One of my summer writing projects is a response to Einer Elhauge’s recent, highly acclaimed article, Tying, Bundled Discounts, and the Death of the Single Monopoly Profit Theory. In the article, which appeared in the December 2009 Harvard Law Review, Elhauge defends current tying doctrine, which declares tie-ins to be per se illegal when the defendant has market power in the tying product market and the tie-in affects a “not insubstantial” volume of commerce in the tied product market.
Efficiency-minded scholars from both the Chicago and Harvard Schools of antitrust analysis have criticized the prevailing liability rule for failing to require that a challenged tie-in foreclose a substantial percentage of marketing opportunities in the tied product market. Absent such foreclosure, those scholars have argued, tying cannot create anticompetitive harm and should not constitute a per se antitrust violation.
Elhauge insists that these scholars are wrong. Even in the absence of substantial tied market foreclosure, he observes, a tie-in can facilitate price discrimination and can extract additional surplus from consumers. He contends that these price discrimination and surplus extraction effects are anticompetitive effects that are properly addressed by antitrust.
No one seriously doubts Elhauge’s initial observation about tying’s potential non-foreclosure effects. Indeed, even Chicago School scholars have long recognized that a monopolist may use a tie-in to meter demand for its monopoly tying product, charging higher effective prices to consumers with high reservation prices. A printer monopolist, for example, may expand its profits by lowering its printer price from profit-maximizing levels but requiring customers to purchase its ink cartridges at supracompetitive prices. High-value customers, who use lots of ink, will end up paying higher effective prices than will low-value customers. Thus, a “requirements tie” of this type may permit a monopolist to price discriminate and to redistribute surplus from high-value consumers to itself.
Elhauge’s second claim — that price discrimination and surplus extraction are anticompetitive effects that tying doctrine should police — is another matter entirely. In support of that claim, he asserts both positive and normative arguments. As a normative matter, he contends that the sort of price discrimination occasioned by tie-ins is likely to reduce both consumer and total surplus and should thus be condemned. As a positive matter, he argues that Supreme Court precedent does, in fact, deem tying’s price discrimination/surplus extraction effects to be anticompetitive.
In future posts, I’ll address Elhauge’s normative argument. For now, let me take a stab at his positive claim.
Elhauge’s positive argument first infers from the structure of tying doctrine that the Supreme Court views price discrimination and surplus extraction as anticompetitive effects of tying. The liability rule for tying doesn’t require that a plaintiff establish substantial foreclosure of the tied product market. Ergo, Elhauge reasons, the Court must deem tying’s nonforeclosure effects (price discrimination and surplus extraction, which he collectively refers to as “power effects”) to be anticompetitive.
This assumes, of course, that the Supreme Court has succeeded in crafting a liability test that is narrowly tailored to prevent only those effects it deems anticompetitive. In reality, there is frequently a “disconnect” between the liability rule the Court adopts and the policy concerns underlying that rule. This is especially true in antitrust, where developments in economic theory lead the evolution of the law, and the legal doctrine changes only as new cases, presenting narrow legal issues for resolution, come before the Supreme Court through the common law process. The maximally prohibitory per se rule against vertical minimum resale price maintenance, for example, persisted long after the Court realized that a number of the effects that initially motivated the rule (e.g., interference with dealer freedom, restraints on alienation of chattels) are not, in fact, anticompetitive. It would have been silly to argue in 2006, the year before the per se rule was overruled, that the persistence of the rule proved that the Supreme Court deemed some or all of those effects to be anticompetitive. Similarly, one should not infer much about the Supreme Court’s understanding of tying policy from its adherence, on stare decisis grounds, to a liability rule that predates the modern understanding of tying’s various effects.
