The causal effect of option ownership on employee performance: Empirical evidence from personnel data