Prestação de serviços de consultoria por auditores independentes: há reflexos no gerenciamento de resultados em empresas brasileiras de capital aberto?

• Ano de defesa: 2011
• Autor(a) principal: Fabiana Lucas de Almeida
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Minas Gerais; UFMG
• Palavras-chave em Português: Gerenciamento de resultados; Auditoria; Independêcia; Contabilidade
• Link de acesso: http://hdl.handle.net/1843/BUOS-8M3H8J

In the Context of Agency Theory, the independent audit works as a monitoring mechanism of the managers actions, aiming its alignments with the shareholders interests. This monitoring is effective only if the auditor is independent. According to the literature, there are practices thatmay harm the auditors independence, as non-audit services and auditor tenure. The literature also indicates that reputation, expertise and size are factors that could strengthen the monitoring capability of the audit firm. Therefore, this research aimed to analyze the relations between nonauditservices and the earnings management in Brazilian public companies, assuming that the auditor is the agent in the agency conflict, in the period of 2006 to 2009. Specifically, we aimed to verify if the independent auditors economic bond with the client can become stronger, reducing, thereby, his or her monitoring capability. Also, we intend to verify if the corporate governance rules required by Sarbanes-Oxley Act have any influence on this practice. To achieve our objectives, we found the relationship between the ratio of non-audit fees to total fees and the discretionary accruals, estimated through Modified Jones (1995) model and KS (1995) model. Moreover, we use an adaptation of Frankel, Johnson e Nelson (2002), model (2002), that differs companies which are audited by big four firms from the others and that considers the audittenure. The effect of the factors that the literature claims to impact the earning management was controlled by the inclusion of variables related to these factors. Furthermore, to verify the effect of the ratio of non-audit fees to total fees lower than 5%, since that CVM 381 Standard exempts the companies from this disclosure, we included dummy variables to the Modified Jones (1995) model e KS (1995) model, by OLS. The results show a reduction of the negative discretionary accruals in the presence of non-audit services, in 2009, and a increase in total accruals in 2008 and in the combined cross section and time series analysis when we include dummy variables. However, these results should be analyzed carefully, once neither the observations by the application of the modified Jones (1991) model was confirmed by the application of the KS (1995) model and vice versa nor the yearly analysis was unanimous. It is considered, still, as a research limitation, the reduced number of companies that disclosed non-audit fees and that have all the required accounting data, which restricted the sample size to 52 observations per year and 208 observations to combined cross section and time series analysis