Decision Theory and the Roberts Court’s Antitrust Decisions

In considering how the Roberts Court’s antitrust jurisprudence reflects a decision-theoretic approach to antitrust adjudication, it is helpful to divide the

Court’s antitrust decisions into two groups: those concerning substantive liability rules and those concerning procedures and immunities.[1] [2]

Decisions Concerning Substantive Liability Rules


The nonprocedural Roberts Court antitrust decision that has received the most criticism for being probusiness, anticonsumer, and “radical” is Leegin Creative Leather Products, Inc. v. PSKS Inc.,13 which held that instances of minimum resale price maintenance (RPM)—an upstream seller’s setting of the minimum price downstream sellers may charge for its product—are not per se illegal but must be evaluated on a case-by-case basis under antitrust’s rule of reason. For example, in the wake of the Leegin decision, a Baltimore Sun columnist boldly predicted that the “[electronic bargains of today will be gone by next year” and stated that “for that you can thank that radical activist Gang of Five on the U.S. Supreme Court,” which “slipped through a decision that overturned 96 years of antitrust law.”[3]

The perception that Leegin represents a radical, probusiness/anticonsumer shift stems from the facts that (1) it overruled a long-standing precedent (the 1911 Dr. Miles decision, which had declared minimum resale price maintenance to be per se illegal[4]); (2) it was a 5-4 decision that pitted the Court’s traditional “conservatives” (Scalia, Kennedy, Thomas, Roberts, and Alito) against its traditional “liberals” (Stevens, Souter, Ginsburg, and Breyer); and (3) it permitted the imposition of price floors, which one would expect to raise consumer prices.

In contrast to its characterization in the popular press, however, Leegin was not particularly controversial among mainstream antitrust scholars. For example, Harvard Law School’s Einer Elhauge, who chaired the Obama campaign’s antitrust advisory committee and is in no sense a “conservative” on antitrust issues, praised the Leegin holding and characterized it as reflecting a consensus view among the leading schools of antitrust analysis.[5] As Elhauge observed, the first two “troubling” facts about Leegin—that it overturned a ninety-six-year- old precedent and was decided on a 5-4 basis—are not really troubling at all. Deference to precedent is less important, and less expected, in antitrust cases than in other statutory cases, for courts have long viewed the Sherman Act’s broad, amorphous text as a delegation to craft a quasicommon law reflecting the ever-expanding insights of economics. Given that economic thinking on RPM has evolved dramatically since Dr. Miles was decided in 1911, it is entirely appropriate that the substantive antitrust rules evolve accordingly.17

But what about the third “troubling” fact about Leegin—that it sanctions manufacturer-imposed price floors that are likely to increase consumer prices? Viewed through the lens of decision theory, that aspect of the decision is likely proconsumer. To see why this is so, consider how Dr. Miles’s per se rule compares to Leegin s rule of reason approach in terms of the three decision-theoretic considerations set forth in the preceding text: (1) the likelihood that the rule will generate an incorrect result; (2) the magnitude of loss that will result from the sort of errors the rule is likely to produce (collectively, these two considerations determine the rule’s expected error costs); and (3) the difficulty of administering the rule (this determines the rule’s expected decision costs).

With respect to the third consideration, difficulty of administration, the per se rule performed fairly well, though perhaps not as well as one might initially suppose. Although per se rules, which impose antitrust liability without regard to actual anticompetitive effect, are usually easy to implement, Supreme Court precedents creating exceptions to Dr. Miles had rendered the inquiry somewhat complicated in RPM cases. Because of the Supreme Court’s decision in United States v. Colgate & Co.,18 a plaintiff complaining of RPM had to show that the manufacturer had not simply adopted a unilateral policy of refusing to deal with discounters. In cases involving the termination of price-cutting dealers, the Court’s decisions in Monsanto Co. v. Spray Rite Service Corp.19 and Business Electronics Corp. v. Sharp Electronics Corp.,20 required the plaintiff to show that the defendant manufacturer had entered an “agreement” with compliant dealers and that the agreement at issue dictated minimum prices, not simply nonprice [6] [7] [8] [9]

matters. Borne out of misgivings about Dr. Miles, the Colgate, Monsanto, and Business Electronics decisions substantially enhanced the decision costs associated with the per se rule. Nevertheless, Leegins rule of reason approach will likely increase the costs of administration.

With respect to the other two decision theory considerations, however, the Leegin approach represents a tremendous improvement over Dr. Miles. When it comes to likelihood of error, Dr. Miles was a disaster. Because, as the Leegin majority explained, both economic theory and empirical evidence suggest that most instances of RPM are procompetitive, the per se rule’s automatic condemnation of all RPM was highly likely to generate Type I errors (false convictions).

