ABSTRACT
We conduct a laboratory experiment using a stylized model of civil litigation bargaining. This is a screening model with a continuum of types based on Bebchuk (1984). Most previous experimental work on the screening model has utilized two-type models. In a screening model, the recipient of the settlement offer is informed, but the party making the offer is uninformed. Among other things, this setting embeds an ultimatum game as the subjects need to divide the joint surplus which is achieved via settlement rather than trial. After the informed plaintiff learns her type, she indicates the minimum offer she is willing to accept. This allows us to measure the demand for fairness within this setting. Previous work with twotype models has shown a reduction in fairness behaviors within this embedded ultimatum game. While there is significant dispersion about the median, by this measure of the central tendency, the demand for fairness essentially disappears. We also test a prediction of the theory that the dispute rate is independent of the distribution of trial costs, depending only on their sum. In the baseline, costs are equal, but there are two treatments, where under one, the plaintiff bears most of the costs of trial, while under the second, the defendant bears most of these costs. The theory does fairly well, as we find only weak evidence that dispute rates are affected by the distribution of dispute costs.
Pecorino, Paul and Solomon, Michael and van Boening, Mark, The Screening Model of Litigation with a Continuum of Types: An Experimental Analysis (September 15, 2026).
Leave a Reply