Statutory inflation and institutional choice.

• Author: Solan, Lawrence M.

INTRODUCTION

1. FOUR MODELS FOR INTERPRETING STATUTES WITH CIVIL AND CRIMINAL REMEDIES

   1. The Standard Model

      1. Two Basic Principles

         1. Broad Interpretation of Remedial Statutes

         2. The Rule of Lenity

      2. Application of the Standard Model:

         Two Examples

         1. The Copyright Act

         2. Bankruptcy Crimes

   2. The Inflationary Model

      1. Insider Trading and Inflation of the Securities Laws

      2. Antitrust Laws

      3. Environmental Law

   3. The Lenity Model

   4. The Law Enforcement Model

      1. Mail Fraud as a RICO Predicate Act

      2. Criminal Mail Fraud and Criminal RICO

2. CONTROLLING STATUTORY INFLATION

   1. How Much Should Statutory Inflation Be Controlled?

   2. Imposing Strong Mens Rea Requirements

   3. Writing Rules of Construction into the Statute Itself: The Wrong Path

   4. Modifying the Chevron Doctrine in Criminal Cases

3. CONCLUSION: STATUTORY INFLATION AND INSTITUTIONAL CHOICE

   INTRODUCTION

   Statutes that contain both civil remedies and criminal penalties are typical in the modern regulatory state. Among them are RICO, (1) the antitrust laws, (2) the securities laws, (3) environmental laws such as the Clean Water Act, (4) various tax statutes, (5) the Copyright Act, (6) and the Bankruptcy Code. (7) These statutes most often regulate business affairs and provide for criminal penalties to punish willful violators. The Sarbanes-Oxley Act of 2002, (8) the statute enacted in response to the financial scandals of Enron, WorldCom, and other major corporations, provides a recent example of this remedial scheme, which remains popular with lawmakers.

   This Article argues that the standard array of interpretive tools employed by judges and academics does not adequately account for important irregularities in the ways that courts construe these dual-remedy statutes, especially in criminal cases. Most of the literature focuses on two institutions: the legislature and the courts. To the extent that administrative agencies become part of the mix, the main issue addressed is how much deference courts should give to the interpretation of a statute by the agency that has been charged with enforcing it. This Article, in contrast, suggests that the institutional setting in which a statute is enforced has a profound effect on how courts construe the statute over time. Although deference doctrines play a role, judicial decisions also reflect the resources committed to litigation by governmental institutions in civil and criminal cases.

   Even on the surface, this kind of legislative scheme creates complex interpretive problems. Deeply entrenched in our system of statutory jurisprudence are two complementary canons of construction: Remedial statutes are interpreted liberally; penal statutes are interpreted narrowly. (9) The former reflects concern about public health and welfare, the latter the due process rights of those accused of committing crimes. Should courts interpret these dual-remedy statutes narrowly in criminal cases, but broadly in civil cases, thus creating two different readings of each statute? Should they interpret the statute narrowly in both kinds of cases to be fair to criminal defendants without creating multiple interpretations of the same language? Should they interpret the statute broadly in both criminal and civil cases, forsaking the value of lenity in order to promote both uniformity and aggressive regulation in civil cases? Should they sometimes interpret a statute broadly in criminal cases but narrowly in civil cases, perhaps to limit private litigation when the statute is basically a criminal one? Each of these possible solutions requires the courts to weigh values that would be noncontroversial were they not in conflict with one another. Institutional choices that determine the ways these statutes are enforced influence the weight that courts ultimately give in different statutory settings.

   Table 1 shows the four interpretive possibilities for dual-remedy statutes which I have named the standard model, the inflationary model, the lenity model, and the law enforcement model.

   Much of this Article will be devoted to examining the inflationary model. When courts interpret a statute broadly enough in civil cases to further its regulatory goals, the broad interpretations sometimes spill over to criminal cases, causing an increase in criminal liability. Conduct that was not criminal in the past has become criminal without any legislative action. (10) I call this phenomenon "statutory inflation."

