Do CEOs possess any extraordinary ability critical to organizational success? Do CEOs receive large pay because of their unique abilities? We examine those questions using one such ability – the ability to forecast future firm risks – because over the last 40 years, firms’ survival and success has become highly dependent on firms’ preparedness to face uncertain future business environments. We measure CEOs’ risk-forecasting ability by the information contained in their personal equity trades. We find that CEOs’ earlier-than-normal stock-option exercises often represent modifications of their personal portfolios in anticipation of future risks unknown to the market today. These results suggest that on average, CEOs can forecast at least some future firm risks better than the market can, and they use that information to modify their personal portfolios. Additional tests show that firms design incentive-compatible compensation contracts by tying high-powered components of CEO compensation to their risk-forecasting abilities, arguably, to competitively sort and attract CEOs with better risk-forecasting abilities and/or to motivate them to expend greater efforts on forecasting risks. We conclude that increases in CEO pay over the last 40 years can be explained, at least partly, by the increasing importance of risk-forecasting skills, especially because firms have become smaller and are more susceptible to failure due to sudden changes in the business environment, but are also more likely to benefit from opportunities created by those changes.