Voluntary disclosure and increases in earnings.

Abstract

Although intuition suggests that managers of firms that report large earnings increases have incentives to increase disclosure, prior empirical literature fails to document a strong association. In this paper, I study changes in disclosure made by managers of 80 firms undergoing at least eight contiguous increases in seasonally-adjusted quarterly earnings. By studying firms with persistent earnings gains, I provide a powerful test of the relation between disclosure and positive earnings performance. I find that total disclosure increases during the eight quarters of increased earnings. Examination of earnings forecasts indicates that managers begin to preempt the earnings increase several quarters before the announcement of the first large increase in earnings and continue to increase the number of forecasts provided throughout the increased earnings period. Consistent with an attempt to provide complementary information to help market participants correctly assess the persistence of the earnings increase, managers increase the amount of information provided with earnings announcements during the increased earnings period. Cross-sectional returns tests indicate a positive relationship between market reactions and the amount of complementary information provided, indicating that this information influences market participants' interpretation of the earning information. An examination of disclosure choice during the second year of increased earnings (quarters five through eight) provides evidence consistent with managers strategically using voluntary disclosure in an attempt to influence market participants' evaluation of current earnings news. After dividing the sample into two groups according to whether firms maintain the current earnings increases over the two years following the increased earnings period (quarters nine through sixteen), I find that the magnitude of the increases in total disclosure and disclosure with earnings announcements do not predict long-term performance. However, managers of firms with strong future prospects increase the frequency of long-term forecasts while managers of firms with weaker future performance (decline firms) increase the frequency of short-term forecasts. Additional analysis finds that, relative to quarters five and six, managers of decline firms decrease the frequency of long-term forecasts provided in quarters seven and eight, indicating the managers are aware of the upcoming earnings decline.