In 2012, a prominent scholar lamented the diminished significance of antitrust in the United States.¹ Although there was once a flourishing antitrust movement, he argued, the subject appeared to attract little interest from lawmakers, academics, and the public.² Political candidates rarely mentioned competition issues, opinion polls reflected indifference toward economic concentration, and the enforcement agencies seemed to operate with a narrow view of antitrust's goals.³

Things have changed. In the past several years, antitrust has resurfaced as a topic of both popular and political concern.⁴ The White House has issued an executive order outlining a "whole-of-government" approach to competition policy;⁵ advocates of reform have been appointed to lead the Federal Trade Commission (FTC) and the Department of Justice's (DOJ's) Antitrust Division;⁶ and Congress has considered a suite of proposals to overhaul various aspects of antitrust doctrine.⁷

In the words of one commentator, antitrust now "stands at its most fluid and negotiable moment in a generation."⁸ The subject has not had such political salience, another contends, since 1912.⁹ Interest in reform has been wide-ranging: "[e]verything is up for grabs, and nothing is free of scrutiny."¹⁰

One of the driving forces behind this trend has been the rise of a handful of large technology firms: Facebook (now Meta Platforms), Google, Amazon, and Apple. In 2020, a House subcommittee released a detailed report (the HJC Report) concluding that the four companies had obtained and solidified dominant positions through anticompetitive conduct.¹¹ These "Big Tech" firms have also faced antitrust lawsuits from regulators and private plaintiffs, both in the United States and abroad.¹²

This report provides an overview of antitrust issues involving the four Big Tech firms and related proposals for legislative reform. It is divided into four parts. First, the report provides an introduction to basic antitrust principles. Second, it reviews selected antitrust allegations against the Big Tech companies. Third, it discusses conceptual issues with proposals to reform the competition laws governing Big Tech. Fourth, the report analyzes the substance of specific categories of reform proposals.

Antitrust Law: The Basics

The antitrust laws aim to protect economic competition by prohibiting unreasonable restraints of trade,¹³ exclusionary conduct by dominant firms,¹⁴ and mergers and acquisitions that may "substantially" lessen competition or "tend to create a monopoly."¹⁵

The following subsections provide a high-level overview of antitrust doctrine to lay the groundwork for later discussions of competition issues in tech markets and proposals for legal reform.

Restraints of Trade

Section 1 of the Sherman Act prohibits "every" contract or conspiracy "in restraint of trade."¹⁶ Despite this categorical language, the Supreme Court has interpreted Section 1 to bar only unreasonable restraints of trade that harm competition.¹⁷

Applying this general standard, the Court has identified some types of agreements that are so likely to be anticompetitive that they are deemed per se illegal, meaning courts need not inquire into their effects in individual cases.¹⁸ Restraints in this category include agreements among competitors ("horizontal" agreements) to fix prices,¹⁹ divide markets,²⁰ and restrain output.²¹

While some types of agreements are per se illegal under Section 1, most restraints are evaluated using a standard called the "rule of reason."²² Under the rule of reason, courts conduct fact-specific assessments of a defendant's market power and the details of a challenged agreement to determine a restraint's competitive effects.²³

This inquiry typically proceeds via a three-step burden-shifting framework. In that framework, the plaintiff has the initial burden to prove that the challenged restraint has a substantial anticompetitive effect.²⁴ Plaintiffs can make this showing directly or indirectly. Direct evidence of anticompetitive harm involves "proof of actual detrimental effects on competition," such as reduced output, increased prices, or decreased quality.²⁵ The indirect route involves proof of market power,²⁶ plus "some evidence that the challenged restraint harms competition"—for example, evidence that the restraint is the type of restriction that tends to produce anticompetitive outcomes.²⁷

If the plaintiff makes a prima facie case of anticompetitive harm, the burden shifts to the defendant to show a procompetitive justification for the challenged restraint.²⁸ For example, a defendant might argue that the restraint increases output, creates operational efficiencies, makes a new product available, enhances product quality, or broadens consumer choice.²⁹

If the defendant makes this showing, the burden shifts back to the plaintiff to demonstrate that the relevant procompetitive benefits could be reasonably achieved through less anticompetitive means.³⁰ Some courts have also added a fourth step in which they balance a restraint's anticompetitive and procompetitive effects.³¹

Although most agreements are evaluated under this framework, courts have also recognized a third standard that lies between the full rule of reason and per se illegality. This intermediate approach—often called "quick look" review—has been applied to agreements that are not per se unlawful but nevertheless exhibit characteristics that make their likely anticompetitive effects clear.³² Different courts have described quick-look analysis in different ways.³³ The basic idea, however, is that the plaintiff in a quick-look case can discharge its initial burden without the type of detailed evidence of competitive harm required under the full rule of reason.³⁴ Defendants in quick-look cases, meanwhile, have the opportunity to offer procompetitive justifications for their conduct, distinguishing quick-look review from per se analysis.³⁵

Monopolization

While Section 1 of the Sherman Act governs agreements between firms, Section 2 prohibits dominant companies from engaging in concerted or unilateral exclusionary conduct by making it unlawful to monopolize, attempt to monopolize, or conspire to monopolize.³⁶

The monopolization offense has two elements:

• 1. the possession of monopoly power; and
• 2. "the willful acquisition or maintenance of that power as distinguished from growth or development as a consequence of a superior product, business acumen, or historical accident."³⁷

The second element is often referred to as "exclusionary" or "anticompetitive" conduct.³⁸

Monopoly Power

The Supreme Court has explained that a firm possesses monopoly power if it has the ability to "control prices or exclude competition."³⁹ Although that standard is similar to many descriptions of market power,⁴⁰ the Court has clarified that monopoly power under Section 2 of the Sherman Act requires "something greater" than market power under Section 1.⁴¹ Courts have thus concluded that monopoly power entails a large degree of market power.⁴²

