Working Paper: CEPR ID: DP13581
Authors: Marco Battaglini
Abstract: We propose a new model of legislative bargaining in which coalitions may have different values, reflecting the fact that the policies they can pursue are constrained by the identity of the coalition members. In the model, a formateur picks a coalition and negotiates for the allocation of the surplus it is expected to generate. The formateur is free to change coalitions to seek better deals with other coalitions, but she may lose her status if bargaining breaks down, in which case a new formateur is chosen. We show that as the delay between offers goes to zero, the equilibrium allocation converges to a generalized version of a Nash Bargaining Solution in which --in contrast to the standard solution-- the coalition is endogenous and determined by the relative coalitional values. A form of the hold-up problem specific to these bargaining games contributes to generate significant inefficiencies in the selection of the equilibrium coalition. We show that the model helps rationalize well known empirical facts that are in conflict with the predictions of standard non-cooperative models of bargaining: the absence of significant (or even positive) premia in ministerial allocations for formateurs and their parties; the occurrence of supermajorities; and delays in reaching agreements.
Keywords: legislative bargaining; hold up; head of state
JEL Codes: D72; D78
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| timing of offers (C41) | efficiency of coalition selection (C71) |
| coalition values (C71) | likelihood of forming efficient coalitions (C71) |
| holdup problem (D86) | inefficiencies in coalition selection (D79) |
| formateur's strategic choices (A29) | inefficiencies in coalition selection (D79) |
| bargaining breakdowns (C78) | coalition outcomes (D74) |