Estrutura de capital, maturidade da dívida e market timing: evidências do desempenho financeiro em empresas de capital aberto

• Ano de defesa: 2026
• Autor(a) principal: Worley Junio Ribeiro Borges
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Minas Gerais
• Palavras-chave em Português: Sociedades por ações; Finanças; Empresas; Contabilidade; estrutura de capital; maturidade da dívida; custo da dívida; market timing; desempenho financeiro
• Link de acesso: https://hdl.handle.net/1843/4807

This dissertation examined the relationship between capital structure, debt maturity, and market timing, as well as their effects on the cost of debt and the financial performance of non-financial firms listed on B3. Grounded in the Trade-Off, Pecking Order, and Market Timing theories, the study investigated how financing decisions, the time profile of liabilities, and market conditions are associated with leverage, debt maturity, the cost of debt, and corporate profitability. Methodologically, a quantitative approach was adopted using an unbalanced quarterly panel of Brazilian firms from 2017 to 2024, estimated through pooled, fixed-effects, and random-effects models, complemented by dynamic specifications with lags and GMM. The results provided favorable evidence of a positive association between debt maturity and leverage. For debt maturity, the findings were partial, with emphasis on the positive relationship between market timing and debt lengthening. In the cost of debt model, leverage showed a negative and significant association, whereas debt maturity presented only complementary evidence in alternative specifications. Regarding financial performance, the results were more consistent, with leverage being negatively associated with both ROA and ROE, debt maturity showing a positive association with performance, and market timing being favorably associated with profitability, particularly in the lagged specifications. It is concluded that the estimated relationships are heterogeneous across models and performance measures, providing conditional evidence rather than an integrated confirmation of all the channels examined.