We consider a moral hazard problem where the agent has limited wealth which limits his possible actions. This may be due to different reasons: the opportunity cost can be monetary, the effort provided by the agent can actually be an investment, or the agent can invest in training activities in order to improve his capability. In such cases, the lower the level of wealth is, including transfer from or to the principal, the lower the maximum effort level that can be provided. The principal and the agent are risk neutral, so that limited wealth which limits possible actions is the distortion we consider compared to the standard model. We show then that the optimal contract is, in some cases, a sharing contract and the optimal up-front transfer is a payment from the principal to the agent. Moreover, whereas incentives and aid are substitutes in the case where the agent has sufficient wealth, they are complements when the agent has limited wealth. We also show that, if the agent can consume his wealth before the contract is signed, he gets all the surplus of the relationship. We discuss the implications of our findings in a variety of settings, including payments for ecosystem services, venture capital, and a current debate on wealth and cognitive functions.