THE ROLE PLAYED BY SIMPLE OUTCOMES IN COALITION FORMATION PROCESS OF THE CORE OUTCOMES

In the one-sided Assignment game any two agents can form a partnership. If this is done, the partners undertake some joint activity, which produces a gain that is split between them. We approach this model by focusing on simple outcomes - feasible and individually rational outcomes where only unmatched agents can block. We prove that this blocking can be done in such a way that the payoffs from the trades done are not changed as players reach the core. The core is non-empty iff every simple and unstable outcome can be extended to a simple outcome by a sequence of adjustments in which, at each step, payoffs are preserved for agents already matched and increased only for those newly matching. Hence, starting from the simple outcome where everybody stands alone, we can gradually increase cooperation by making Pareto improvements (and still staying within simple outcomes), until we reach the core, or until the payoff cannot be simple anymore. That is, increase in payoffs is only available through non-optimal cooperation of some agents. In addition, the total sum of these payoffs is the same at any core outcome. The gains in insight with this approach allows a necessary and sufficient condition for the non-emptiness of the core to be identified. Several properties of the core outcomes of economic interest are proved.