Power has long been associated with dishonesty. Here, we examined the contributions of personal and structural factors associated with power. Across five studies ( N = 1,366), we tested the hypothesis that being dominant, more than having power and felt prestige, predicts dishonesty in incentivized tasks, moral disengagement, and breaking of Covid-19 containment rules. Dominance and dishonesty were positively associated (Study 1). Furthermore, dominance contributed to the positive relationship between occupational power and dishonesty in natural settings (Studies 2 and 5). Different types of power had inconsistent effects on dishonesty (Studies 3 and 4). Prestige was unrelated to dishonesty. Dominant individuals were overrepresented at the top, suggesting that the association between power and dishonesty may derive from self-selection processes, rather than power itself.

Power activates social influence goals and goals linked to power roles-Power elicits goal orientation, energizing people towards actions that help initiate and advance any salient aims and desires-Power facilitates goal related cognition-Powerholders' goal orientation can be beneficial for the performance of some tasks-However, power can trigger the neglect of secondary goals, social inattention and social objectification.

Power has long been associated with corruption, yet most evidence has been linked to abuses for gains (money, resources, sex). In this article, we propose a conceptual framework that considers unethical conduct to obtain gains and to avoid losses.Following the situated focus theory of power (Guinote, 2007), we propose that power flexibly orients individuals' cognitions and efforts in line with active goals.Under a gains frame, compared to the powerless, the powerful should be more motivated to obtain gains and cheat more, in order to protect these gains. Under a loss frame, the powerful should experience a temporary activation of loss aversion goals, while the powerless should experience a chronic activation of loss aversion goals.Consequently, power differences in corruption levels should only occur for gains and not when losses are at stake. The effects of power and frame were demonstrated in one study (N = 321). The findings provided initial evidence supporting the notion that an understanding of the effects of power on corruption necessitates a consideration of contextual framing.