ATOMISTIC ANTITRUST.

AuthorFeldman, Robin C.

TABLE OF CONTENTS Introduction 1872 I. Antitrust's Focus on the Trees 1876 A. Standards of Proof 1876 B. Mergers and Future Harm 1879 C. Petitioning Agencies and Courts 1884 D. Market or Markets? 1889 II. The Forest Antitrust Misses 1892 A. Probabilistic Antitrust Harm 1893 1. Mergers 1893 2. Noncompete Agreements and Exclusive Dealing 1901 B. Synergistic Antitrust Violations 1906 1. Multiple Actions 1907 2. Multiple Markets 1914 III. Clearing a Path 1922 A. Too Big to Buy 1923 B. Banning Noncompetes 1928 C. Consider Markets and Conduct as a Whole 1930 D. Remedies for Anticompetitive Conduct 1934 Conclusion 1936 INTRODUCTION

A fundamental premise of modern antitrust law is that monopoly itself is not illegal. Monopolies can exist--and charge monopoly prices--without fear of antitrust liability. Big is not bad if a company's strength is earned through legitimate competition. (1) Only specific anticompetitive acts of monopolization or certain agreements with rivals are forbidden by the antitrust laws. Antitrust enforcement therefore focuses its primary attention on particular acts claimed to be unlawful, albeit in the context of market power or other competitive harm. And consistent with the normal rules of evidence and procedure, showing an antitrust violation generally requires proving by a preponderance of the evidence that the particular challenged act harmed competition. (2) And it has become even harder to meet that burden in recent years as courts imposed new, antitrust-specific barriers and burdens of proof in an effort to reduce the risk of false positives. Antitrust, in short, is atomistic: deliberately focused on trees, not forests.

That focus is misplaced. Companies and markets don't focus on one particular act to the exclusion of all else. Business strategy emphasizes holistic, integrated planning. (3) And market outcomes aren't determined by a single act, but by the result of multiple acts by multiple parties in the overall context of the structure and characteristics of the market.

The atomistic nature of modern antitrust law causes it to miss two important classes of potential competitive harms. First, the focus on individual acts, coupled with the preponderance of the evidence standard for proving a violation, means that antitrust can't effectively deal with what we might call probabilistic competitive harm: multiple acts, any one of which might or might not harm competition. A monopolist might, for instance, buy a startup that could potentially turn into a competitor. At the time of the merger, there is no way to know for sure whether the startup would have matured into a competitive threat. And the probability that any particular startup would have displaced the monopolist might be small--say 10 percent. Atomistic antitrust says there is no violation here because that particular merger is not more likely than not to restrict competition. (4) But statistically, a monopolist that buys one hundred such startups has almost certainly restrained competition, even though antitrust treats no one purchase as illegal and even though we don't know which startups would have succeeded.

Second, atomistic antitrust tends to miss synergistic competitive harm: acts that are lawful when taken individually but combine to produce anticompetitive effects. A pharmaceutical company, for instance, might take many acts to delay generic competition--from filing lawsuits, to acquiring follow-on patents, to filing citizen petitions at the FDA, to denying generic companies access to samples of the drug, to repeatedly changing the formulation of the drug. Some of these acts might themselves be unlawful, but often the law views them, in isolation, as permitted, even constitutionally protected, activity. (5) But in the context of pharmaceutical regulation, they work together to prevent competition that would otherwise have occurred, not because of a genuine effort to persuade the government or the courts, but because of the combined effect of multiple obstacles to generic competition. (6)

Synergistic effects also can arise from the interactions of multiple markets. Patent aggregators that amass vast portfolios can force rational companies to pay for a license simply by the threat of the combined power of the portfolio. (7) The aggregator may not hold sufficient patents in a traditional product market to constitute a monopoly. Rather, the aggregator's power comes from the synergistic effects of lower levels of power in multiple intellectual property (IP) markets. An atomistic lens cannot capture these types of effects.

