Research on allocation bias, inspired by social identity theory, has dealt with the relative status between groups but seldom examined the individual's personal status relative to others within the group. This experiment examined the effect of personal status on bias, and how such an effect might vary with group status. When asked to allocate money rewards directly to ingroup and outgroup members who had done equal works, subjects allocated equally regardless of status manipulations. When asked instead to assign task weightings which would then determine the rewards, (1) subjects with low personal status favoured the ingroup, regardless of their group's status; and high-status subjects either (2) allocated equally (when group status was high) or (3) favoured the outgroup (when group status was low). Thus, personal status could lead to bias either by itself or jointly with group status. Additionally, members of the low group exaggerated the ease of the task that gave them high personal status, whereas members of the high group exaggerated the difficulty of the task that gave them low personal status.