First consider theory. Manufacturers that distribute their products through retailer-dealers make their money on the sale to the dealers, not the resale to consumers. Their chief objective is to maximize purchases by dealers, whose demand will be determined by consumer demand at the retail level. If a retailer’s markup rises but the retailer does not otherwise alter its conduct, consumers will reduce their purchases, and the manufacturer, who captures none of the retail markup, will lose sales and profits. Because RPM has the effect of increasing retail markups, manufacturers will normally avoid it unless the greater retailer margin it provides induces dealers to provide services that enhance demand for the manufacturer’s product and thereby increase consumer sales, in which case the RPM is output enhancing and procompetitive.

As the Leegin majority explained, RPM tends to promote demand-enhancing dealer services upon which other dealers may free ride (training, product demonstration, etc.), because it prevents low-service, low-cost dealers from profiting by underselling their high-service rivals.[10] In addition, RPM may promote demand-enhancing dealer services that are not susceptible to free riding. By coupling an RPM policy (which guarantees dealers an attractive profit margin) with a liberal right of termination (which can be exercised against dealers with poor sales records), a manufacturer encourages dealers—retailing experts—to use their own energy and innovation to promote the manufacturer’s brand so as to protect the RPM-protected profit margins. RPM thus provides an efficient mechanism for inducing output-enhancing dealer services that are difficult to secure through a contract.[11] In addition, RPM may provide great assistance to a manufacturer that is a new entrant into a product market; by assuring retailers of a guaranteed margin on the manufacturer’s brand, the manufacturer’s RPM policy encourages retailers to take a chance on the new brand, afford it favorable shelf space, promote it over established brands, and so forth.[12]

But manufacturers may also demand (or at least concede to) RPM for anticompetitive reasons. Scholars have identified, and the Leegin majority acknowledged, four such reasons. First, a retailer cartel may demand that manufacturers impose RPM as a means of shoring up a retailer-level conspiracy. Second, a dominant retailer may demand RPM in order to protect itself from more efficient retailer rivals. Third, colluding manufacturers may collectively impose RPM as a means of increasing price transparency (which aids in cartel enforcement) and discouraging cartel participants from cutting prices to dealers. And finally, a dominant manufacturer may use RPM to “bribe” retailers to disfavor nondominant brands (lest they lose the RPM-guaranteed profit margin).

Thus, as the Leegin majority acknowledged, we have two sets of theories— one procompetitive, one anticompetitive—as to why manufacturers would demand or concede to enhanced retail margins using RPM. Which set is likely to explain most instances of RPM? Probably the procompetitive theories, for the preconditions to those theories, unlike those for the anticompetitive theories, are frequently satisfied.

The anticompetitive explanations for RPM are plausible only under highly restrictive sets of conditions:

  • • The retailer cartel theory is plausible only where the retailer market is susceptible to cartelization (which is extremely uncommon, given the low barriers to entry into retail markets) and either (1) the manufacturer’s brand is unique; or (2) RPM is also imposed on competing brands of the product (otherwise, consumers would respond to the RPM by switching to another brand).
  • • RPM may operate as a device by which a dominant retailer excludes more efficient retailer rivals only where it is imposed so broadly (i.e., on so many brands) that the more efficient retailers are unable to gain a foothold.
  • • The manufacturer cartel theory is plausible only where the manufacturer market is susceptible to cartelization (e.g., it is concentrated or entry barriers are high), and RPM is commonly employed throughout the market (otherwise, the RPM could not operate to police the cartel).

• The manufacturer exclusion theory is plausible only where the profit margin provided by RPM is significant enough to bribe retailers to drop or disfavor other brands, and the RPM scheme is implemented broadly among retail outlets (so that “foreclosed” brands cannot simply distribute through other retailers—such as any of the ubiquitous discount retailers who compete primarily on price and would be unlikely to forego carrying a lower priced product in exchange for a higher retail margin).

Each of these sets of circumstances is uncommon.

By contrast, the preconditions for procompetitive uses of RPM are frequently satisfied. RPM may be used to ensure point-of-sale services that might be the subject of free riding whenever such dealer-provided services enhance demand for a manufacturer’s product and are susceptible to free riding (because, e.g., dealers are in located within close proximity of each other). RPM may provide an optimal means of ensuring dealer performance of unspecified agreements whenever dealer activities would enhance the attractiveness of a manufacturer’s offerings, and the quality-enhancing activities are difficult to delineate in advance or to monitor. RPM may facilitate entry whenever a new producer seeks to gain access to or promotion by retail outlets that already stock and provide favorable shelf space to well-established brands. Because these various conditions quite often exist, procompetitive rationales for instances of RPM, unlike anticompetitive effects, are frequently plausible. Thus, economic theory predicts that most instances of RPM will be procompetitive, not anticompetitive, and that a rule of per se illegality will have a high error rate.

