查看文章

This paper tests for ratcheting in a dynamic principal–agent game where the principal does not have complete information. In such situations the principal often uses any information revealed by the agent's actions to extract the latter's informational rent in future periods – the “ratchet principle”. This in turn induces the agent to underproduce in order to avoid more demanding schedules in the future. We find little evidence of ratcheting. Agents play the game in a naive fashion and reveal their types even when such revelation is not optimal and the principal often does not exploit such type revelation.