The theory of relative deprivation (RD) distinguishes between personal and group RD, and links the two to personal and group identity. Models linking RD and identity suggest that identity causally precedes RD. These models are extended in the present study to allow for causal effects of RD on personal and group esteem. An experiment is reported which was designed to examine the causal effects of personal and group RD on personal self-esteem and collective self-esteem. It was predicted that experiencing personal RD would lower personal self-esteem but would not affect collective self-esteem, and, conversely, that experiencing group RD would lower collective self-esteem but would not affect personal self-esteem. The results of a 2 (sex of subject) X 2 (sex of others in group: either 3 males or 3 females) X 2 (reward condition: fair or unfair) experiment supported this prediction for personal self-esteem, but only partially supported the prediction for collective self-esteem. The implications of the results for understanding the failure of RD to translate into action, and the usefulness of the experimental paradigm for inducing personal or group RD are briefly discussed.