Civil procedure serves a multitude of goals, from regulating the cost of fact gathering to dictating the rules of advocacy in court to promoting public participation in trials. To what extent can procedural design serve them all, or must rules sacrifice some interests to serve others? In this Article, we are the first to introduce a theory of procedural design that answers this question. We build upon the fundamental insight that the goals of civil procedure, as varied as they are, all occupy a common conceptual space—each addresses an externality, positive or negative, that litigation creates. This insight allows us to tie together distinct strands of scholarship on procedural design, develop a taxonomy of externalities that civil procedure addresses, and propose (sometimes radical) reforms that would allow procedure to serve more of its goals at once.

First, we show that the literature on procedural design has unraveled into three distinct strands. The first strand centers on the interest in reducing cost and delay in litigation. The second strand centers on the interests in limiting gamesmanship between the parties and improving court accuracy in decision-making. The third strand centers on the many related interests in the positive effects of procedure on society, such as developing legal precedent, deterring unwanted (primary) behavior, and so on.

Second, we tie together these strands of the literature by observing that each strand is focused on how procedure can address one type of externality. The first strand of the literature addresses what we call “system externalities”—the effects of actions on other cases in the same court or court system. The second strand addresses what we call “strategic externalities”—the effects of a party’s actions on opposing parties in the same case. The third strand implicates external effects on society as a whole, which we call “public-goods externalities.”