Earnings management constraints and market reactions to subsequent earnings surprises

In this dissertation, I examine investors' use of balance sheet information to infer earnings management constraint and the extent to which they utilize that information to assess the quality of subsequent earnings surprises. Ex-ante constrained firms may not have sufficient ability to manage earnings towards desired earnings thresholds and thus their reported earnings surprises are more likely to be the result of real performance. Ex-ante flexible firms, however, have more room to manage earnings and so it becomes less clear to investors whether the reported earnings surprises are the result of real performance or earnings management. My tests provide mixed support for the constraint theory. I find evidence that the market reacts more to small positive earnings surprises when they are reported by ex-ante constrained firms than when reported by ex-ante flexible firms. This suggests that the market interprets the earnings surprise reported by constrained firms to be of higher quality. However, I also find that earnings surprises reported by ex-ante constrained firms are no more persistent with regard to one-year-ahead earnings than those reported by ex-ante flexible firms, a result that is not consistent with the differential reaction to earnings surprises.