Determinants of Corporate Voluntary Disclosure in Brazil

Abstract

The central objective of this paper is to identify the factors that explain the level of voluntary disclosure of Brazilian public companies. The theory underpinning this work is the Discretionary-based Disclosure (Verrecchia, 2001). Sample is composed by the top 100 largest non-financial public companies listed in Bovespa. Information has been gathered from Financial Statements for the years ended in 2006, 2007 and 2008 with the use of content analysis. For these, a disclosure framework based on 27 studies has been elaborated. This framework contains a total of 92 voluntary items, divided in two dimensions: economic (43), social-environmental (49). Based on the existing literature, a total of 12 hypotheses have been elaborated and tested using a panel data approach. We have also analyzed the association between voluntary disclosure and company’s sector, which has been confirmed by the Chi-square test at a significance level of 1%. Correspondence Analysis, by means of the perceptual map, has evidenced that (i) retail sector is associated with a ‘bad’ level of disclosure and (ii) electric sector is associated with a ‘good’ and ‘great’ level of voluntary disclosure. Panel data analysis has evidence that: (i) ‘Sector’ and ‘Origin of Control’ are statically significant in all three models tested: economic, social-environmental, and total; (ii) ‘Profitability’ is relevant in the economic model and in the total model; (iii) Tobin’s Q is relevant in the social-environmental model and in the total disclosure model, (iv) ‘Leverage’ and ‘Auditing Firm’ are relevant, only, in the economic disclosure model; (v) ‘Size’, ‘Governance’, ‘Stock Issuing’, ‘Growth Opportunities’ and ‘Concentration of Control’ are not statistically significant in any of the three models of disclosure.