Abstract
This paper tries to address the problem of donator 's efficiency. In spite of the fact, that social services and public goods are not fully determined by market forces, they are not beyond the subject of economics.
The state or other donator can allocate grants among recipients either efficiently or inefficiently. In the paper the authors demonstrate the idea, that if a donator does not trust in grant recipients and states too strict limits for them, he/she risks the lower allocation efficiency.
Two models are analyzed: one in which postponing of grant funds to the next budget period is not allowed and second with maximum of allowed portion for overhead cost. It is shown that such limits could be contra-productive as far as the initial donator's aim (maximization of the probability of survival of the recipients) is concerned.