Daniel A. Crane, Mixed Bundling, Profit Sacrifice, and Consumer Welfare

• Publication year: 2006

MIXED BUNDLING, PROFIT SACRIFICE, AND CONSUMER WELFARE

Daniel A. Crane^*

INTRODUCTION ...............................................................................................425

I. A TAXONOMY OF EXPLANATIONS FOR MIXED BUNDLING ..................428

A. Procompetitive and Competition-Neutral Explanations .............430

1. Cost Efficiencies ....................................................................430

2. Instilling Customer Loyalty ...................................................433

3. Eliminating Double Marginalization ....................................434

4. Price Discrimination .............................................................436

5. Playing Behavioral Games ....................................................438

6. Exploiting Fragmented Customer Decision-Making .............440

7. Responding to Monopsonist or Oligopsonist Demand for

Discounts ...............................................................................441

B. Exclusionary Explanations ..........................................................443

1. Forcing Competitor to Price Unprofitably ............................443

2. Creating Entry Barriers by Raising Rivals' Costs ................446

II. PROFIT SACRIFICE THEORY AND MIXED BUNDLING ...........................447

A. Theories of Exclusion with Sacrifice ...........................................449

1. Bundling by a Monopolist .....................................................450

2. Bundling by a Firm that Faces Competition in All

Covered Markets ...................................................................455

B. Theories of Exclusion Without Sacrifice .....................................459

III. CONSUMER WELFARE AND ADMINISTRABILITY ..................................463

A. Perils of Administration ..............................................................465

1. The Law of Unintended Consequences ..................................465

2. The Shortcomings of an Efficiency Justification Defense ......469

B. Categorical Rules for Promoting Price Competition ..................473

1. Ordinary Predation Rules (May) Still Apply .........................473

2. Above-Cost Effective Prices Per Se Legal .............................474

3. Immunity for Above-Cost Bundles Where Rivals Compete in All Covered Markets ..........................................................480

4. Pro Rata Allocation of Discounts Straddling Multiple

Competitive and Monopoly Markets .....................................481

5. Foreclosure and Coercion Required .....................................482

6. Actual Economic Harm to Consumers ..................................484

CONCLUSION ...................................................................................................485

ABSTRACT

Mixed bundling-the conditioning of discounts in one market on the purchase of products or services in another market-is a pervasive business practice that is attracting attention because of its potentially exclusionary effects on competitors. The practice has been analogized to anticompetitive tying and, as with tying a generation ago, the reasons for its existence have been misunderstood. In most cases, mixed bundling benefits consumers. Firms adopt mixed bundling pricing strategies for many procompetitive or competitively neutral or indeterminate reasons, including achieving efficiencies, responding to demand by powerful customers, price-discriminating, and playing behavioral games with customer psychology. Courts and the antitrust agencies are struggling to formulate legal tests capable of addressing mixed bundling as a potential monopolization offense. The prominent sacrifice test-under which an act is considered unlawfully exclusionary if it makes no rational sense absent the exclusion of competitors-cannot be meaningfully applied to most mixed bundling schemes in a litigation context because there are too many variables that determine the "rationality" calculus. But nor do theories of exclusionary bundling without profit sacrifice provide a compelling reason to be worried about the competitive effects of mixed bundling in most cases. Given the generally procompetitive effects of mixed bundling and the propensity of plaintiffs strategically to misuse monopolization law to stymie vigorous price competition, courts should be reluctant to create new prohibitory rules and should, at most, use existing predatory pricing principles to assess the practice.

INTRODUCTION

Mixed bundling describes a pricing strategy where a seller offers a discount if separate items are purchased together as a package.¹If you have ever purchased skis together with bindings, a value meal in a fast food restaurant, or a vacation package including airfare and lodging, you have probably encountered mixed bundling. Mixed bundling pricing strategies are pervasive in the U.S. economy. They can take a number of different forms, such as loyalty rebates, graduated discount programs, incentive targets, and "buy one, get one for half price" schemes.