Some courts have held that monopoly power can be established through direct evidence of supra-competitive prices and restricted output.⁴³ However, this type of direct proof is rarely available.⁴⁴ As a result, courts typically evaluate allegations of monopoly power by examining structural factors like a defendant's market share and any entry barriers in the relevant market.⁴⁵

These inquiries require a plaintiff to define the scope of the relevant market—an exercise that turns on the range of items that are reasonable substitutes for one another.⁴⁶

There are two general approaches to market definition. One approach—the hypothetical monopolist test (HMT)—attempts to identify the smallest grouping of products over which a single seller could exercise significant market power.⁴⁷ Under a version of the HMT known as the SSNIP test, this inquiry starts with the product at issue in a given case and asks whether a hypothetical monopolist selling that product could profitably increase its price by a significant amount (typically 5%-10%) for a non-transitory period of time (typically one year or more).⁴⁸ If the answer is yes, then the product represents a relevant antitrust market. However, if consumer substitution would render such a price increase unprofitable, the SSNIP test prescribes that the market must be expanded to include substitute products. This process continues until the point at which a "small but significant non-transitory increase in price" for one of the products would be profitable.⁴⁹

A second approach to market definition involves an evaluation of qualitative similarities and differences between products and services. This methodology is derived from the Supreme Court's 1962 decision in Brown Shoe Co. v. United States, which identified a series of "practical indicia" that may be relevant to an evaluation of a market's boundaries.⁵⁰ These qualitative factors include industry or public recognition of separate markets, a product's peculiar characteristics and uses, unique production facilities, distinct customers, distinct prices, sensitivity to price changes, and specialized vendors.⁵¹

The HMT and Brown Shoe's qualitative inquiry both attempt to determine the range of substitutes that constrain a firm's exercise of market power,⁵² and judicial decisions often employ both approaches in defining markets.⁵³

As mentioned, once a market has been defined, courts typically assess claims of monopoly power by evaluating a defendant's market share and other structural factors like entry barriers.⁵⁴ When entry barriers are present, a market share in excess of 70% can establish a prima facie case of monopoly power.⁵⁵ Courts rarely find monopoly power, by contrast, when a firm's market share is less than 50%.⁵⁶ Shares between 50% and 70% present "the greatest uncertainty,"⁵⁷ with some courts deeming shares in that range to be insufficient absent additional evidence.⁵⁸

To establish monopoly power, plaintiffs also typically must show that a defendant's dominant position is likely to be durable—for example, with evidence of significant barriers to entry.⁵⁹ Entry barriers may include legal and regulatory requirements, control of an essential resource, entrenched buyer preferences, and economies of scale.⁶⁰ In some digital markets, entry barriers may also emerge from network effects (which cause a product's utility to increase as it gains more users) and significant switching costs (high costs that users of a product would face in switching to a substitute).⁶¹

Exclusionary Conduct

As noted, the second element of a monopolization claim is exclusionary conduct. The Supreme Court has described this element as involving "the willful acquisition or maintenance of [monopoly] power as distinguished from growth or development as a consequence of a superior product, business acumen, or historical accident."⁶²

As a general standard, many have found that description unhelpful. Firms often "willfully" try to obtain monopoly status by developing superior products and by deploying business acumen.⁶³ Moreover, in offering the above formulation, the Supreme Court did not define "business acumen," leaving little guidance as to when practices like aggressive price cutting, bundling separate products, or refusing to share property with rivals represent savvy strategy as opposed to unlawful exclusion.⁶⁴

While academics have made several attempts to develop an alternative general standard, courts have not decisively embraced any of them.⁶⁵ Instead, the doctrine contains a variety of tests that govern specific categories of conduct, along with a burden-shifting framework that is similar to the usual rule-of-reason inquiry in Section 1 cases.

The following sections review efforts to develop a unified theory of monopolization and several of the conduct-specific tests that courts have adopted in place of such a theory.

Predatory Pricing

Some monopolization cases involve allegations that a defendant aggressively cut prices in an attempt to exclude rivals from the market—a practice commonly known as predatory pricing. Predation has been described as "a play in two acts."⁶⁶ In the first stage of a predation scheme, a firm charges unsustainably low prices to drive rivals from the market or deter entry.⁶⁷ In the second, the firm attempts to recoup the losses incurred in the first stage by raising prices to monopoly levels.⁶⁸

Predatory pricing has played a notable role in antitrust history and was a part of the federal government's landmark monopolization case against the Standard Oil Company, which was broken up in 1911.⁶⁹ The practice was also a common target of antitrust enforcement through the 1960s,⁷⁰ when some courts evaluated predation claims by focusing on whether the defendant intended to harm rivals.⁷¹

Today, matters are different. Beginning in the 1950s, academics affiliated with what came to be known as the Chicago School of antitrust analysis mounted a critique of prevailing theories of predatory pricing.⁷² They claimed, among other things, that predation is typically an irrational strategy, because monopoly prices charged during the recoupment period will often invite entry, which will in turn drive prices down to competitive levels.⁷³ Chicago School scholars also contended that monopolists will usually suffer greater losses from a price war than their competitors, because large firms tend to make more sales than smaller ones.⁷⁴ Other academics from what is often called the modern Harvard School later offered arguments for a less interventionist posture that were grounded in institutional concerns about the ability of courts to distinguish predatory pricing from vigorous price competition.⁷⁵

These criticisms proved highly influential.⁷⁶ In the 1970s and 1980s, many lower courts took a more restrictive approach to predation claims, often requiring plaintiffs to show that the defendant's prices fell below its costs rather than inquiring into the defendant's intent.⁷⁷ The Supreme Court ultimately ratified this approach in its 1993 Brooke Group decision, which held that predation plaintiffs must establish that (1) the defendant charged below-cost prices, and (2) there is a "dangerous probability" that the defendant will recoup its losses by raising prices upon the elimination of competitors.⁷⁸

These requirements have proven difficult to satisfy. Since the Brooke Group decision, successful predatory-pricing claims have been rare.⁷⁹

Refusals to Deal

Another category of potentially exclusionary conduct involves refusals to deal with rivals. In general, firms—including monopolists—have the right to choose their business partners.⁸⁰ In certain cases, however, courts have held that declining to do business with competitors can harm competition without justification and thus violate the Sherman Act.

