Prices lead earnings no Brasil?

• Autor(a) principal: Santos, Mateus Alexandre Costa dos
• Data de Publicação: 2013
• Outros Autores: Mol, Anderson Luiz Rezende, Anjos, Luiz Carlos Marques dos, Santiago, Josicarla Soares
• Tipo de documento: Artigo
• Idioma: por; eng
• Título da fonte: Revista Contabilidade & Finanças (Online)
• Texto Completo: https://www.revistas.usp.br/rcf/article/view/78835

This article aims to identify the timing of the return-earnings relationship in Brazil, that is, the degree of time lag between the occurrences of the variables. This research was developed based on assumptions from the prices lead earnings hypothesis, the fundamental premise of which is that the stock price is informationally richer than the current and past accounting earnings in terms of future earnings, which invalidates the establishment of a contemporaneous relationship (timing zero) between these variables. This research was conducted by means of pooled regression using panel data (fixed effects and random effects). Four models were employed in total. A total of 205 firms were analyzed over 53 quarters (1999 to 2012), resulting in 8,440 firm-quarters. The results indicated that accounting earnings alone are not informationally contemporaneous to stock price. However, when the effects of future earnings on this relationship were eliminated, it was found that there are signs of timeliness. Furthermore, it was found that the returns anticipated information about future earnings. The identified associations suggest that this anticipation occurs over at least eight quarters. However, it was not possible to determine the timing of the quarterly return-earnings relationship in Brazil because, on the one hand, past returns are associated with current earnings and, on the other, the significance of future earnings in explaining current returns depends on the arrangement of the independent variables in the model. Nevertheless, it is clear that the results converge with a timing equal to 1, in which the return anticipates earnings in the following period, a result that was independent of the addition of the other variables in the model.