A Dynamic Theory of Common Law Courts

We develop a model that determines when and how time-consistent and forward-looking courts should set and reform legal rules (a normative theory for dynamically efficient courts). We explicitly take into account that: 1) the optimal rules most likely are not the same for all periods of time; 2) courts can only rule at trials; 3) the enforcement strategies of courts determine the litigation strategies of present and future parties in conflict; and 4) the parties in conflict can contract around the rules. As main results, we show that: 1) courts should set those rules that maximize the value of the present parties in conflict (statically efficient rules) only under extreme circumstances; 2) if legal rules are the only control variables, courts should adjust the unconstrained first-best rules for society in order to give the parties incentives to partially correct an inefficient frequency of litigation; 3) there always exists a distribution of the litigation expenses between the parties that generates an optimal frequency of trials in which case courts don't need to bias the rules. The model allows us to analyze the social desirability of two commonly suggested strategies to increase the frequency of shareholders' litigation: adding ambiguity to the law or involving public prosecutors as the N.Y.A.G.'s office or agencies as the S.E.C. In addition, the model also allows us to discuss when courts should set contingent rules (rules that adapt to the states of nature) instead of rigid rules (rules that don't adapt to the states of nature).