Abstract
This paper is the first to perform a comprehensive estimation of employee stock option exercise behavior and option cost to firms. We develop a GMM-based methodology, robust to heteroskedasticity and correlation across exercises, for estimating the rate of voluntary option exercise as a function of the stock price path and of various firm and option holder characteristics. We use it to estimate an exercise function for a sample of 1.3 million employee-option grants to 530,266 employees at 103 publicly-traded rms between 1981-2009. We use the estimated exercise functions in a simulation based valuation model to analyze the effect of different firms and option holder characteristics on option value, and show that the true value of these options can di er substantially from values calculated using the usual modified Black-Scholes approximation.