Firms engage in mixed bundling strategies for many reasons, but one possible motivation has recently captured the attention of courts, economists, antitrust regulators, and plaintiffs' lawyers: Mixed bundling strategies may be employed by a dominant firm to foreclose competition by a single-product competitor that is unable to match the multiproduct or multimarket discounts and therefore loses sales in the sole market in which it competes with the dominant firm. The Third Circuit's en banc decision in LePage's Inc. v. 3M involves allegations of such exclusion²and has drawn a formidable amount of attention, much of it critical.³Other important cases involving mixed bundling allegations have recently been decided or lurk on the horizon.⁴Neither the practice of mixed bundling, nor the need for clear antitrust standards governing this practice, is likely to disappear any time soon.

The antitrust community's focus on mixed bundling arrives at an important moment in the wider debate over the proper way of defining the monopolization offense under section 2 of the Sherman Act. Section 2 prohibits monopolizing, attempting to monopolize, and conspiring to monopolize, but never defines these terms.⁵Since Learned Hand's iconic

Alcoa opinion,⁶courts have recognized that the monopolization offense requires an affirmative bad act, an exclusionary act.⁷Of course, most competitive acts "exclude" competitors in some sense,⁸so legal differentiation between permissible and impermissible exclusion is required. The theory currently preferred by the Justice Department and a number of courts and scholars makes unlawfully exclusionary any act by a dominant firm that would make no rational sense unless it assumed the exclusion of a competitor and the obtainment of monopoly profits.⁹This so-called "profit sacrifice" theory of monopolization has come under attack by other scholars who argue that it is both overinclusive and underinclusive when measured against the consumer welfare goals of the Sherman Act.¹⁰But if the sacrifice test fails adequately to account for mixed bundling, nor do non-sacrifice models¹¹provide a compelling basis to create prohibitory rules governing bundled discounts. Such models rely on rarely observed conditions and show that mixed bundling can be used to increase both the seller's profits and consumer welfare without necessarily excluding competitors.¹²

Counterproposals to the sacrifice theory abound,¹³but prove similarly unable to provide a reliable way to adjudge the legality of mixed bundling. Mixed bundling thus provides an important juridical lesson about the difficulty of encapsulating the monopolization offense in a single verbal formulation. In approaching mixed bundling-and probably other forms of competitive conduct as well-courts would do better to stop searching for a metaphysical statement of principle and instead formulate categorical rules likely to advance consumer welfare. As they do, antitrust rulemakers should keep in mind that liability rules setting limits on price discounting must be predictable and easily administrable or they will quickly provide an opportunity for strategic misuse of antitrust law by rent-seeking competitors.

Bundled discounting is similar in many ways to the "tying" of two separate products-a much theorized and litigated competitive practice. The history of tying law reveals some pitfalls that courts should avoid as they formulate the law of mixed bundling. Courts in the early tying cases failed to appreciate the many procompetitive reasons that firms tie products and therefore imposed draconian liability rules on a practice that is usually benign.¹⁴More broadly, the history of common law development under the Sherman Act suggests that the beginning point in any effort to frame liability rules for a category of competitive conduct should be an inquiry into the reasons firms engage in that conduct. Much bad law has been created and lingered for decades simply because courts failed to appreciate the many reasons that firms engage in particular practices.¹⁵This Article seeks to provide a better understanding of the reasons that mixed bundling occurs and the limitations of litigation in controlling the few occasions when it causes competitive concerns.

Part I presents a taxonomy of procompetitive, competitively neutral, and anticompetitive explanations for the existence and ubiquity of mixed bundling. It shows that mixed bundling is pervasive, generally beneficial to consumers, and usually not exclusionary of competitors even when its welfare effects are ambiguous. Part II demonstrates that the sacrifice test cannot do justice to mixed bundling, and calls into question the usefulness of generalized verbal formulations of the monopolization offense. Part II also shows that alternative non-sacrifice theories of exclusion do not provide any compelling reason to be concerned about the competitive effects of bundled discounts. Finally, Part...