| Nome: | Descrição: | Tamanho: | Formato: | |
|---|---|---|---|---|
| 1.56 MB | Adobe PDF |
Autores
Orientador(es)
Resumo(s)
O objetivo deste estudo é analisar os determinantes da rentabilidade do setor bancário brasileiro, medida por três indicadores: ROA (return on assets), ROE (return on equity) e NIM (net interest margin). Tem como variáveis explicativas selecionados a partir da revisão de literatura, fatores específicos dos bancos e fatores externos, incluindo as variáveis macroeconómicas.
A amostra é constituída por dados semestrais de um painel dos dez maiores bancos do Brasil em activos totais, para o período compreendido entre o primeiro semestre de 2007 e o segundo semestre de 2017. A metodologia, usada para testar as cinco hipóteses formuladas, consistiu no desenvolvimento de um modelo econométrico de regressão múltipla e na sua estimação com recurso ao método do painel mínimos quadrados ordinários (Panel Least Squares - PLS) para efeitos fixos.
Os resultados em geral indicam que a rentabilidade das instituições bancárias brasileiras depende da evolução das suas variáveis específicas sendo entretanto em grande medida influenciada pelos determinantes macroeconómicos cuja significância estatística é recorrente.
A preponderância dos fatores para explicar as proxies da rentabilidade (ROA, ROE e NIM) não foram uniforme. Os fatores que melhor explicam o ROA são a qualidade dos activos e a variável exógena do crescimento do produto interno bruto - PIB, enquanto que para o rácio ROE acrescem duas variáveis internas: alavancagem financeira e eficiência. Quando a medida da rentabilidade utilizada é o NIM, as regressões evidenciam um nível explicativo global superior e as variáveis específicas, alavancagem financeira, qualidade do activo, liquidez, custo do financiamento e estrutura do activo, evidenciam influência significativa na rentabilidade.
Além disso, o índice de concentração bancária (ICB) foi estatisticamente significativo, observando-se também uma relação significativa entre os determinantes do NIM e o crescimento do PIB e em menor grau com a taxa de juro e inflação.
The objective of this study is to analyze the determinants of the profitability of the Brazilian banking sector, measured by three indicators: ROA (return on assets), ROE (return on equity) and NIM (net interest margin). It has as explanatory variables, selected from the literature review, specific factors of the banks and external factors, including macroeconomic variables. The sample consists of semiannual data from a panel of the ten largest banks in Brazil for the period between the first half of 2007 and the second half of 2017. The methodology used to test the five hypotheses formulated consisted in the development of a the econometric model of multiple regression and its estimation using the Least Squares (PLS) method, for fixed effects. The results generally indicate that the profitability of Brazilian banking institutions depends on the evolution of their specific variables but is also influenced to a large extent by macroeconomic determinants whose statistical significance is recurrent. The preponderance of the factors to explain the proxies of profitability (ROA, ROE and NIM) was not uniform. The factors that best explain the ROA are the asset quality and the exogenous variable of the GDP growth, while for the ROE index two internal variables are added: Financial Leverage and Efficiency. When the measure of profitability used is the NIM, the regressions show a higher overall explanatory level and the specific variables, financial leverage, asset quality, liquidity, funding cost and asset structure, show a significant influence on profitability. In addition, the Bank Concentration Index (BCI) was statistically significant, with a significant relationship between the determinants of the NIM and GDP growth, and to a lesser extent with the interest rate and inflation.
The objective of this study is to analyze the determinants of the profitability of the Brazilian banking sector, measured by three indicators: ROA (return on assets), ROE (return on equity) and NIM (net interest margin). It has as explanatory variables, selected from the literature review, specific factors of the banks and external factors, including macroeconomic variables. The sample consists of semiannual data from a panel of the ten largest banks in Brazil for the period between the first half of 2007 and the second half of 2017. The methodology used to test the five hypotheses formulated consisted in the development of a the econometric model of multiple regression and its estimation using the Least Squares (PLS) method, for fixed effects. The results generally indicate that the profitability of Brazilian banking institutions depends on the evolution of their specific variables but is also influenced to a large extent by macroeconomic determinants whose statistical significance is recurrent. The preponderance of the factors to explain the proxies of profitability (ROA, ROE and NIM) was not uniform. The factors that best explain the ROA are the asset quality and the exogenous variable of the GDP growth, while for the ROE index two internal variables are added: Financial Leverage and Efficiency. When the measure of profitability used is the NIM, the regressions show a higher overall explanatory level and the specific variables, financial leverage, asset quality, liquidity, funding cost and asset structure, show a significant influence on profitability. In addition, the Bank Concentration Index (BCI) was statistically significant, with a significant relationship between the determinants of the NIM and GDP growth, and to a lesser extent with the interest rate and inflation.
Descrição
Palavras-chave
Rentabilidade bancária Dados de painel Ciclo económico Bancos do Brasil Bank profitability Economic cycle Brazil banks Panel data