), de modo que, nesta abordagem, as empresas de responsabilidade limitada são definidas como grupo de controle. Os resultados são compatíveis com as hipóteses relativas ao financiamento, isto é, a CVM 476 possibilitou o aumento da alavancagem total e de longo prazo das companhias, de modo que há evidências que tais efeitos sejam maiores para as sociedades anônimas fechadas. Por outro lado, não foram encontradas evidências robustas que indiquem que a CVM 476 possibilitou, de forma ampla, ajustes no caixa e nos investimentos em ativos de longo prazo de companhias afetadas pela Instrução / This study analyses the effects of reducing financing constraints to debt issuance in Brazilian bond market on financing and investments of Brazilian public and private non-financial companies. Empirically, this study interprets that CVM Act nº 476, in force since 2009, which deals with public issuance of securities issued with restricted efforts, may have reduced financing constraints faced by Brazilian companies, since this Act, compared to the traditional mechanism to access Brazilian bond market, reduced issuance costs of debentures in domestic bond market, made the access to capital faster and reduced a regulatory barrier for private companies, restricted from conducting public issues of debentures in domestic bond market before CVM 476. The hypotheses related to financing propose that CVM 476, that is, the reduction in financing constraints, allowed companies to increase total leverage, due to the increase in long-term leverage. Regarding investments, hypotheses related to cash holdings and investments in long-term assets were proposed. The hypothesis concerning cash proposes that CVM 476 allowed companies to maintain lower cash holdings, due to the reduction in financing constraint component of the cash policy. Regarding investments in long-term assets, the hypothesis proposes that CVM 476 allowed companies to increase their investments, since they could become less dependent on internally generated cash flow and other sources of capital. These effects are expected to be greater for private companies, since these companies are expected to be more financially constrained than public companies, especially with regard to the access to the domestic bond market. Two empirical approaches were employed to investigate the hypotheses. In the first one, debentures issues according to CVM 476 were added to the models as the explanatory variable of interest, and is represented by a dummy variable that indicates the issuance of the debentures, in the year of the issue. In the second approach, CVM 476 is analyzed as a natural experiment that may have caused an exogenous shock in financing constraints, that is, a decrease in financing constraints faced by Brazilian public and private companies (treatment group) and, in this approach, limited liability firms are defined as the control group. The results are compatible with the hypotheses regarding financing, that is, CVM 476 made possible the increase in total and long-term leverage of the companies, and there is evidence that these effects are greater for private companies. On the other hand, no robust evidence was found that indicate the CVM 476 made possible extensive adjustments in cash holdings and investments in long-term assets of the companies treated by CVM 476.

Cette thèse est composée de 3 essais. Le 1er essai traite de la problématique du risque de pertes extrêmes dans le secteur bancaire dans un contexte du problème d'agence entre les actionnaires et les top managers des banques. Pour pouvoir inciter les banques à ne pas prendre le risque de pertes extrêmes, il est proposé d'appliquer la régulation des fonds propres sous forme d'une politique de recapitalisations obligatoires, dont les paramètres sont choisis pour inciter les actionnaires à rémunérer leurs managers de la manière à les détourner des stratégies au risque de pertes extrêmes.Le 2ème essai développe le design de la supervision bancaire qui vise à éliminer le problème d'aléa moral au sein d'une banque, tout en assurant un coût minimum de supervisions. Les banques, dont la situation financière commence à se dégrader, doivent être soumises à des audits aléatoires. Les banques, dont la valeur de l'actif s'est dégradée considérablement, doivent être mises sous tutelle pour un redressement financier. Les auditeurs externes peuvent être impliqués dans le processus de supervision, mais ne doivent pas complètement remplacer les régulateurs. Le 3ème essai étudie comment la capacité d'emprunt de l'entreprise non-financière affecte sa politique d'investissement en présence des coûts d'émission de la dette. Il est montré que les entreprises, dont la capacité d'emprunt est moyenne, ont intérêt à réaliser un investissement plus important par rapport aux entreprises dont la capacité d'emprunt est relativement faible/forte. Cela est entièrement dû à l'effet des coûts fixes d'émission de la dette, qui émerge dans le contexte dynamique d'investissement. / This dissertation consists of 3 self-contained theoretical essays.Essay 1 brings into focus the problem of "manufacturing" tail risk in the banking sector. This work shows that, in order to prevent banks from engaging in tail risk, bank capital regulation should account for the internal agency problem between bank shareholders and bank top managers. It is proposed to design bank capital requirements in the form of incentive-based recapitalization mechanism which would induce bank shareholders to shape executive compensation in such a way as to prevent top managers from engaging in tail-risk.Essay 2 deals with the problem of moral hazard in bank asset management. It proposes the concept of incentive-based bank supervision aimed at preventing moral hazard at a minimum cost to the regulator. It is shown that the intensity of supervision efforts should be gradually adjusted to the bank's financial health: banks in the mild form of distress should be subject to random audits, whereas deeply distressed banks should be placed under temporary regulatory control. To prevent double moral hazard, external auditors involved in supervision should be offered the optimal incentive contract.Essay 3 examines the impact of credit rationing (debt capacity) on corporate investment in the setting with costly debt financing. It is shown that, when credit constraints are binding, the firms with intermediate levels of debt capacity will establish larger investment projects than the firms with relatively low or high debt capacity. This non-monotonicity of investment on debt capacity arises due to the effect of the lump-sum debt issuance costs in the dynamic context of investment.