   To illustrate with an example that will be discussed in more detail later, courts do not generally apply statutes extraterritorially. (11) Yet, for perfectly good reasons, the Supreme Court held in Hartford Fire Insurance Co. v. California, (12) that the antitrust statutes should be applied extraterritorially when their violation has an effect in the United States. (13) If it were otherwise, companies could legally conspire to gain monopoly power in the United States simply by meeting abroad. Thus, the decision is consistent with the notion that remedial statutes are to be interpreted broadly.

   Prior to Hartford Fire, it would have been unlikely for a court to apply United States antitrust laws internationally in a criminal case. Criminal statutes are generally interpreted narrowly, and nothing in the antitrust laws says that Congress intended any exception to the extraterritoriality rule. But once the courts interpret a statute broadly in a civil case, the ruling becomes part of the meaning of the statute for subsequent judicial analysis. (14) Over the years, antitrust regulators and the antitrust barave bar absorbed the notion that extraterritorial conspiracies are unacceptable. As a result, the First Circuit recently refused to apply the rule of lenity in a criminal antitrust case involving extraterritorial conduct. (15)

   Statutory inflation will most likely occur when both an administrative agency and a division or bureau of the Department of Justice push for expansive interpretation of a statutory scheme. One informal, but important device that serves to limit the expansion of criminal liability is prosecutorial discretion. With heavy caseloads and the chance of losing a case that rests on uncertain legal grounds, prosecutors have an incentive to devote most of their resources to clear instances of criminal wrongdoing. (16) That cost-benefit analysis changes when an agency contains an enforcement division whose only job is to bring civil actions under either a single statute or a small set of related statutes, and the Department of Justice has a bureau whose only job is to bring criminal cases, largely those that the agency refers for prosecution. Together, they are more likely to pursue more marginal cases, and to ask courts to interpret the law expansively.

   Moreover, when an agency interprets a statute through a formal process, its interpretation is ordinarily reviewed deferentially under the Chevron doctrine, according to which courts defer to the agency's interpretation of an ambiguous statute, as long it is reasonable. (17) As a result, agencies are free to promulgate rules that might be much broader than Congress intended when it delegated authority to them. Although the application of the Chevron doctrine is consistent with the broad interpretation of remedial statutes, if Congress also decides to criminalize the willful violation of regulations, statutory inflation is likely to occur. (18)

   Significantly, statutory inflation occurs even when courts believe that they are applying the standard model. It is by and large an unnoticed by-product of the architecture of legal institutions empowered to enforce the laws. Beyond that, inflationary pressure is increased because courts are reluctant to maintain two separate sets of readings for multiple-remedy statutes: a narrow reading for criminal cases and a broad reading for civil cases. This reduces the appeal of the standard model, thereby increasing the need to seek alternatives.

   Although the other two interpretive models--the lenity model and the law enforcement model--play less prominent roles, they are both very much alive. The lenity model calls for a statute to be interpreted narrowly in all cases, in order to protect against statutory inflation in criminal cases resulting from broad interpretation in civil ones. (19) It can result in underenforcement of remedial legislation. The Supreme Court has recently used this model to interpret a statute that taxes certain firearms, (20) and to interpret a statute that governs the conduct of former government officials after they leave public office. (21) Finally, the law enforcement model calls for broad interpretation of a statute in criminal cases, but narrow interpretation in civil cases. It is at odds with the standard model, and on its face inconsistent with basic principles of statutory interpretation. Yet courts use it occasionally. For example, courts are sometimes reluctant to give civil RICO plaintiffs the broad latitude that they give federal prosecutors when the predicate acts for RICO are multiple mail fraud offenses. (22)

   Part I of this Article describes the four models in more detail and illustrates their application. Although courts' descriptions of their interpretive decisions seem to suggest that they randomly apply conflicting canons of statutory construction, this rhetoric masks important institutional considerations that drive the array of decisions. The likelihood of statutory inflation is largely a function of the institutional setting in which the statute is enforced. Thus, substantial inflation has occurred in the interpretation of antitrust, environmental, and securities laws. In contrast, there has been little inflation in criminal cases interpreting the Copyright Act or the Bankruptcy Code, which are basically interpreted according to a weaker version of the standard model. The Bankruptcy Code is especially interesting because the judiciary is concerned about underenforcement, and has been moving in recent years to...