We study dynamic moral hazard, with symmetric ex ante uncertainty and learning. Unlike Holmstroms career concerns model, uncertainty pertains to the di ¢ culty of the job rather than the general talent of the agent, so that contracts are required to provide incentives. Since e¤ort is privately chosen, the agent can always cause a misalignment of beliefs between the principal and himself, by shirking. We show that such a misalignment is always pro\u85table for the agent, and must be dissuaded by providing more high powered incentives. However, high powered incentives in the future only aggravate the incentive problem today, so that the problem is compounded as the interaction becomes longer. We also study the bene\u85ts of long term contract...