Economic Crime: Antitrust Offenses

Congress effectively delegated the formulation of antitrust policy—including criminal antitrust policy—to federal judges. Have they interpreted the Sherman Act to confine its criminal reach only to people who indeed are morally blameworthy? The Supreme Court has delivered mixed results on this score.

Before turning to the cases, however, one must confront an initial question: should not this culpability issue be a concern for prosecutors during case selection, and not one for judges during statutory interpretation? Supreme Court justices could and indeed once did consign concerns about the culpability of defendants entirely to prosecutorial discretion, but in practice they no longer do so (Wiley, 1999, pp. 1058–1068, 1160–1161). Since 1985, the Supreme Court has interpreted federal criminal statutes on the apparent premise that Congress means to permit federal prosecutors to prosecute only morally blameworthy people (Wiley, 1999, pp. 1026–1056).

The Supreme Court's jurisprudence about criminal antitrust predates this interpretive shift. There are two main cases: Nash v. United States, 229 U.S. 373, 377–378 (1913) and United States v. U.S. Gypsum Co., 438 U.S. 422 (1978). Nash ruled that criminal application of the Sherman Act was not unconstitutionally vague, but the decision did not set forth the elements prosecutors must prove in a criminal antitrust action. The Gypsum decision tackled just this task. Gypsum involved three pertinent holdings, which respectively seem (1) attractive; (2) questionable; and (3) unpersuasive and troubling.

The first holding—the attractive one—was the Court's conclusion that "[w]e are unwilling to construe the Sherman Act as mandating a regime of strict-liability criminal offenses" and therefore that "the criminal offenses defined by the Sherman Act should be construed as including intent as an element" (pp. 436, 443). This holding is attractive because criminal violation of the Sherman Act exposes a person to a potential three-year prison sentence. Without a showing of bad intent, there would be no guarantee that this person is morally blameworthy. To imprison a blameless person would be unjust.

The Court's second holding—the questionable one—was its decision (at 448 n.23 and 444) that "knowledge of the probable consequences of conduct [is] the requisite mental state in a criminal prosecution like the instant one where an effect on prices is also alleged." The Court summarily dismissed other levels of culpability with the opaque statement that, "[i]n dealing with the kinds of business decisions upon which the antitrust laws focus, the concepts of recklessness and negligence have no place." This second holding is mystifying and unjustified because the Court did not say why the concepts of recklessness and negligence have no place in dealing with business decisions. In contrast, the Model Penal Code recommends "recklessness" and not "knowledge" as the correct culpability default because recklessness is "the basic norm [that] usually is regarded as the common law position" (Model Penal Code §2.02(3) cmt. 5 (1985)). Jeffries and Stephan likewise observe that "the minimum culpability most widely found in the penal law is recklessness" (p. 1372). The Gypsum Court's preference for culpability at the level of knowledge rather than recklessness remains questionable (see Wiley, 1999, pp. 1111–1128). Procedurally this issue now seems fixed in concrete, however, because it seems extremely unlikely that any prosecutor would seek jury instructions that violate this rule simply on a remote prospect of eventual review in the Supreme Court, which is the only court with power to revise this Gypsum holding. Without any apparent prospect of Supreme Court review or congressional revision, this point seems of purely academic interest.

The Court's third holding—the unpersuasive one—is about the type of criminal intent the government must prove. The Gypsum decision failed to require the government to prove that the defendants were aware that they were acting wrongfully or illegally, which seems a logical state of mind to require if a court seeks (as the Gypsum decision sought, see 438 U.S. at 442) to guarantee that defendants are morally culpable. A standard of this sort is what the government must prove, for instance, in drug and tax prosecutions. The typical Ninth Circuit jury instruction for drug cases (No. 9.13) requires the government to prove that defendants knew that they possessed "some kind of a prohibited drug," while in tax evasion prosecutions the government must prove defendants knew of the duty imposed by law and intentionally violated that duty (Cheek v. United States, 498 U.S. 192, 201 (1991)). At least one district court has imposed a standard of this kind in a criminal antitrust case by requiring the jury to find both that "the defendants knowingly joined a conspiracy whose purpose was illegal and that they understood the illegality of that purpose" (United States v. Brown, 936 F.2d 1042, 1046 n.3 (9th Cir. 1991)). Proof of this kind of culpability has been readily available in recent cartel prosecutions, where defendants have shown their consciousness of guilt by using elaborate concealment precautions, false names, and the other standard tools of people with something to hide. If prosecutors in a particular case find it difficult to prove that defendants were aware of the illegal or wrongful nature of their conduct, this difficulty is a good reason for prosecutors to reexamine their decision to prosecute that case as a criminal matter.