The empirical evidence on RPM’s effects confirms this prediction. As the Leegin majority observed, both a detailed staff report of the Federal Trade Commission’s Bureau of Economics[13] and an investigation of litigated RPM cases[14] suggest that most instances of RPM are procompetitive rather than anticompetitive. While the dissent similarly sought to invoke empirical evidence in support of retaining the rule of per se illegality, the evidence it cited was inapposite. It consisted of studies purporting to show that prices were higher in “fair trade” states (which, for a period of time, were permitted to declare RPM to be per se legal) than in states that did not provide immunity for RPM. Those studies are not convincing because (l) even the procompetitive accounts of RPM involve higher consumer prices (which induce output-enhancing dealer services), so the existence of such price increases says nothing about the competitive effects of RPM; and (2) fair trade states adopted a rule of per se legality, not a rule of reason, for RPM, so the results under fair trade would not necessarily follow from applying the rule of reason to RPM. It seems, then, that both economic theory and empirical evidence suggest that Dr. Miles’s automatic condemnation of RPM arrangements is far more likely to generate errors than is the rule of reason approach Leegin embraced.

The final decision-theoretic consideration is the magnitude of loss from errors. The Dr. Miles rule produced more false convictions (Type I errors) than will Leegin s rule of reason, which will likely tend to reduce errors overall but may raise the incidence of false acquittals (Type II errors). When it comes to antitrust, though, false convictions tend to produce greater social loss than false acquittals. When an anticompetitive instance of RPM is improperly approved, social cost (allocative inefficiency) may result from market power that is created or maintained. When a procompetitive instance of RPM is improperly condemned, by contrast, the social cost consists of the immediate benefit foregone by stopping the challenged instance plus any future benefits that are thwarted because of the precedent condemning that particular type of efficient conduct. Whereas the former harm—market power—is generally self-correcting by entry or, in the case of collusion, cheating, the latter harm—economy-wide thwarting of an outputenhancing practice—may be undone only by a court decision (or legislative or regulatory development) that corrects the bad precedent. False convictions are, therefore, more likely to cause greater and more durable harm than false acquittals and should thus be more stridently avoided by the governing liability rule.[15]

The holding of Leegin is thus wholly defensible from the perspective of decision theory. Both the “likelihood of error” and “magnitude of loss from expected errors” considerations weigh heavily in favor of rule of reason adjudication. While the “difficulty of administration” consideration might support adherence to the Dr. Miles rule, that concern is not so compelling. The Dr. Miles rule was actually fairly difficult to implement, and the Leegin majority helpfully constrained decision costs by noting specific factors courts should consider in evaluating instances of RPM and by directing the lower courts to craft a structured rule of reason. Decision theory therefore supports the outcome in Leegin.

  • [1] I will not discuss the Court’s decision in Volvo Trucks North America, Inc. v. Reeder-Simco GMC,Inc., 546 U.S. 164 (2006). While that decision technically involved a dispute under the antitrust laws,the provision at issue, the Robinson-Patman Act, 15 U.S.C. §13(a), is expressly not focused on consumer welfare and is thus not readily amenable to the decision-theoretic approach set forth in thepreceding text.
  • [2] 551 U.S. 877 (2007).
  • [3] Mike Himowitz, Electronic Bargains of Today Will Be Gone by Next Year, Baltimore Sun, June5, 2007, at 7D.
  • [4] Dr. Miles Med. Co. v. John D. Park & Sons Co., 220 U.S. 373 (1911).
  • [5] Einer Elhauge, Harvard, Not Chicago: Which Antitrust School Drives Recent U.S. Supreme CourtDecisions?, 3 Comp. Pol’y Int’l 59, 61-62 (2007).
  • [6] The dissent was authored by Justice Breyer, a former antitrust professor who was certainlyaware of the consensus to which Elhauge refers. A careful reading of this dissent, however, suggeststhat the dissenters were more concerned with cementing the notion of “super-precedent,” a key concept for defenders of the prevailing Supreme Court case law on abortion rights, than with contestingthe actual majority holding. Leegin, 551 U.S. 877. Thus, if one focuses solely on the antitrust issuespresented in the case, Leegin is far less controversial than the Court’s 5-4 vote would suggest.
  • [7] 250 U.S. 300 (1919).
  • [8] 465 U.S. 752 (1984).
  • [9] 485 U.S. 717 (1988).
  • [10] Lester G. Telser, Why Should Manufacturers Want Fair Trade?, 3 J.L. & ECON. 86, 89-96(1960).
  • [11] Benjamin Klein & Kevin M. Murphy, Vertical Restraints as Contract Enforcement Mechanisms,31 J.L. & Econ. 265 (1988).
  • [12] Kenneth G. Elzinga & David E. Mills, The Economics of Resale Price Maintenance, in 3 ISSUESin Competition Law and Policy 1841, 1848 (Wayne D. Collins ed., 2008) (“To secure entry, anew entrant may seek to gain retail distribution by offering independent retailers protections againstdiscounting, in the hope that margin protection will induce retailers to market and promote the newproduct”).
  • [13] Thomas R. Overstreet Jr., Resale Price Maintenance: Economic Theories andEmpirical Evidence (1983).
  • [14] Pauline M. Ippolito, Resale Price Maintenance: Empirical Evidence from Litigation, 34 J.L. &Econ. 263 (1991).
  • [15] See Frank H. Easterbrook, The Limits of Antitrust, 63 Tex L. Rev. 1 (1984).
< Prev   CONTENTS   Source   Next >