In Aspen Skiing Co. v. Aspen Highlands Skiing Corp., for example, the Supreme Court held that a monopolist of downhill skiing services in Aspen, Colorado violated Section 2 by terminating an "all-Aspen" ski ticket that it had offered with the plaintiff.⁸¹ The defendant also made it difficult for the plaintiff to replicate the "all-Aspen" package, refusing to sell the plaintiff lift tickets or accept bank-guaranteed vouchers included in the plaintiff's replacement ticket package.⁸² After concluding that the defendant had failed to offer a plausible efficiency justification for its conduct, the Supreme Court affirmed a verdict of Section 2 liability.⁸³

However, the Court later cabined the scope of its decision in Aspen Skiing, explaining that the case lies "at or near the outer boundary" of monopolization law.⁸⁴ The Court offered this guidance in Verizon Communications Inc. v. Trinko, in which it held that Verizon did not violate Section 2 by refusing to provide interconnection services to a rival local telephone service provider.⁸⁵ The Court distinguished Aspen Skiing on the ground that the monopolist in the latter decision had terminated "a voluntary (and thus presumably profitable) course of dealing" with the plaintiff, which suggested a willingness to sacrifice short-term profits for an anticompetitive end.⁸⁶ The Court also emphasized that the monopolist in Aspen Skiing refused to deal with its rival even if compensated at the prices it charged to other customers, which also revealed an anticompetitive purpose.⁸⁷ Because those factors were not present in Trinko, the Court held that Verizon's conduct did not fall within Aspen Skiing's "limited exception" to the principle that firms are free to refuse to deal with their competitors.⁸⁸

Based on the Supreme Court's reasoning in Trinko, some lower courts have concluded that refusal-to-deal plaintiffs must establish that a defendant's refusal entailed a sacrifice of short-term profits for an exclusionary purpose.⁸⁹ Some courts have also required plaintiffs to establish this type of profit sacrifice with proof that the defendant terminated a voluntary course of dealing.⁹⁰

Many circuit courts have also accepted a specific theory of refusal-to-deal liability called the "essential facilities" doctrine, which the Supreme Court has declined to either recognize or repudiate.⁹¹ To prevail under the essential-facilities doctrine, plaintiffs must establish

• 1. the control of an "essential facility" by a monopolist;
• 2. an inability to "practically or reasonably" duplicate the facility;
• 3. the denial of the use of the facility to a competitor; and
• 4. the feasibility of providing access to the facility.⁹²

While that doctrine remains on the books as a formal matter,⁹³ two commentators have described the Supreme Court's treatment of it as inflicting "death by dicta."⁹⁴ Its viability thus remains uncertain.

Refusal-to-deal doctrine implicates a well-recognized trade-off. On the one hand, compulsory dealing will often increase static efficiency. Requiring a vertically integrated monopolist to supply necessary inputs to downstream rivals, for example, may promote price competition in the downstream market and thereby eliminate allocative inefficiencies.⁹⁵ Those benefits, however, may come at the expense of dynamic competition insofar as they reduce incentives to invest and innovate.⁹⁶

As Trinko makes clear, antitrust doctrine currently places greater emphasis on the latter concern.⁹⁷ The Supreme Court has also expressed skepticism about the institutional competence of courts to craft appropriate remedies in refusal-to-deal cases. The worry is that compulsory dealing will often require generalist judges to set prices and other contract terms—tasks that are typically the province of a sectoral regulator.⁹⁸

As discussed below, in many cases, permissive refusal-to-deal doctrine has significant implications for the viability of antitrust action against major tech platforms under existing law.

Tying

Tying arrangements are vertical restraints of trade (i.e., restraints involving individuals or firms in a customer-supplier relationship) that can be challenged under several provisions of the antitrust laws, including Sections 1 and 2 of the Sherman Act.⁹⁹ Tying involves a refusal to sell one product (the tying product) unless buyers also purchase another product (the tied product) from the seller.¹⁰⁰

The basic concern with tying arrangements is that they may allow a firm with market power for the tying product to harm competition in and even monopolize the tied product market.¹⁰¹ Tying may also help a dominant firm preserve a monopoly in the tying market by eliminating potential rivals that may enter via the tied market.¹⁰²

However, tying can also produce procompetitive benefits. For example, tying may dissuade consumers from using an inferior substitute to the tied product with the tying product, mitigating the risk of reputational damage to a seller's brand.¹⁰³ Producing and selling different products together may also reduce production, marketing, and distribution costs.¹⁰⁴

Some ties can also serve as a means of price discrimination—for example, by allowing firms to discriminate between high-intensity and low-intensity users of a product.¹⁰⁵ Commentators have debated the effects of these "requirements" or "variable proportion" ties, whereby consumers purchase a durable tying product (e.g., a printer) and amounts of the tied product (e.g., ink) that vary with their use of the tying product. Firms may employ these types of ties to lower the price of the tying product and raise the price of the tied product, benefitting low-volume users and harming high-volume users.¹⁰⁶ Some commentators have argued that "requirements ties" typically increase total and consumer welfare,¹⁰⁷ while others have come to the opposite conclusion.¹⁰⁸

Like predatory-pricing doctrine, tying law has changed significantly over the course of antitrust history. Throughout much of the 20th century, courts were highly skeptical of tying arrangements, which were deemed per se illegal under Section 1 of the Sherman Act.¹⁰⁹ During this period of disapproval, the Supreme Court consistently described tying as inherently anticompetitive.¹¹⁰

