This paper considers the effects of an interim performance evaluation on the decision of a principal to delegate authority to a potentially biased but better informed agent. Assuming the agents’ outside option to be determined by market beliefs about their type, interim evaluations (a) provide a possibility for the principal to potentially separate biased agents from unbiased agents and (b) induce an incentive for biased agents to imitate unbiased ones in order to retain the decision authority and to increase their wages in later periods (in case of public evaluation). We show that the principal always profits from a private evaluation while a public evaluation is only beneficial if the corresponding wage effects are not too costly. Nevertheless, the principal prefers public over private evaluation if the imitation incentive for the biased type is high enough. Finally, regarding implications for economic policy, we show that in view of aggregate welfare any evaluation conducted ought to be disclosed to the public.