Rather than require awareness of wrongful or illegal conduct, however, the Gypsum Court apparently held the government must prove an entirely different intent: awareness of the probable consequences of conduct. The decision phrased this requirement in slightly different ways: "knowledge of the probable consequences of conduct" (438 U.S. at p. 448 n.23); "action undertaken with the knowledge of its probable consequences" (id. p. 444); and "the perpetrator's knowledge of the anticipated consequences" (id. p. 446). The Court did not elaborate upon this point, but it appears that all of these formulations fail to guarantee that convicted persons are morally culpable. Take United States v. Topco, for instance. This oft-cited 1972 antitrust decision involved independent grocers from different regions who formed a buying cooperative and a private label brand called Topco. Each participating grocer owned an equal share in Topco, and each received an exclusive territory in which to market the new Topco brand. These grocers were the small firms in their regions; each one held only six percent of the market on average. Scholars have demonstrated convincingly that the grocers' conduct was beneficial to consumers and society (Baxter and Kessler, pp. 628–629; Bork, pp. 274–279; Hovenkamp, pp. 205–206). Yet the Supreme Court ruled that these defendants had committed a per se violation of the Sherman Act. Had the government opted to proceed criminally instead of civilly against the Topco defendants, it apparently could have obtained a conviction against them under the Gypsum standard because it seems clear that these defendants had knowledge of at least very many of the probable consequences of their group conduct. As far as one can tell, however, the Topco defendants were morally blameless people whose only sin was a good-hearted attempt to compete against the much larger grocery chains of A&P, Safeway, and Kroger—the three firms that the Court's decision identified as the market leaders. To imprison the Topco defendants would have been unjust, yet the Gypsum interpretation of the Sherman Act would have permitted it. Topco is not a fluke; a similar analysis could be performed with Sealy, Associated Press, and other cases. Gypsum's third holding thus is unpersuasive and troubling.

Supreme Court developments since 1985 may have cast doubt on the soundness of Gypsum's interpretation of the Sherman Act. (For the moment, put aside the mental state debate about knowledge versus recklessness and accept Gypsum's holding that knowledge is the right level of awareness.) On the matter of Gypsum's third holding, a better formulation would require the government to prove that the defendant in a price fixing case (1) agreed with a competitor (2) about price (3) knowing that this conduct was illegal or wrongful. This formulation goes beyond Gypsum and is at odds with some lower court case law (see, e.g., United States v. W.F. Brinkley & Son Construction Co., 783 F.2d 1157, 1162 (4th Cir. 1986)), but it would accord with the thrust of the Supreme Court's general criminal interpretive jurisprudence since 1985. It would assure that federal antitrust prosecutions can imprison only people who are morally culpable.

One might hope that the Department of Justice would adopt this formulation as a matter of self-restraint, and would propose it in the jury instructions the Antitrust Division offers to district courts in criminal cartel cases. Should the Antitrust Division decline this measured self-restraint, it will open itself to challenges from defense counsel seeking jury instructions that require proof that defendants knew their conduct was illegal or wrongful. There are two risks for the government in this course. The first is that the district court may agree with the government but that an appellate court may not. The government then would face the need for a costly retrial of a case, and the passage of time never improves a case-in-chief. The second risk is that the appellate court may formulate jury instructions differently than would the government, and these instructions then would be chiseled in appellate stone. Self-imposed matters of discretion retain more flexibility than do judicial dictates. Arguments of tactical prudence as well as of principle thus support the case for prosecutorial self-restraint in antitrust cases.