As in other areas of antitrust, academic work challenged this attitude. Beginning in the 1950s, Chicago School scholars criticized the theory that a firm could leverage power in one market to extract additional profits from another market. They argued that when consumers use complementary products in fixed proportions—for example, nuts and bolts—a monopolist cannot extract additional profits by tying one product to the other.¹¹¹ In such cases, they reasoned, there is one profit-maximizing price for the product set, meaning a monopolist of nuts could extract only one monopoly profit from the nut-bolt set. If the market for bolts is competitive, charging a monopoly price for nuts while tying them to bolts sold at a supra-competitive price would result in a price for the nut-bolt set that exceeds the profit-maximizing level.¹¹² The Chicago critique of leverage theory thus contended that, in these circumstances, firms likely employ tying arrangements because they generate efficiencies.¹¹³

The single monopoly profit theory (SMPT) described above applies only under certain restrictive assumptions.¹¹⁴ In addition to being limited to complementary products used in fixed proportions,¹¹⁵ the SMPT does not eliminate the possibility that a firm may employ a tying arrangement to impair the efficiency of rivals in the tied market. If there are necessary scale economies in the tied market, for example, tying can potentially allow a firm to deny those economies to rivals and thus decrease the competitiveness of that market.¹¹⁶ The SMPT also does not preclude the use of a tying arrangement to maintain market power in the tying market (i.e., in cases where firms may enter the tying market via the tied market).¹¹⁷

Despite these limitations, the Chicago critique of traditional leverage theory—along with the development of various efficiency-based rationales for tying—ultimately led courts to move away from the view that ties are almost invariably anticompetitive.¹¹⁸ This change prompted an erosion of the per se rule. In decisions in the 1970s and 1980s, the Supreme Court retained the label of per se illegality for tying arrangements, but limited the rule's application to firms with sufficient market power in the tying market to force purchases of the tied product.¹¹⁹

Lower courts have adopted different formulations of this modified per se rule, but the inquiries are generally similar.¹²⁰ One commentator has summarized the doctrine as establishing the following requirements for a per se tying claim under Section 1:

• 1. The defendant offered two distinct products;
• 2. The defendant conditioned the sale of one product (the tying product) on the purchase of the other product (the tied product);
• 3. The defendant possessed sufficient economic power in the tying product market to coerce purchasers into acceptance of the tied product; and
• 4. The defendant's conduct affected a "not insubstantial" amount of interstate commerce in the tied product (an inquiry that focuses on the absolute dollar amount of affected commerce).¹²¹

Some lower courts have also required plaintiffs to demonstrate that a tying arrangement had anticompetitive effects in the tied product market.¹²² Others have entertained and accepted business justifications for challenged ties.¹²³ In practice, then, the modified per se rule against tying appears to be more similar to the rule of reason than it is to traditional per se rules.¹²⁴

Courts have also declined to apply the modified per se rule to ties involving platform software products. In its 2001 decision in United States v. Microsoft, the D.C. Circuit held that a tie involving Microsoft's Windows operating system and its Internet Explorer web browser was governed by the rule of reason, rather than the modified per se rule.¹²⁵ In rejecting application of the per se rule, the D.C. Circuit noted that none of the Supreme Court's tying cases had involved the physical and technological integration of separate products.¹²⁶ Condemning such ties without evaluating their competitive effects, the court reasoned, would create an unacceptable risk of error and deter innovation.¹²⁷ In 2023, the Ninth Circuit adopted the D.C. Circuit's reasoning to conclude that the rule of reason applied to a tying claim challenging Apple's requirement that software developers use Apple's payment processor for in-app purchases as a condition of distributing apps through its App Store.¹²⁸

As mentioned, tying arrangements can be challenged under Sections 1 and 2 of the Sherman Act. The key differences between the provisions are Section 1's requirement of an agreement; the availability of the modified per se rule under Section 1; and Section 2's requirement that challenged conduct contribute to the creation or maintenance of monopoly power (or produce a dangerous probability of those effects).¹²⁹

In evaluating product-design or "technological tie" claims under Section 2, some decisions have held that the integration of separate products is lawful when it improves quality or reduces cost, even if that conduct forecloses rivals.¹³⁰ The Microsoft decision, by contrast, employed a rule-of-reason-like burden-shifting framework to the government's Section 2 claims in that case.¹³¹ Another appellate decision has affirmed liability for product integration where evidence of an exclusionary motive cast doubt on the defendant's argument that the challenged design represented a genuine improvement.¹³²

Exclusive Dealing

Like tying arrangements, exclusive contracts—in which a firm commits to refrain from dealing with its counterparty's rivals—are vertical restraints of trade that can be challenged under Sections 1 and 2 of the Sherman Act.¹³³

Exclusive contracts can harm competition when a dominant firm uses them to foreclose rivals from key inputs or distribution channels.¹³⁴ They can also produce procompetitive benefits. For example, exclusivity may induce manufacturers to make relationship-specific investments in dealers by providing sales training, technical support, and other promotional assistance.¹³⁵ To the extent that a dealer can use any of this support to promote rival brands, manufacturers may lack the incentive to provide it. Exclusive dealing can eliminate this free-rider problem and thereby encourage investment.¹³⁶ Exclusivity may also mitigate uncertainty about future sales or purchases¹³⁷ and encourage more intense competition for distribution, which may result in lower consumer prices.¹³⁸

While exclusive dealing has never been deemed per se illegal, its treatment has evolved considerably. In its 1949 Standard Stations decision, the Supreme Court affirmed a decision finding that foreclosure of 6.7% of the relevant market was sufficient to render an exclusive contract illegal.¹³⁹ In doing so, the Court appeared to approve the lower court's refusal to engage in a full rule-of-reason analysis of competitive harm.¹⁴⁰ The decision thus stood for what came to be called the "quantitative substantiality" approach to exclusivity, which focused on the percentage of the relevant market foreclosed by a challenged agreement.¹⁴¹

