Settlement Delay as a Sorting Device

This paper examines the use of sorting by defendants as a strategy for lowering their overall costs of litigation. The effectiveness of this strategy relies on the defendant’s ability to bundle settlement offers with some other variable that is valued differently by plaintiffs. The example I examine in this paper is settlement delay. For example, if plaintiffs have different time costs of pursuing a case, then the defendant can induce those plaintiffs with higher time costs to accept lower settlement offers now, while promising higher offers at a later date. The analysis thus extends the basic asymmetric information model of settlement by allowing the defendant to dictate not only the dollar amount of the settlement offer, but also the time at which it will be paid, where the latter can range anywhere from the time the dispute arises up to the (fixed) date of the trial. The model uses standard self-selection techniques to derive the conditions under which sorting is profitable for the defendant. The analysis also contributes to the growing literature, both theoretical and empirical, that examines the factors affecting the timing of the settlement of legal disputes. Interest in this topic is motivated in part by a desire to find better policies for facilitating settlement, thereby reducing the social costs associated with delay. The first model to introduce time explicitly into the economic model of the settlement-trial decision was by Spier (1992). Her model showed that the highest probabilities of settlement are at the beginning and the end of the negotiation process, a pattern that is consistent with observations of the actual settlement process [Spier (1992), p. 93]. Fournier and Zuehlke (1996) developed an empirical model, loosely based on Spier’s theoretical framework, of the causes of settlement delay. They found that several of the comparative statics that arise from static models of settlement continue to hold in a