Corrupção no Brasil: comportamento do mercado acionário à divulgação de notícias de práticas corruptivas

Abstract

This paper aimed to understand how news involving corruptive practices can affect the financial market in terms of volatility, stock returns and traded volumes, comparing companies that adhered to the best practices of corporate governance in relation to those that did not adhere. The methodology adopted was evet studies and for that, the period from January 1, 2011 to March 31, 2016 was analyzed. The sample consisted of seventy-three companies, thirty-five were classified as best governance practices, thirty-five classified as worst practices and three companies involved in the corruption events analyzed. The data were collected in Economática, at daily frequency, and the generalized autoregressive model with conditional heteroscedasticity (GARCH) was used to calculate stock volatility. In defining the corruption events participating in the survey, they were divided into news items that positively affected the market and those that affected negatively. The hypothesis tests were performed through the event studies, considering an event window of 11 days (-5 to +5). The significance of the abnormal variations was verified by means of Student's t test and Wilcoxon test. The results found for volatility suggest that companies with worse practices are more affected than companies with best practices, indicating that the informational content regarding the dissemination of corruption news is asymmetrical. While results referring to returns and volumes indicate that the behavior of companies with best practices behaves similarly to the group of worst practices and does not present statistically significant differences. Another finding of the study refers to the volumes traded, since all groups analyzed reacted instantly, however only in the scenario where the market reacted negatively to the event.