We experimentally investigate a simple version of Holmström?s career concerns model in which firms compete for agents in two consecutive periods. Profits of firms are determined by agents? unknown ability and the effort they choose. Before making second-period wage offers firms are informed about first-period profits. In a different treatment firms additionally learn the abilities of agents. Theory suggests high first-period equilibrium effort in the hidden ability treatment but no effort elsewhere. However, we find that effort is significantly higher in the revealed ability treatment and therefore conclude that transparency does not weaken, but strengthen career concerns incentives.