The study tests and confirms the hypothesis that costs of Brazilian firms are sticky, i.e. that costs increase with more intensity when revenues increase than in the opposite direction, as shown by recent empirical evidence involving American firms. As opposed to this evidence, however, the sticky behavior does not seem to decrease when periods over one year are considered. The possibility of a partial reversion of the asymmetry when lagged periods are tested is confirmed in the study. The adopted methodology involves different types of panel data regressions. The paper intends to contribute to a better understanding of Brazilian firms' cost behavior related to changes in the firms' activity level, which is a relevant theme for business administration, accountants and external financial analysts. By using a sample of 198 firms within a 17-year period, it was found that the properties of the sticky costs model proposed by Anderson, Banker e Janakiraman (2003) are partially applicable to Brazil.
Sticky Costs; Brazilian Firms; Cost Behavior; Empirical Tests
