only $75 from maximizing the bonus compared to $2,295 for manager L4. Because the incremental benefit from maximizing their bonus in the upper end relative to playing it safe by selecting the average historical accrual is so much less for those in the upper bound, behavior in the upper bound, particularly with monitoring, is an interesting scenario that affected the way we measure the dependent variable as is discussed in the next section.Table 1 and 2Table 1 reports the average percentage of individuals who maximized their bonuses over the two periods by MONITORING and EXPERIENCE variables, and is partitioned by MAXLOWER and MAXUPPER. A separate Logit model was used in each of the MAXLOWER and MAXUPPER bounds. The main effects of the two independent variables (MONITORING and EXPERIENCE) were measured as well as their interaction. Table 2 reports the results of the Logit models. The hypotheses jointly
predict that monitoring would curb upward earnings management for more experienced managers, but not for less experienced managers, and monitoring would have no effect for downward earnings management. A statistically significant interaction (MONITORING EXPERIENCE) in the MAXLOWER
You've reached the end of your free preview.
Want to read both pages?
- Spring '17
- Bu Rini
- Management, Logit, Berlin U-Bahn