Real earning management no Brasil em duas condições de avaliação

Abstract

This dissertation analyzes the theme of results management through operational decisions in the Brazilian capital market, this type of management affects the quality of the profit and the operational cash flow, the suboptimal choices in relation to the levels of operational practices and the moment of the company. The general objective of this dissertation was to analyze Brazilian capital market companies by means of their accounting information in two conditions (annual and quarterly) in the period 2008 to 2015 listed on the São Paulo Stock Exchange, BM&FBOVESPA and understand Management of operational results. The annual and quarterly data come from the 324 companies of the Brazilian capital market listed on the BM&FBovespa from the Economática provider in the period from 2008 to 2015. Four hypotheses were formulated in this study to compare the quarterly and annual period: (i) identify if The companies suspected according to the profit margin (net profit divided by total assets) used the actual earning management (using the quarterly period is hypothesis 1, using the annual period is hypothesis 3) (ii) analyze whether companies suspected of agreement (Using the quarterly period is hypothesis 2, and using the annual period is hypothesis 4). In this case, the hypothesis is that the hypothesis of the hypothesis is that the hypothesis of the hypothesis of the hypothesis According to Roychowdhury (2006), therefore, several studies derived from this research, in order to be possible the comparability was used the three operational activities mentioned by the author: (1) high level of production with intention to reduce CPV (variable NPROD); (2) reduction of discretionary expenses (administrative, general and sales) (DD) and (3) use of sales (FCO). They were initially used to estimate the normal levels of operating activities (production, discretionary expenditure and sales) by means of regressions with panel data following the econometric models developed by Dechow, Kothari and Watts (1998) and empirically implemented by Roychowdhury ( 2006) and the residuals of these regressions are the levels of abnormality of these operational activities and the variables dependent on the regressions applied in the second moment. In these estimates of operating activity abnormality, we identify the companies that are suspected of making use of management of operating results through the variable profit margin (net profit minus prior period net profit divided by total assets) and profit margin Net of prior period net income divided by total assets). An aggregate measure was created, which allowed to analyze the operational activities at the same time and thus to answer the four hypotheses raised in this study. The results show that the companies suspected of using real earning management according to the profit margin had statistical evidence in the annual and quarterly period of the confirmed suspicions and so there was manipulation in the results in order to avoid disclosing losses. When the suspect companies are analyzed according to the variation of the profit margin between 0 and 1%, the suspicions are confirmed in the quarterly period that the companies made use of real earning management, analyzing the annual period the suspicions are not confirmed. Therefore, companies using quarterly period are more likely to avoid disclosing losses incurred in the period.