The Supreme Court departed from that approach twelve years later in Tampa Electric Co. v. Nashville Coal Co., where it rejected a challenge to an exclusive contract that foreclosed less than 1% of the relevant market.¹⁴² In Tampa Electric, the Court did not limit its analysis to the low foreclosure percentage, explaining that it was necessary to also engage in a qualitative analysis of the agreement's competitive effects.¹⁴³

The quantitative aspect of foreclosure analysis has also become more permissive. In the Court's 1984 Jefferson Parish decision, the concurring opinion of four Justices concluded, without a detailed inquiry, that foreclosure of 30% of the market was not sufficient to render an exclusive contract unlawful.¹⁴⁴

Since these decisions, reviewing courts have tended to require foreclosure of at least 40% of the market before condemning exclusive contracts under Section 1, while also analyzing the duration of the restrictions, any business justifications, and other factors that may bear on an agreement's competitive effects.¹⁴⁵

Some courts have indicated that the standards for assessing exclusive dealing are more plaintiff-friendly under Section 2, and that a monopolist's use of exclusive contracts may be illegal even if they foreclose less than the 40% figure that is typically necessary for a Section 1 violation.¹⁴⁶

Mergers & Acquisitions

The antitrust laws also place limitations on mergers and acquisitions.¹⁴⁷ Section 7 of the Clayton Act prohibits a merger if its effect "may be substantially to lessen competition, or to tend to create a monopoly."¹⁴⁸ Though less common, Section 2 of the Sherman Act has also been used to challenge mergers that help a firm acquire or maintain monopoly power.¹⁴⁹

Analysis of mergers varies based on the relationship between the merging parties—specifically, based on whether a merger is horizontal, vertical, or conglomerate.

Horizontal mergers (i.e., mergers between competitors) receive the greatest scrutiny and can raise two primary types of concerns. First, horizontal mergers may allow a firm to unilaterally increase its prices or decrease the quality of its products by eliminating competition between rivals.¹⁵⁰ Second, horizontal mergers may facilitate tacit or express collusion by increasing market concentration (so-called "coordinated effects").¹⁵¹

Vertical mergers (i.e., mergers between firms in the same supply chain) receive less exacting scrutiny than horizontal ones, because they do not eliminate direct competitors and are thought to often generate efficiencies.¹⁵² The main concern with vertical mergers is foreclosure; when a firm acquires an important source of inputs or a key distribution channel, it may have the ability and incentive to raise rivals' costs or refuse to do business with rivals altogether.¹⁵³ A vertical merger may also prompt concerns if it gives a firm access to competitively sensitive information about rivals or facilitates collusion by allowing the merged entity to monitor compliance with tacit pricing agreements.¹⁵⁴

Conglomerate mergers are mergers that are neither horizontal nor vertical.¹⁵⁵ Challenges to such mergers are rare.¹⁵⁶ Conglomerate mergers may raise antitrust concerns, however, if they allow a firm to acquire a potential competitor.¹⁵⁷

Merger law has evolved significantly over the last 50 years. During the Warren Court era from the early 1950s through the 1960s, the Supreme Court heard 12 merger cases, siding with the plaintiff in each case where it reached the merits.¹⁵⁸ Some of the Court's decisions blocked small mergers in unconcentrated markets, based in part on a concern about stopping an incipient trend toward concentration and a desire to effectuate congressional intent to protect small businesses.¹⁵⁹

In this period, courts were heavily influenced by an approach to industrial organization often called the "structure-conduct-performance" (SCP) paradigm, which held that market concentration tended to produce less competitive markets with higher prices.¹⁶⁰ The impact of these theories was made clear in the Supreme Court's 1963 decision in United States v. Philadelphia National Bank, which recognized a presumption of illegality for mergers that would result in a firm controlling "an undue percentage share of the relevant market" while significantly increasing market concentration.¹⁶¹

This "structural presumption" remains good law, but subsequent developments have chipped away at its strength. In its 1974 decision in United States v. General Dynamics Corp., the Supreme Court held that the defendant coal-mine operator had successfully rebutted the presumption with evidence that almost all of the acquired firm's coal reserves were depleted or committed under long-term contracts.¹⁶² Lower court decisions later interpreted General Dynamics as demanding a more detailed inquiry into a merger's competitive effects than was evident in the Warren Court's merger decisions.¹⁶³ This shift coincided with a wave of academic criticism directed at SCP theories. Among other things, SCP's detractors argued that high levels of market concentration are often necessary for firms to achieve economies of scale and scope and that many concentrated markets perform competitively.¹⁶⁴

Today, much of the action in merger enforcement takes place in the antitrust agencies rather than the courts. This is partly the result of Congress's adoption of the Hart-Scott-Rodino Antitrust Improvements Act (HSR Act) in 1976, which created a pre-merger notification regime that allows the DOJ and FTC to review mergers exceeding certain numerical thresholds before they close.¹⁶⁵

Since 1968, the agencies have published guidelines outlining their analytical approach to merger review, including their application of the structural presumption.¹⁶⁶ Starting with the 1982 guidelines, the agencies have relied on the Herfindahl-Hirschman Index (HHI) measure of market concentration in applying the presumption.¹⁶⁷ Revisions to the guidelines in 2010 increased the minimum concentration levels at which the agencies regard horizontal mergers as potentially problematic.¹⁶⁸ While the guidelines are not legally binding, courts often treat them as persuasive authority and appear to accord some significance to the relevant HHI thresholds.¹⁶⁹

As discussed, the structural presumption can be rebutted—for example, with evidence that the proposed market is poorly defined or that market shares do not reflect a merger's likely competitive effects; that the entry of other firms will discipline any pricing power; or that the merger will produce efficiencies that offset any anticompetitive effects.¹⁷⁰ Upon rebuttal of a prima facie case, the burden of producing further evidence of anticompetitive harm shifts back to the plaintiff and merges with the burden of persuasion.¹⁷¹

While the case law on vertical mergers is sparse,¹⁷² the most recent appellate decision reviewing a vertical deal employed a burden-shifting approach that is similar to the framework used to evaluate horizontal mergers.¹⁷³ However, the court indicated that plaintiffs challenging vertical mergers cannot rely on the structural presumption to discharge their initial burden.¹⁷⁴ Instead, the court explained that such plaintiffs must make a fact-specific showing that a transaction is likely to be anticompetitive,¹⁷⁵ which will presumably often involve foreclosure concerns.

