| dc.description.abstract | The purpose of this study was to analyze the level of disclosure and its relation to the cost of debt and cost of equity of publicly-traded companies, after the conversion of the Brazilian accounting standards to IFRS. The main theory used to support the findings of this research, concerning the level of disclosure was the Disclosure Theory, demonstrating as well the importance of accounting information to its several users. The research was guided by a descriptive and explanatory analysis, where 66 companies were analyzed from 2005 to 2011, and it was calculated the cost of capital and the disclosure index for each organization. It was applied the regression method with panel data, in order to choose one, among the three main models, Pooled, Fixed Effects and Random Effects, the one which was the most suitable to demonstrate wheter and how the disclosure level affects the cost of capital of companies. The model was chosen according to the results obtained in the Chow, Breusch Pagan and Hausman tests, which were used to determine which model was the most suitable one. It was decided to add to the model some control variables, which are classified into internal and external. The internal variables are: indebtedness level, leverage ratio of companies, the variable firm size obtained by measuring the total assets of the organizations. As external variables, the value concerning the cumulative inflation rate in the years surveyed, the gross domestic products and the dummy IFRS variable tha verified if companies were preparing their financial statements according to the stadards of IFRS. The results show that there is no statistical evidence that the level of disclosure is related to the cost of capital, after the conversion to international accounting standards. The conclusions obtained are consistent with the hypothesis of this study, which mentioned that the level of disclosure of accouting information did not impact on the cost of capital of companies. It is emphasized that this conclusion should not be generalized, once that the sample used in the study, despite being representative, was obtained in a limited temporal space, with the possibility that increasing the amount of time, the findings presented here might change. | |
