Decision-making in committees. Game-theoretic analysis (Q1047596)

From a game-theoretic perspective, two general approaches to problems of decision-making in committees exist. First, traditional power indices, which are rooted in cooperative game theory, can be used to quantify the ``a priori voting power'' of a committee member, defined as the ability to affect the outcome of a vote. For this purpose, they abstract from decision makers' preferences. Yet, as equations would suggest, looking at these indices may tell us very little about the players' capacities to shape actual outcomes: the latter still depend on preferences and the feasible alternatives. Second, the collective decision process may be described as a non-cooperative game. In principle, allowing for the specific analysis of complex decision structures, no cooperative modeling necessitates a detailed description of who can do what and when, which, in many contexts, is simply not available. Thus, studies that analyze individual influence in this vein have been limited so far to purely distributional decisions or rely on both cooperative and no cooperative solution concepts. The borderland between these two approaches is exactly the location of the reviewed dissertation. This work consists of two key parts: the first is the review of existing theories that aim to assess the influence conferred upon committee members by the decision rules; the second presents new ground by combining elements of these theories to obtain fresh insights into committee decision processes. It presents an innovative model for choice of ``adequate'' voting rules for a democratic committee of representatives who act on behalf of groups of different sizes. The model presented here considers decisions which are elements of a one-dimensional convex policy space and may result from strategic behavior consistent with the median voter theorem. A mathematical investigation suggests that, under limit conditions and simple majority rule, a square root allocation of weights would be optimal -- a result, which is then confirmed by means of extensive Monte-Carlo simulations. In order to give individual citizens from different constituencies a priori equal chances to influence the collective decision, the voting weights of representatives need to be such that their power as measured by the Shapley-Shubik index is proportional to the constituency size. In the last part of the dissertation decision-making in a legislative committee under lobby influence is considered: lobbyists who have similar policy preferences wish to manipulate the collective decision by offering payments to committee members. A model of endogenous coalition formation among lobbyists is developed. Its analysis uncovers a new link between the status quo bias of a legislative committee and the severity of the collective action problem of the lobbyists. It turns out that small or no bias in favor of the status quo encourages the formation of the grand coalition of lobbyists, whereas, in the case of a large status quo bias, cooperation is partial in the sense that parts of the lobbyists will free-ride on others' lobbying efforts. The equilibrium coalition structure then involves under-provision of lobby efforts. To the extent that the legislative bargaining game used in the model can be taken as a stylized description of real world procedures, the model may help to understand the lobbying of supranational legislative institutions.