In a cheap‐talk setting where the conflict of interest between sender and receiver is determined endogenously by the choice of parameters θ i for each agent i , conditions are provided that determine the sign of each agent's inverse demand for θ without assuming that the most informative equilibrium will necessarily be played in the cheap talk game. For two popular functional forms of payoffs, we derive analytically tractable approximations for agent i 's demand for θ . In an application where the θi 's are purchased on a competitive market, we provide conditions for a competitive equilibrium to feature maximal information transmission. In a principal–agent application where the agent's θ is set by the principal, our results show that information transmission will be partial. We consider extensions where: (1) the θ 's are acquired covertly rather than overtly and (2) the θ 's are traded after the sender has received the information.

We study an agency model in which the principal knows the agent-optimal actions in response to K “known” contracts but is unaware of other actions available or their costs, and seeks a contract to maximize worst-case profits. The optimal contract is a mixture of the known contracts and output. Moreover, when K = 1, the single known contract maximizes the principal's profit guarantee, whereas with two known contracts, the optimal mixture puts positive weight on one of the known contracts. Our methodology is straightforward to implement, a point that we demonstrate using data from an experimental study of different incentive schemes.

Two players with common interests exchange information to make a decision. But they fear scrutiny. Their unencrypted communications will be observed by another agent with different interests who can object to their decision. We show how the players can implement their ideal decision rule using a back-and-forth conversation. Such a subversive conversation reveals enough information for the players to determine their best decision but not enough information for the observer to determine whether the decision was against his interest. Our results show how conversations can maintain deniability even in the face of leaks, hacks, and other public exposures.