I extend the basic moral hazard model to include the possibility that the agent faces non-contractible uncertainty that is payoff relevant to him. These rewards may or may not be of a monetary nature. The agent devotes effort towards working for the principal and towards pursuing private rewards. Thus, the agent is multi-tasking. Despite the addition of private rewards and multi-tasking, I establish conditions under which the first-order approach remains valid. Interestingly, the model identifies a new source of economic rents for the agent. Moreover, to induce higher effort on the job, it may be optimal to write a contract that to an outsider appears to offer higher base utility but weaker or flatter incentives on the margin. Thus, the paper contributes to the literature on extrinsic versus intrinsic motivation. Here, the explanation is that a contract change leads the agent to reevaluate his "work-life balance" and to adjust, jointly, the effort devoted to both types of rewards. The principal optimally induces higher effort by manipulating this balance through the contract design. Finally, implications for common agency are examined.