The current state of merger law is something of an oddity. Although the Supreme Court's 1960s merger decisions have not been formally overturned, they do not accurately reflect the "law on the ground" as applied by the antitrust agencies and the lower courts.¹⁷⁶ Since the Warren Court, for example, merger doctrine has abandoned "non-economic" goals like the protection of small businesses.¹⁷⁷ While structural evidence continues to play a role in merger analysis, its centrality has diminished as regulators and courts also consider a broader range of factors that may illuminate a transaction's competitive effects.¹⁷⁸ That may be changing, however. In 2023, the DOJ and FTC released draft merger guidelines that appear to place more weight on structural considerations than previous iterations of the guidelines.¹⁷⁹ It remains to be seen whether courts will follow the agencies in this regard, should the regulators finalize the guidelines in similar form.

Theoretical Approaches to Antitrust

As the above discussion makes clear, antitrust doctrine has changed significantly over time, often in response to shifts in political ideology and economic theory. Congress has played a limited role in this evolution; the language of the core antitrust statutes has not meaningfully changed since the Celler-Kefauver Act amended Section 7 of the Clayton Act in 1950.¹⁸⁰ Because the flexible nature of the antitrust laws gives the judiciary broad powers to shape competition policy based on prevailing economic and political thinking,¹⁸¹ this section provides a brief overview of the leading theoretical approaches to antitrust and their historical influence.

As discussed, SCP theories exerted a strong influence on antitrust policy in the middle of the 20th century.¹⁸² This approach to industrial organization was developed by scholars working in a tradition often referred to as the Harvard School, which posited a close causal link between market concentration, firm conduct, and competitive performance.¹⁸³ In particular, the SCP literature held that there was a tight connection between high levels of market concentration and certain undesirable outcomes, such as high price-cost margins.¹⁸⁴

For much of antitrust history—including during the heyday of the SCP paradigm—"non-economic" goals also played a major role in shaping antitrust doctrine. These goals included the protection of small businesses, the dispersion of economic power, the preservation of economic freedom, and the elimination of concentrated political power.¹⁸⁵

From the 1940s through the 1960s, structuralist economic theories and the above normative concerns provided the theoretical architecture for a highly interventionist approach to antitrust, judged by today's standards. As discussed, the Warren Court's merger jurisprudence was quite restrictive, invalidating small mergers based in part on a desire to "promote competition through the protection of viable, small, locally owned business," even if "occasional higher costs and prices might result from the maintenance of fragmented industries and markets."¹⁸⁶

Conduct cases during this era reflected similar attitudes. In the federal government's monopolization case against Alcoa, for example, Judge Learned Hand of the Second Circuit reasoned that the Sherman Act was motivated in part by a belief that "great industrial consolidations are inherently undesirable, regardless of their economic results."¹⁸⁷ He thus construed the statute as an attempt to "put an end to great aggregations of capital because of the helplessness of the individual before them."¹⁸⁸ Based on these principles, the Second Circuit held that Alcoa violated Section 2 by expanding its capacity in ways that deterred entry.¹⁸⁹ In its Section 1 cases during this period, the Supreme Court likewise condemned a wide range of conduct as per se illegal.¹⁹⁰

The 1970s witnessed a marked shift in theory and doctrine. As discussed, beginning in the 1950s, lawyers and economists affiliated with what came to be known as the Chicago School challenged much of prevailing antitrust thinking.¹⁹¹ Chicago School scholars criticized SCP theories on a variety of grounds. Among other things, they argued that markets tend to self-correct; that high levels of concentration often reflect growth by the most efficient firms; and that many business practices that attracted antitrust scrutiny had efficiency-based rationales.¹⁹² The Chicago School's most influential contribution, however, was its prescription that antitrust should be limited to promoting economic welfare.¹⁹³ An antitrust system that instead committed itself to a series of often-conflicting social objectives, Chicago School scholars claimed, offered no principled method for distinguishing anticompetitive behavior from permissible conduct.¹⁹⁴

Chicago's empirical claims did not go unchallenged. Scholars working in the "Post-Chicago" tradition generally embraced the Chicago School's focus on economic goals, but developed theories of anticompetitive harm that were not present in the Chicago literature.¹⁹⁵ Many of these Post-Chicago models highlighted the possibility that dominant firms could employ strategic behavior to raise their rivals' costs, relying heavily on game theory.¹⁹⁶ Another group of academics from the so-called "modern Harvard School" tended to fall somewhere between the Chicago School and the ideology of mid-20th century antitrust, focusing on the administrability of antitrust doctrine and the institutional limitations of courts.¹⁹⁷ Like Post-Chicago scholars, the modern Harvard School endorsed the Chicago view that the ultimate purpose of the antitrust laws is to promote economic welfare.¹⁹⁸ The three approaches differ primarily in their empirical claims about market functioning and the competence of courts to remedy market failures.¹⁹⁹

In general, the Chicago School and the modern Harvard School have had the greatest impact on the shape of current doctrine.²⁰⁰ Since the 1970s, the Supreme Court has overturned several decisions establishing per se Section 1 liability for certain categories of conduct²⁰¹ and established restrictive standards for various types of monopolization claims.²⁰² Similarly, the lower courts and the antitrust agencies have de-emphasized structural merger analysis in favor of more detailed inquiries into the competitive effects of individual transactions.²⁰³

