Executive equity compensation and earnings management: A quantile regression approach
Prior research has investigated the association between executive equity compensation and earnings management but the evidence is not conclusive. We investigate this question using the quantile regression approach which allows the coefficient on the independent variable (equity compensation) to s...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | text |
| Language: | English |
| Published: |
Institutional Knowledge at Singapore Management University
2011
|
| Subjects: | |
| Online Access: | https://ink.library.smu.edu.sg/soa_research/1213 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Singapore Management University |
| Language: | English |
| Summary: | Prior research has investigated the association between executive equity compensation and earnings management but the evidence is not conclusive. We investigate this question using the quantile regression approach which allows the coefficient on the independent variable (equity compensation) to shift across the distribution of the dependent variable (earnings management). Based on a sample of 18,203 U.S. non-financial firm-year observations from 1995 to 2008, we find that chief executive officer (CEO) equity compensation is positively associated with the absolute value of discretionary accruals at all quantiles of absolute discretionary accruals, but the association becomes weaker as the quantile decreases. The association between CEO equity compensation and signed values of discretionary accruals is positive (negative) when the discretionary accruals are at the high (medium and low) quantiles. The results are robust to alternative measures of equity incentives and earnings management and alternative model specifications. Overall, the quantile regression results suggest that equity compensation motivates income-increasing earnings management when the firm has low financial reporting quality, but mitigates income-increasing earnings management when the financial reporting quality is high. The results also demonstrate that the least-squares and least-sum optimization techniques which are used commonly in prior research do not capture the behavior of firms at the high and low quantiles of financial reporting quality. |
|---|