Essays on committee decision making

This doctoral dissertation studies relationships between voting rule, social preference, and information structure with an economic model of jury decision making.;In Chapter 1 we consider three kinds of conservativeness in a committee decision making problem: a conservative voting rule, a conservative threshold of reasonable doubt, and a conservativeness in the legal information system. The reason why unanimity performs worst in Feddersen and Pesendorfer (1998) is that strategic voters compensate for the bias brought upon by the voting rule. If the legal information system has the opposing bias (less conservative) to that of the voting rule (more conservative), we can restore informative voting to being rational. We show that a sufficient condition to lower the lower bound of the probability of a convicted defendant being innocent is to make information system more informative in the sense of Blackwell (1953). We also show that another sufficient condition is to have a more conservative threshold of reasonable doubt in committee members' homogeneous preferences.;In Chapter 2 we adopt the canonical utility specification of the information aggregation through voting literature but use it in a cost minimization and preference aggregation setup. Using this model, we show that the judge's decision is always better than that of the jury if the judge has the same utility parameter as the social planner. A judge internalizing social cost is better for the society than jury members randomly selected from the population since the jury's decision is inherently noisy. In general cases where the judge's utility parameter is different from that of the social planner, we identify a condition under which the decision by jury is better than that of the judge. When the social planner can choose a voting rule ex ante, the optimal voting rule is a weakly increasing function of the societal threshold of reasonable doubt. Under certain distributional assumptions, we illustrate thresholds of reasonable doubt that can justify unanimity and other majority rules. Considering jury selection, we show that peremptory challenges are conferred in a way that favors the defense if the societal threshold of reasonable doubt is high enough. Finally, we show, with an example, that unanimity rule is necessary to make the voting result representative of unbiased population.;In Chapter 3 we review commonly identified differences between common law and civil law legal traditions. Although it becomes harder to draw a clear line between the two legal traditions, identifying differences can be done at least in theory. We review the economic literature that has modeled and studied these features. Lastly, we show how legal tradition study and other observations can be linked to the study in previous chapters of this dissertation.