Modern antitrust doctrine has also abandoned explicit consideration of "non-economic" goals like small business protectionism and the sociopolitical effects of concentrated economic power.²⁰⁴ Since the 1970s, the Supreme Court has repeatedly described the antitrust laws as being principally concerned with the economic welfare of consumers.²⁰⁵ This proposition—often called the "consumer welfare standard"—has generated an enormous amount of scholarly attention, especially in recent years. While there is disagreement about what the standard does and should mean in practice,²⁰⁶ contemporary doctrine clearly recognizes economic welfare as the lodestar of antitrust analysis.²⁰⁷

Over the past decade or so, this economic orientation has been criticized by a group of academics and policymakers often described as "Neo-Brandeisians." Members of this movement have criticized much of existing antitrust doctrine as unduly permissive and called for increased attention to some of the "non-economic" goals that played a more prominent role in earlier periods of antitrust history.²⁰⁸ The Neo-Brandeisian movement's prescriptions are discussed in greater detail below.²⁰⁹

The Big Tech Firms: A Summary of Selected Antitrust Allegations

The Big Tech firms have achieved tremendous financial success. As of the publication of this report, the combined market capitalization of Meta, Alphabet (Google's parent), Amazon, and Apple is more than $6.6 trillion—a figure that exceeds the value of all but the largest national equity markets.²¹⁰

While some have emphasized the quality of the firms' offerings as the primary driver of their ascent,²¹¹ others have alleged that Big Tech has obtained and cemented monopoly power through anticompetitive conduct.²¹²

This section of the report reviews selected antitrust allegations against the Big Tech firms.

Meta Platforms

Meta describes itself as a company that builds technology that "helps people connect, find communities, and grow businesses."²¹³ More specifically, Meta offers a "family of apps" related to social networking and messaging.²¹⁴ This family of apps consists of

• Facebook (a social network);
• Instagram (a photo-sharing platform);
• Messenger (a messaging app for Facebook users); and
• WhatsApp (a messaging app).²¹⁵

In October 2023, Meta reported that Facebook had 2.09 billion daily active users and 3.05 billion monthly active users.²¹⁶ The company's family of apps reportedly features 3.14 billion daily active people and 3.96 billion monthly active people.²¹⁷

Allegations of Market Power

Some observers have argued that Meta possesses significant market power in the market for social networking.²¹⁸ The Federal Trade Commission (FTC) shares that view. In an ongoing monopolization lawsuit, the Commission alleges that Meta has held monopoly power in the market for "personal social networking services" (PSNS) since at least 2011.²¹⁹ To support such claims, Meta's critics have argued that the firm has persistently maintained a large market share and benefited from substantial entry barriers, including powerful network effects and high switching costs.²²⁰

Others disagree. Meta has argued that it operates in a "dynamic, intensely competitive" industry in which there are many substitutes for its services.²²¹ In its litigation with the FTC, the firm has criticized the Commission's alleged PSNS market as unduly narrow insofar as it excludes rivals like YouTube, TikTok, LinkedIn, and Twitter.²²²

Meta and some commentators have also rejected the notion that entry barriers have caused the market to decisively tip in Meta's favor.²²³ For example, observers have highlighted the ability of differentiated firms like TikTok and Snapchat to rapidly gain scale despite Meta's ostensible network advantages.²²⁴

Allegations of Anticompetitive Conduct

Meta has also been accused of engaging in anticompetitive conduct. The FTC's lawsuit contends that Meta has maintained its dominant position through its 2012 acquisition of Instagram and its 2014 acquisition of WhatsApp.²²⁵ The Commission argues that Meta's Instagram purchase allowed it to neutralize a rapidly growing competitive threat, giving the firm control over what became two of the most popular social networks in the world.²²⁶ The FTC also contends that Meta's acquisition of WhatsApp preserved its monopoly by preventing WhatsApp from entering the PSNS market.²²⁷

Besides targeting Meta's major acquisitions, the FTC and some commentators have criticized the company's treatment of software developers.²²⁸ These allegations involve access to Facebook Platform—an initiative whereby Meta encouraged developers to create apps that interoperate with Facebook.²²⁹ As part of this initiative, Meta provided software developers with application programming interfaces (APIs) and other tools that allowed them to access certain Facebook data and functionalities.²³⁰ According to the FTC, Facebook Platform ultimately became key infrastructure for app developers because of Facebook's large user base.²³¹ The Commission's lawsuit alleges that Meta leveraged control of this infrastructure to preserve its monopoly, requiring developers that participated in Facebook Platform to refrain from creating apps that would compete with Facebook products.²³²

Meta has denied engaging in anticompetitive conduct. The company has argued that its Instagram acquisition allowed it to invest resources and expertise in a young startup, hastening the small firm's growth.²³³ Meta has also defended its WhatsApp purchase, arguing that the FTC has failed to present evidence that WhatsApp would have likely entered social networking absent the acquisition.²³⁴ Finally, Meta has argued that its policies governing access to Facebook Platform—which it has since revised—were lawful under duty-to-deal doctrine.²³⁵

As of the publication of this report, the FTC's monopolization case against Meta is in discovery, a pre-trial stage of litigation in which the parties develop evidence that can be used at trial. Although the district court dismissed the agency's initial complaint for failing to plausibly allege monopoly power,²³⁶ the court ultimately allowed the case to proceed after concluding that the Commission's amended complaint was sufficiently plausible to survive a motion to dismiss.²³⁷

Google

Google is a ubiquitous presence in the digital economy. The firm began as an internet search company and is now also a major player in digital advertising, mobile operating systems, app distribution, digital maps, email, and web browsing.²³⁸ The following subsections discuss antitrust allegations involving Google's conduct related to online search, mobile operating systems and app distribution, and digital advertising.

