Although lay intuition and some academic theories suggest that power increases variability in mood, the prevailing view in the literature is that power elevates mood—a view that is not consistently borne out in empirical data. To rectify these discrepancies, we conducted five studies examining the impact of high and low power on mood in, and across, contexts of differing valence (negative vs. neutral vs. positive). Drawing on 19,710 observations from 1,042 participants, we found that high (vs. medium/control) power elevated, and low (vs. medium/control) power dampened, individuals’ mood at baseline/in neutral contexts and in positive contexts. However, neither high (vs. medium/control) power nor low (vs. medium/control) power modulated individuals’ mood in negative contexts. Overall, high (vs. medium/control) power tended to increase, and low (vs. medium/control) power decreased variability in mood across contexts (the former effect was marginally significant). We discuss how these findings corroborate, but also qualify, lay intuition and social psychological theories of power.