Perhaps recognizing the danger in attempting to divine the Supreme Court’s policy conclusions from the structure of prevailing antitrust liability rules, Elhauge seeks to buttress his legal argument by pointing to explicit statements in Supreme Court opinions that, he says, demonstrate that the Court deems tying’s price discrimination effects to be anticompetitive. The most direct such statement first appeared in a dissenting opinion but was later quoted in majority opinions of the Supreme Court. In Fortner I, dissenting Justice White, joined by Justice Harlan, purported to state “the rationale on which the illegality of tying arrangements is based.” The dissenters first pointed to tying’s ability “to work significant restraints on competition in the tied product” by foreclosing competitors from the tied product market and by raising barriers to entry in that market. They then observed that tying may have additional effects. Elhauge mentions two of the effects noted: tying arrangements may be used “as a counting device to effect price discrimination” and “to force a full line of products on the consumer so as to extract more easily from him a monopoly return on one unique product in the line.” He infers from this mention of tying’s potential price discrimination and surplus extraction effects that the dissenting justices deemed those effects to be anticompetitive.
A more complete review of the White-Harlan dissent suggests such an inference is unwarranted. In quoting from the Fortner I dissent, Elhauge excludes one of the effects the dissenting justices recognized. Replacing Elhauge’s ellipsis with the actual text, the White-Harlan dissent reads as follows:
In addition to these anticompetitive effects in the tied product [i.e., those related to foreclosure from, and the creation of entry barriers into, the tied product market], tying arrangements may be used to evade price control in the tying product through clandestine transfer of the profit of the tied product; they may be used as a counting device to effect price discrimination; and they may be used to force a full line of products on the customer so as to extract more easily from him a monopoly return on one unique product in the line.
This is a significant omission. If, as Elhauge suggests, the dissenting justices had meant that the additional effects they noted are anticompetitive effects, then they would have been implying that evasion of price controls is anticompetitive. But quite often — e.g., any time price controls are implemented for reasons other than to constrain a seller’s exercise of market power — evasion of price controls is actually output-enhancing and thus pro-, not anti-, competitive. The inclusion of a frequently procompetitive effect on the list of non-foreclosure effects of tying suggests that the dissenting justices did not mean to imply that the listed effects should be deemed anticompetitive.
This passage from Fortner I is of little persuasive value for the obvious reason that it occurred in a dissent. But Elhauge also cites two majority Supreme Court opinions — Jefferson Parish and Independent Ink — in support of his claim that the Court deems tying’s price discrimination effects to be anticompetitive. In Jefferson Parish, the Court favorably quoted the aforementioned passage from Justice White’s Fortner I dissent and cited that dissent in observing that tying “can increase the social costs of market power by facilitating price discrimination, thereby increasing monopoly profits over what they would be absent the tie.” Elhauge thus concludes that the Jefferson Parish Court deemed these various non-foreclosure effects of tying to be anticompetitive effects.
But Elhauge ignores a passage in which the Jefferson Parish Court more directly addressed the preconditions to anticompetitive harm from tying and clarified unequivocally that tied market foreclosure is, in fact, such a prerequisite. After cataloguing the various effects of tying arrangements (including, as Elhauge emphasizes, tying’s potential to facilitate price discrimination and thereby enhance the tying monopolist’s profits), the Court turned to the conditions under which tying may create anticompetitive consequences that justify per se condemnation, and it emphasized two situations in which such anticompetitive harm would not occur:
[A]pplication of the per se rule focuses on the probability of anticompetitive consequences. If only a single purchaser were ‘forced’ with respect to the purchase of a tied item, the resultant impact on competition would not be sufficient to warrant the concern of antitrust law. It is for this reason that we have refused to condemn tying arrangements unless a substantial volume of commerce is foreclosed thereby. Similarly, when a purchaser is “forced” to buy a product he would not have otherwise bought even from another seller in the tied product market, there can be no adverse impact on competition because no portion of the market which would otherwise have been available to other sellers has been foreclosed. Once this threshold is surmounted, per se prohibition is appropriate if anticompetitive forcing is likely.
