Economic rewards versus economic sanctions in international relations

I use game-theoretical models to compare a sender's expected payoff under two methods of wielding influence under incomplete information: offering rewards or threatening punishments. Attempts to influence another's behaviour can have the perverse effect of actually encouraging the behaviour that one is trying to discourage by creating, albeit different, incentives to bluff. Under a reward regime, targets can bluff in order to extort larger inducements. Under a sanction regime, targets can bluff in order to deter coercion attempts. I find that while sanctions are a weakly dominant tool of influence under complete information, rewards can be preferable under incomplete information. The sender often could do better by committing to a certain course of action for the entire duration of interaction. Moreover, using rewards and sanctions in combination can mitigate their adverse effects on targets’ bluffing incentives when deploying only one instrument of influence separately.