Unfortunately, modern antitrust law has strayed too far down the atomistic pathway. Courts and agencies too often take a narrow, transaction-specific focus to challenged conduct. Instead of asking "is the overall behavior of this company reducing competition in the market," they focus on a particular merger or challenged monopolistic practice in isolation. They also create rules of thumb that allow them to turn away antitrust cases early on. Those rules of thumb are easier to administer, and they provide a sense of comfort for companies. The problem is not that everything involved in a particular rule of thumb is wrong or that the notion of developing a proxy itself is wrong; the danger lies in allowing the proxy to take on a life of its own, disembodied from the underlying concepts. (8) Modern antitrust law has reached that dangerous domain.

Courts and agencies need to move beyond atomistic antitrust and take a more holistic look at the circumstances and effects of an overall pattern of conduct. Doing so doesn't require new legislation. Existing antitrust doctrine permits a broader focus, and indeed that was part of the original purpose of antitrust law. Section 7 of the Clayton Act, for instance, allows the government to stop monopoly "in its incipiency" (9) by blocking mergers whose effect "may be substantially to lessen competition, or to tend to create a monopoly." (10) That is rather less than a requirement that a plaintiff prove a merger will create or preserve a monopoly. Similarly, the law on the books permits the aggregation of conduct by monopolists accused of an overall scheme to monopolize. (11) Courts just haven't been enforcing that law. And because antitrust rules are almost entirely created by courts, courts have the power to change those rules to better protect competition.

Our goal in this Article is to set out a framework for integrated antitrust in which individual actions can be understood not just on their own but also as part of a comprehensive whole. Only by doing so can the legal system both return antitrust to its roots and bring antitrust into the modern context of the business decisions that courts must analyze today. It is a rare moment in legal theory when one can accomplish both at the same time.

In Part I, we show how antitrust law over the last fifty years has become increasingly atomistic, trading a broad focus on the structure of markets for a narrow approach anchored in atomistic ends. In Part II, we argue that this atomistic focus misses important modern harms to competition, including those related to large tech companies buying startups, employers imposing noncompete clauses, pharmaceutical company behaviors blocking generics, and patent aggregators asserting large portfolios. Finally, in Part III, we argue for a return to an integrated analysis, one that focuses on competitive effects as a whole.

We also set out a framework for specific improvements--ranging from more comprehensive reforms requiring congressional action to more narrow reforms appropriate for judicial implementation--that could help avoid the shortsightedness of atomistic antitrust. Among other things, we suggest creating a presumption of anticompetitive harm from mergers by monopolists, allowing antitrust law to punish a pattern of conduct that harms competition even if no one act in that pattern is itself provably anticompetitive, and encouraging courts that do find an antitrust violation to impose remedies sufficient to undo the competitive harm rather than just limiting themselves to stopping an ongoing violation. More important than the particular solutions, however, is replacing today's atomistic antitrust with a broader focus.

  1. ANTITRUST'S FOCUS ON THE TREES

    Antitrust law was conceived broadly, as the "Magna Carta of free enterprise." (12) When the first antitrust law, the Sherman Antitrust Act, was passed in 1890, it condemned two basic types of conduct--monopolization (an effort by a dominant firm to acquire or maintain control of a market) and cartels (agreements among competitors to fix prices or otherwise restrict competition). (13) The statutes are written in sweeping fashion to cover a wide variety of conduct that threatens competition.

    Despite the breadth of the statutes, antitrust law has become increasingly atomistic. Over the last fifty years in particular, a number of procedural rules and substantive doctrines have pushed antitrust into narrow corners, leaving it ill-equipped to manage a comprehensive analysis of competitive effects as a whole. In this Part, we discuss a range of these doctrines.

    1. Standards of Proof

      Antitrust law requires plaintiffs to prove causation and injury with more detail and particularity than other bodies of law. (14) That is particularly important because integral aspects of antitrust law rely much more on predictions of future harm than other areas of law. (15) Most law is backward-looking, asking whether a defendant breached a contract, committed a tort, infringed a patent, and the like. At most, remedies may require some effort to predict the future. (16) But important parts of antitrust law--in many ways the most important parts--are designed not to identify past illegal acts but to prevent future ones. Assessing a merger, a regulatory filing, a change in prices, or an exclusive dealing arrangement requires courts to compare what would...

To continue reading

Request your trial

VLEX uses login cookies to provide you with a better browsing experience. If you click on 'Accept' or continue browsing this site we consider that you accept our cookie policy. ACCEPT