5. Robustness Checks
In this section, we perform other empirical tests to check the robustness
of the empirical results presented in the previous section. First, we run a
similar QR in Table 4, but control for industrial and yearly effects. Second,
to further control the industry effect, we replicate our regression
using the industry-adjusted RoE. Third, we use the pay-for-performance
measure from stock options as an alternative measure for the impact of
stock-based compensation. Fourth, to assess whether the estimate
parameters of our QR model might be biased due to an omitted-variables
problem, we include the book-to-market ratio as an additional
control variable. Fifth, to check whether the result could extend to firm
market performance, we use Tobin’s q as the dependent variable in QR.
Finally, to alleviate the problem of endogeneity, we use the lagged
stock-based/total compensation ratio as the explanatory variable and
perform the same QR as that in Section 4.2. We adapt one change at a
time and keep other variables the same as the setting in the previous section
to trace the impact of the change in the particular variable.