On the Costs of Civil Justice

Commentary Charles Silver's paper, Does Civil Justice Cost Too Much?, takes on the daunting task of evaluating transaction costs in the American litigation system and assessing the results of empirical studies on the costs of litigation and its alternatives. His goals are threefold. First, he wants to raise skeptical questions about the ubiquitous claim that American litigation is too expensive-that we have too many lawsuits, too many lawyers, too broad discovery, that lawyers charge too much, that the system is slow, complex, and ridden with inefficiency. Second, he wants, in evaluating this claim, to offer a more adequate conceptual model of what we're talking about-to clarify terminology, clear away confusion, and endogenize factors that are often omitted from the conventional analysis. Finally, he wants to evaluate the existing empirical literature on litigation costs with a view towards assessing the models employed and the results obtained. Silver has made an excellent start at grappling with these complex and important issues. Silver is on target in observing that litigation is really a species of claims settlement. A lawsuit is essentially a sale. The defendant buys a valuable asset from the plaintiff, in the form of a release of claims if the case is settled, or a verdict with res judicata effect if the case goes to a verdict. If the defendant wins the suit, the sale price is zero. Otherwise, the sale price is the amount of any settlement or verdict. As a form of sale, a lawsuit has much in common with other economic transactions. But it has certain features that, taken together, make it an unusual, even unique transaction. Four features are most salient. First, the sale in litigation occurs in a bilateral monopoly. In most cases, the plaintiff can sell res judicata only to the defendant, and the defendant can buy it only from the plaintiff. This generalization requires qualification, since there is a market in litigation claims, however imperfect-the most obvious example of which is the contingent fee, which is a partial sale of the claim to the attorney. However, it is still usually the case that the parties to litigation must transact with one another and not with third parties. Second, the sale effected by litigation is a forced sale. The verdict forces the defendant to pay the verdict amount (which, as noted, will be zero if the defendant wins). If the defendant does not pay, the plaintiff can take him to court to enforce the judgment. Eventually, the government will send out men or women with guns to encourage the defendant to make good. Of course, most cases are settled, as Professor Silver observes. But this does not detract from the fact that a sale will be forced if settlement negotiations fail. A third distinctive feature of litigation sales is that the value of the claim to be sold is often very difficult to appraise. Litigation claims are unique, not fungible, and often involve important elements of value known only to one party at the outset of the transaction. The value of a litigation claim usually becomes more determinate as the case progresses, both because the parties will have the benefit of discovery and because the legal issues may be clarified by the judge through preliminary rulings or in settlement conferences. Civil litigation is, in a sense, a process of appraisal. But it is conducted differently from ordinary appraisals. Fourth, the sale transaction in litigation creates externalities or third party effects. Litigation may generate judicial precedents, which are a public good no matter how much we law professors may criticize them. And, equally if not more important, litigation sets a price for behavior. It is because litigation sets a price that law accomplishes its deterrent effect. Again, qualifications are in order. Price-setting is not a unique feature of litigation sales. All sale transactions in any commodity contribute to setting price so long as the transactions and the price are publicly observable. …