The Effect of CEO Stock-Based Compensation on the Pricing of Future Earnings

Abstract

This paper examines whether CEO stock-based compensation has an effect on the market’s ability to predict future earnings. When stock-based compensation motivates managers to share their private information with shareholders, it will expedite the pricing of future earnings in current stock prices. In contrast, when equity-compensated managers attempt to temporarily manipulate the stock price to maximize their own benefit rather than that of shareholders, the market may not fully anticipate future performance. We find that a CEO’s stock-based compensation strengthens the association between current returns and future earnings, indicating that more information about future earnings is reflected in current stock prices. In addition, we find that the positive effect is weaker for firms that have a high level of signed discretionary accruals or a low management forecast frequency. Overall, our study suggests that on average, equity-based compensation improves the informativeness of stock prices about future earnings, while opportunistic discretionary accruals or lowered earnings guidance hamper this improvement.

Acknowledgements

We are very grateful to the Editor and the anonymous referee for their constructive comments. We also thank attendees at the 2014 Asian Finance Association Conference as well as seminar participants in University of Technology Sydney and University of Newcastle for their helpful comments.

Notes

1 We later explain why the FERC is more appropriate for examining the effect of stock-based incentives on the informativeness of stock prices than other short-term-oriented measures.

2 This notion is further supported by Choi et al. (2011), who stress the importance of the FERC, especially when examining the effect of management forecasts. They report that characteristics of management forecasts are closely related to the FERC but rarely influence the degree to which returns reflect current-period earnings, that is, the ERC.

3 In particular, the mean and median of PMDTA are similar in magnitude to those reported in previous studies such as Cohen, Dey, and Lys (2008) and Kothari et al. (2005).

4 Because the coefficients on X_t3×GROWTH are negative, which is inconsistent with our expectation, we conduct a test using the market-to-book ratio (MTB) as an alternative proxy for growth. The coefficient on X_t3×MTB is positive and significant, as expected, while our main findings are qualitatively unchanged (untabulated).

5 We repeat the tests in Table 4 using the absolute discretionary accrual measure instead of our signed discretionary accrual measure. The coefficient for the highest earnings management group is still negative but is not significant when the absolute measure is employed. This result, together with that from the analysis in Section 5.3, indicates that the reported negative impact of EM on the FERC is mainly driven by managers who engage in temporary income-boosting for their future selling activities.

6 We posit that this weaker effect for the post-adoption period could be due to the small sample size in the post-adoption period. In addition, the post-adoption period overlaps with the economic downturn of the financial crisis, during which stock prices and earnings were very volatile. The high volatility associated with these great uncertainties may not have been fully anticipated by managers in 2007, for which the current stock return data are collected. While it would be interesting to further examine the implications of the new accounting standard related to stock-based compensation on managers’ disclosure behaviors, we propose that such an examination be conducted in a future study due to the sample limitation and scope of the present study. Further implications are discussed in the conclusion.

7 We appreciate the referee’s suggestion to conduct this analysis to strengthen our inferences.

8 In addition, we test whether the three instruments are associated with the FERC by estimating the FERC model augmented with these three variables and their interaction terms with future earnings. The untabulated results depict that all of the coefficients on the interaction terms (between the three instrumental variables and future earnings) are not significant, indicating that these variables are not associated with the FERC. This result provides additional evidence for the validity of our instruments.

9 Alternatively, we also employ the lagged variable of stock-based compensation as an explanatory variable. To the extent that stock-based compensation from the previous year is based on determinants that are available only before the current year, the previous year’s stock-based compensation is more likely to be exogenous to the factors that potentially affect the FERC in the current year. The inferences from the test using the lagged variable of stock-based compensation are similar to those from the main analyses (untabulated).

10 To obtain a meaningful inference, for both events, we only consider cases in which a firm has at least two observations in both the pre- and post-periods.

11 We thank the anonymous referee for suggesting this argument and the test.

12 We obtain qualitatively similar results if we use an alternative measure of net sales that uses the market value of equity as the deflator instead of total shareholdings, following Cheng and Warfield (2005).

13 Holding other factors constant, the FERC is likely to be positively associated with the transparency of a firm’s information environment because many determinants of the FERC (such as analysts following, institutional trading, and corporate disclosures) relate to information flows in the market. However, individual factors may not capture the overall transparency due to the existence of private information and other external market factors. The opacity measure developed by Anderson et al. (2009) provides a composite index of the transparency of a firm’s information environment by combining four commonly used proxies for information uncertainty and asymmetry.

14 The other variables are as defined earlier, and their exact measurements are provided in Appendix A.

15 Although stock movement has an effect, it should strengthen the results in the opposite direction of our main findings (i.e. a negative association between stock-based compensation and the FERC).

16 However, in testing for earnings management, we explore cases in which managers are more likely to engage in earnings management (i.e. high insider sales in the following period).

17 We thank the editor for suggesting this case.

Additional information

Funding

We gratefully acknowledge the financial support of the School of Accountancy Research Center at the Singapore Management University.