This passage makes clear that tied market foreclosure is necessary for anticompetitive consequences to result from tying. If, as Elhauge contends, price discrimination or the extraction of additional consumer surplus constituted an anticompetitive harm that could justify prohibiting tying under the antitrust laws, then both scenarios in the quoted passage would give rise to antitrust liability. It would be entirely possible for a monopolist to price discriminate against, and extract additional surplus from, a single high-value consumer by requiring that consumer to purchase a tied product in order to obtain the monopolist’s tying product. Similarly, a monopolist could price discriminate or extract additional consumer surplus by imposing a tie-in that forced a purchaser to buy a tied product he would not have purchased from another seller in the tied product market. Neither scenario, however, would likely generate significant foreclosure in the tied market. In insisting that a single customer tie-in would not impact competition sufficiently “to warrant the concern of antitrust law” and that a tie-in involving a tied product the customer would not otherwise have bought would have “no adverse impact on competition because no portion of the market which would otherwise have been available to other sellers has been foreclosed,” the Court made clear that price discrimination and surplus extraction are not anticompetitive effects of tying and that market foreclosure is a prerequisite to anticompetitive harm.
The Supreme Court’s most recent tying decision removed any doubt about that proposition. In Independent Ink, the Court ruled that a tying defendant’s possession of a patent on its tying product is not, in itself, sufficient to establish that the defendant possessed market power in the tying product market. The defendant, a printer manufacturer, had required purchasers of its printers to use its ink exclusively. The Court of Appeals for the Federal Circuit had concluded that the defendant’s possession of a patent on its printhead technology established its market power in the printer market. The Supreme Court reversed on that point and returned the case to the district court so that the plaintiff could have an opportunity to define the relevant tying product market and prove the defendant’s possession of power within it. Elhauge contends that the Court’s remand instruction “confirm[s] that power effects [i.e., price discrimination and surplus extraction] suffice” to establish anticompetitive harm. Had the Court believed that tied market foreclosure is a necessary prerequisite to anticompetitive harm, Elhauge asserts, its remand instruction “would have required evidence of a substantial tied foreclosure share, which would have been implausible because the ink used for one specialized sort of printer is hardly likely to be a big share of all ink.”
Elhauge’s reading of Independent Ink is unpersuasive for two reasons. As an initial matter, the Supreme Court’s focused remand instruction was dictated by the narrow legal issue before the Court. The Independent Ink Court granted certiorari on a precise question — whether a tying defendant’s possession of a patent on the tying product should give rise to a presumption of market power in the tying product market — and the briefing and arguments of the parties therefore focused exclusively on that question. It would have been improvident for the Court to address the broader issue of whether substantial tied market foreclosure should be required for tying liability when the petitioner had not sought review of the question and neither the parties nor amici had briefed the issue. Accordingly, the Court addressed only the narrow issue at hand and crafted its remand instruction to reflect only its resolution of that discrete issue.
Moreover, while the Independent Ink Court was not presented with the question of whether substantial tied market foreclosure should be a prerequisite to tying liability, it did consider — and rejected — the argument that requirements ties resulting in price discrimination and extraction of consumer surplus should be condemned as anticompetitive. In seeking to sustain the judgment in its favor, the plaintiff-respondent presented the Court with a narrower alternative to its requested holding that possession of a patent creates a presumption of market power. That narrower alternative would have created a presumption of tying market power when a defendant with a patent on its tying product imposes a “requirements tie” on purchasers of that product, requiring them also to purchase unpatented complements from the defendant. As professors Barry Nalebuff, Ian Ayres, and Lawrence Sullivan explained in an amicus brief advocating this narrower holding, the presumption of tying market power in cases involving patented tying products and requirements ties would enable antitrust to police the use of tie-ins to price discriminate and extract additional consumer surplus by metering consumer demand for the tying product. Professors Nalebuff, Ayres, and Sullivan argued that such price discrimination and surplus extraction are anticompetitive effects that are properly addressed by antitrust. The Supreme Court acknowledged that metering tie-ins may result in price discrimination; it even referenced the Jefferson Parish footnote asserting that point. Yet, the Court rejected the narrower holding advocated by professors Nalebuff, et al. because it concluded that price discrimination “occurs in fully competitive markets” and that “[m]any tying arrangements, even those involving patents and requirements ties, are fully consistent with a free, competitive market.” The Court thus made clear that, while it understands that price discrimination and additional surplus extraction are possible effects of tying, they are not appropriately deemed anticompetitive effects that justify condemnation under the antitrust laws.
Of course, the more interesting issue is the normative question of whether tying-induced price discrimination and surplus extraction should be deemed anticompetitive effects. Elhauge says yes. I’ll address that issue as the summer progresses.