|
Article Excerpt
Abstract The idea that changes in Supreme Court decision rules should have measurable effects on the volume of cases litigated has a compelling plausibility, and several models of litigation predict this result. The prediction is a fragile one, however, because it implies very restrictive assumptions about the probability distributions of the cases subject to dispute. The period studied includes four Supreme Court decisions widely regarded as changing the rules and altering the level of uncertainty surrounding the legality of the anti-tying provisions of the antitrust laws. Broad trends in antitrust activity generally and changes in firm profitability statistically explain over three-quarters of the observed variation in tying litigation. Changes in legal precedent have only modest effects upon litigation.
Keywords common law efficiency * empirical antitrust * tying litigation
JEL Classification K21 * L42
Introduction
Antitrust policy in the United States has gone through a variety of changes since its genesis in the Sherman Act over a century ago. The treatment of some types of actions, such as horizontal conspiracy, evolved toward a standard of per se illegality in a relatively straightforward fashion, with the exception of an infrequent aberrant decision such as Appalachian Coals. (1) Other areas of antitrust have changed in the opposite direction. Tolerance of certain types of mergers, for example, has increased significantly since the period of populist hostility that characterized many decisions of the Warren Court era. Change in the legal status of other types of business behavior appears to be cyclical. For example, enforcement against tying arrangements has ranged from relative leniency during the very early years, to per se illegality beginning in the 1950s, toward a de facto rule of reason in the mid-1980s, and then back toward a more stringent standard during the last 10 years. The volume of tying litigation fluctuated substantially over the period in which standards were changing. This paper investigates these volume changes empirically and seeks to measure the influence of changes in Supreme Court precedent.
Most empirical analyses of antitrust activity date from Posner's massive statistical description of antitrust enforcement over the period 1890 through 1969 (Posner, 1970). Subsequent work has focused on explaining the determinants of enforcement. Some have studied antitrust litigation in specific industries (Lean, Ogur, & Rogers, 1985; Preston & Connor, 1992), while others have concentrated on the role of the allocation of public enforcement resources (Asch, 1975; Coate & McChesney, 1994; Ghosal & Gallo, 2001; Long, Schramm, & Tollison, 1973; Siegfried, 1975), and the effect of macroeconomic variables upon the volume of litigation (Bachmeier, Guaghan, & Swanson, 2004; Donohue & Siegelman, 1993; Siegelman & Donohue, 1995). The political determinants of antitrust litigation have been studied by economists (Amacher, Higgins, Shughart, & Tollison, 1985; Faith, Leavens, & Tollison, 1982) and political scientists (Eisner & Meier, 1994; Lewis-Beck, 1979). None of these papers, however, considered the possible influence of changes in Supreme Court precedent as a determinant of antitrust litigation.
The balance of this paper is organized as follows. The next section explains why changes in Supreme Court decisions would affect the volume of subsequent litigation and develops a simple model to develop testable predictions. A brief review of the changing status of tying illustrates why tying law provides a useful venue on which to test the model. The data and empirical results are then presented and a final section gives concluding perspectives on the results.
Why Does Precedent Matter?
The Supreme Court is the final arbiter of conflict between lower courts and its decisions help define and delimit the statutes created by the legislative branch. The power to set the rules has important implications for the efficiency of law enforcement and for the private and public costs of litigation. A strand of the law and economics literature suggests that common law legal systems tend toward an efficient set of rules through evolutionary selection. Laws that are privately inefficient generate legal challenges from those who stand to gain from removal of the inefficiency. Efficient laws therefore will be subject to less litigation than will be inefficient ones, and the excess burden produced by the latter will eventually be minimized through changes in case law as litigation is provoked by the initial inefficiency. The Supreme Court can affect this process through the impact of its decisions on the level of legal uncertainty. A decision that establishes a clear precedent for future cases will lower the level of uncertainty faced by potential litigants. If, on the other hand, a Supreme Court decision is ambiguous or contradictory then potential litigants face new uncertainties regarding the meaning and reach of the law. Some of the variance in litigation volume over time therefore may reflect adjustment to changes in precedent.
All litigation takes place within a context of uncertainty that has multiple sources. The degree of guilt of the defendant, the likely actions of the judge and jury, the possible strategic initiatives of the opposing side, and the legal standard to be applied cannot be known or predicted with complete confidence. A new legal precedent can potentially affect potential litigants in at least two ways. First, rule changes can alter the types of cases that are brought to trial. Consider a change in liability rules that substantially increases the plaintiff's burden of proof, for example. Suppose further that the rule change is explicit and clearly understood by all potential plaintiffs and defendants so that the level of uncertainty is unchanged. Disputes that would have been litigated under the previous looser standard now will not result in suits or will be settled before trial. The cases that become subject to litigation under the stricter liability regime now will be drawn from those that likely would have resulted in a decision for the plaintiff under the looser standard and therefore would have been settled without trial.
A second way in which changes in legal precedent can affect litigation is by creating additional uncertainty if a new rule is ambiguous. A decision that is contradictory or that stipulates standards of proof that are untested and unfamiliar will increase the variance in the estimates of liability by potential litigants. The additional uncertainty may be transitory to the extent that it can be reduced by subsequent decisions. (2)
These ideas are illustrated below. Figure 1 contains two alternative probability density functions over liability which differ with regard to the degree of legal uncertainty. Each distribution is centered on the plaintiff's estimate of the degree of liability of the defendant in a given case. (3) The decision rule specifies the degree of liability necessary for a finding of guilt and is assumed to be determined by the precedents established by previous court decisions. Possible decision rules are illustrated by vertical lines D and D*. The proportion of the distribution lying to the right of the decision rule therefore represents the plaintiff's probability estimate, before trial (4), that the defendant will be found guilty. A court decision that simply raised the standard of proof without increasing legal uncertainty would move the decision...
|
