I propose that CEO employment history is an observable characteristic that reveals the CEO’s unobservable risk-taking preferences. I hypothesize that CEOs that change employers more frequently (mobile CEOs) have a propensity to bear risk and implement riskier firm policies. Using a sample of S&P 1500 CEOs, I find that firms are more likely to hire mobile CEOs when the firm’s prior risk is high, firm-specific human capital is less important, the prior CEO turnover is forced, the prior CEO has a shorter tenure and the board is smaller and has fewer insiders. Mobile CEOs increase financial leverage, invest more in advertising and less in capital expenditures, and increase firm-specific risk. Mobile CEOs invest more (less) in R&D in homogenous (heterogeneous) industries where firm-specific knowledge is less (more) important in making investment decisions. Shareholders react positively to appointments of CEOs who change employers more frequently. I find no difference in long-run accounting performance for CEOs with different employment histories. Firms’ annual stock returns and sales growth are higher for CEOs who change employers more frequently. The cost of debt increases after the firm appoints a mobile CEO. These findings suggest that lower CEO risk aversion and the potential risk-shifting from shareholders to bondholders are sources of shareholder value increases. In sum, my findings provide evidence that CEO employment history is an observable characteristic that reveals the risk-taking preference of the CEO.