Abstract: People are more likely to take risks when they involve losses rather than gains, and are reluctant to part with arbitrary endowments. Prospect theory explains these and related findings in terms of a deep asymmetry in how we value gains and losses. Based on two sets of experiments, we suggest an alternative psychological account. The first set of studies finds risk seeking only for losses quantified on standard numerical scales but not for losses quantified on other numerical scales or for qualitative losses of consumer goods. The second set of studies shows that effects usually attributed to loss aversion are mirrored in “no-loss paradigms,” including an endowment-like effect in the absence of any endowment. Our alternative account suggests that prospect theory’s value function may reflect the psychophysics of numerical cognition (and not a more general psychophysics of value), and that the endowment effect may be a response to value imprecision (not value asymmetry). It further suggests that psychological asymmetries between gains and losses may be less pronounced than previously thought, and raises questions about the generalizability of prospect theory’s behavioral predictions beyond standard paradigms.