Games Real Actors Play: Actor-Centered Institutionalism in Policy Research

In comparison with the Nash equilibria that may be reached through unilateral action, the outcomes achieved through negotiated agreements are attractive in welfare-theoretic terms. However, negotiations run into prohibitive transaction costs when the number of independent parties increases. By contrast, the choices of very large numbers of actors may be coordinated at very moderate transaction costs if collectively binding decisions can be imposed by majority rule or by hierarchical fiat. In order to be considered "collectively binding," these must be decisions for which compliance can be expected from a given set of actors, even though they have not agreed to them, even though these decisions go against their interests, and even though these actors might have options of unilateral action that would improve the outcome for them individually (i.e., even though the outcomes are not Nash equilibria). But what could account for this binding force?

Nash equilibria reached through unilateral action among independent actors will generate their binding force endogenously within the interaction itself -- none of the parties has a better option that it could choose unilaterally. In the case of negotiations, as I pointed out at the beginning of Chapter 6, this is also true of contracts that can be classified as instances of "joint production," in which neither party has a motive to renege on its commitments. This is not so in contracts involving the "exchange" of goods or services, in which, if performance is not simultaneous with the agreement, implementation becomes a problem when purely self-interested rational actors are assumed. As long as alter is keeping its side of the contract, ego could benefit from defaulting on its own commitment. This is a Prisoner's Dilemma constellation that may have an endogenous solution under network conditions, in which actors will have to deal with one another repeatedly, will observe each other, and will invest in valuable reputations that would be lost in case of default ( Milgrom/ North/ Weingast 1990; Scharpf 1994; de Jasay 1995). Where these conditions do not exist, even the implementation of negotiated agreements depends on enforcement mechanisms that are exogenous to

-151-