Purpose This study aims to examine the effect of CEO compensation risk, as measured by the proportion of equity-based pay (option and stock awards) relative to total compensation and pay sensitivity to stock volatility, on CEO pay for luck asymmetry. This paper also empirically examines CEO compensation risk as a mediating variable between the regulatory changes and CEO pay for luck asymmetry. Design/methodology/approach This paper test the proposed two hypothesis that CEO compensation risk is positively associated with the degree of CEO pay for luck asymmetry; and the pay related regulations implemented around 2006 could mitigate the degree of CEO pay for luck asymmetry using the fixed-effects regression models. Findings Consistent with the managerial talent retention hypothesis, this paper finds that CEO compensation risk, as measured by the equity-based pay as a proportion of CEO total compensation and CEO pay sensitivity to stock volatility, is positively associated with the degree of CEO pay for luck asymmetry. In addition, this paper find that CEO pay for luck asymmetry is significantly reduced by the major regulatory changes on executive compensation implemented around 2006. Research limitations/implications This study is among the very few studies exploring the impact of CEO compensation risk on pay for luck asymmetry in the literature. While the major purpose of the widely used stock options is to align executive interests and shareholder values, it also tends to increase the risk level of CEO compensation. So, a well-designed CEO pay package should protect risk-averse CEOs from bad luck for the retention purpose, which is also beneficial to shareholder wealth maximization. Therefore, future research on executive compensation needs to examine the issue from various perspectives. Practical implications For board of directors who is responsible for the compensation of CEOs, it is necessary to consider a broad range of factors when designing an optimal CEO pay package. Social implications The findings on the impact of regulations on CEO pay for luck asymmetry suggest that the executive-pay-related regulations around 2006 have indeed achieved some of their intended goals to significantly lower pay for nonperformance asymmetry, whereby CEO pay sensitivity to stock volatility has been identified as a major mediating variable. Originality/value This study contributes to the literature on executive pay for luck asymmetry in several perspectives. First, this paper finds that CEO compensation risk has a positive impact on the degree of CEO pay for luck asymmetry. Second, this paper finds that the CEO pay for luck asymmetry has been mitigated after 2006 when various regulatory changes on executive compensation began to be implemented in the USA. To the best of the authors’ knowledge, this study is among the very few studies investigating these issues in the literature.