Abstract
This paper examines auctions of incentive contracts, where the principal (a government agency) contracts with a private vendor to supply a given quantity. The contract is incomplete as the agent's research behaviour is unobservable and is not in the principal's objective function. The agent, however, has an incentive to engage in research. Therefore, the principal's actions inadvertently affect research spending. It is shown that the agent's optimal research spending responds to changes in contractual design. In general, the agent's research response is dependent upon the size of the bid and the probability of product innovation. Policy implications are discussed.