Online Search

Allegations of Market Power

Some commentators have argued that Google has significant market power in the market for general online search.²³⁹ The DOJ agrees. In an ongoing monopolization lawsuit, the DOJ contends that Google has monopoly power in the market for "general search services" based on an alleged market share of 88% and the presence of substantial entry barriers, including economies of scale.²⁴⁰ The DOJ has also alleged monopolization of separate markets for "general search text advertising" and "search advertising."²⁴¹

For its part, Google has claimed that it operates in a "highly competitive environment" and faces a "vast array of competitors."²⁴² The company also argues that, for particular search queries, it competes against a range of firms—such as Amazon, eBay, and Yelp—that would not fall within a market for general search services.²⁴³

Allegations of Anticompetitive Conduct

Search Distribution

The DOJ's monopolization lawsuit contends that Google has maintained its search monopoly through exclusionary agreements with firms that control search distribution.²⁴⁴ The agreements make Google the default search engine on various products in exchange for a share of Google's advertising revenue.²⁴⁵

The lawsuit focuses on Google's agreements with two categories of counterparties: (1) browser developers, and (2) manufacturers and wireless carriers that sell devices running the Android mobile operating system, which Google acquired in 2005.²⁴⁶

Under Google's agreements with browser developers—primarily Apple and Mozilla—the developers have agreed to make Google the default search engine for all search access points on their browsers in exchange for payments from Google.²⁴⁷

The DOJ's allegations regarding device manufacturers and wireless carriers involve two types of agreements. One set of contracts requires manufacturers to preinstall Google Search and place an associated search widget on device home screens as conditions of licensing other proprietary Google apps.²⁴⁸ Under another set of agreements, manufacturers and wireless carriers commit to make Google the only preinstalled search engine on covered devices and the default for all search access points in exchange for payments from Google.²⁴⁹

In its lawsuit, the DOJ contends that Google's agreements amount to exclusive contracts that foreclose substantial channels of search distribution, depriving rivals of the scale needed to serve as effective competitors.²⁵⁰

Google has made several arguments in response. While Google concedes that its revenue-sharing agreements with manufacturers and carriers require exclusivity, it has denied that its contracts with browser developers and its Android licensing agreements amount to exclusive dealing.²⁵¹

The browser agreements are not exclusive, Google contends, because they do not prevent developers from promoting rival search engines and users can change a browser's default search engine.²⁵² Similarly, Google argues that its Android licensing agreements do not prohibit manufacturers from preinstalling rival search apps or browsers.²⁵³

Google further maintains that the relevant agreements would not be anticompetitive even if they did require exclusivity. Rather, Google claims that it has successfully competed with rivals to secure the challenged agreements with browser developers, who have chosen Google as their default search engine based on considerations of quality and price.²⁵⁴ Google claims that this "competition for the contract" is the type of merits competition that antitrust encourages.²⁵⁵

Google also denies that its Android agreements result in substantial foreclosure, arguing that the appropriate measure of foreclosure requires an analysis of consumer behavior absent the agreements, rather than the percentage of the market covered by those agreements.²⁵⁶ Because few consumers would switch from Google to another search engine if the challenged agreements did not exist, Google argues, the agreements do not substantially foreclose rivals.²⁵⁷

In August 2023, a federal district court allowed the claims discussed above to proceed, denying in part Google's motion for summary judgment.²⁵⁸ The court concluded that there were disputed issues of material fact as to whether Google's browser agreements and Android licensing agreements were de facto exclusive, including disagreement over the competitive significance of default status.²⁵⁹ The court likewise held that Google's "competition for the contract" defense and the appropriate measure of foreclosure raised issues that could not be resolved on summary judgment.²⁶⁰

Foreclosure metrics are likely to be a key issue at trial, which began in September 2023 and is ongoing as of the publication of this report. While courts regularly look to the share of distribution covered by exclusive contracts in evaluating foreclosure, some commentators have advocated alternative approaches, including the type of counterfactual analysis that Google proposes.²⁶¹ The choice between alternative metrics may present the court with a trade-off: while Google's preferred methodology arguably involves a more accurate assessment of the competitive impact of the challenged agreements, the DOJ's simpler approach may be more manageable and appears to be more firmly rooted in the case law.²⁶²

The issue of procompetitive justification may also prove significant. The DOJ's case relies heavily on the importance of scale in improving search-engine quality. If Google can establish that it has not exhausted the relevant scale economies, those economies may constitute a procompetitive justification for its distribution agreements.²⁶³ To the extent that the court accepts this justification, the DOJ may need to prove that Google can secure those economies through less restrictive means or that the anticompetitive effects of the agreements outweigh their benefits.

Self-Preferencing

Commentators and some foreign regulators have also argued that Google has leveraged its dominance in general search to favor its own vertical offerings. For example, the HJC Report concluded that Google has adjusted its search algorithms to automatically elevate some of Google's vertical services, like its video-sharing platform YouTube, in search results.²⁶⁴

This type of self-preferencing prompted the European Commission—which enforces European Union competition law—to fine Google €2.42 billion in 2017 for giving prominent placement to its comparison-shopping service and demoting rival services in search results.²⁶⁵

The FTC investigated similar allegations of self-preferencing involving Google Search in 2012, but concluded that it had not found sufficient evidence of an antitrust violation.²⁶⁶ The agency determined that Google's favorable placement of its own verticals could plausibly be viewed as an improvement in the quality of Google's search product.²⁶⁷ The Commission also did not find sufficient evidence that Google had manipulated its search algorithms to unfairly disadvantage rival vertical websites.²⁶⁸

Mobile Operating Systems and App Distribution

Allegations of Market Power

Mobile Operating Systems

In addition to operating a major search engine, Google controls Android—a leading mobile operating system. Android and Apple's iOS represent the two dominant mobile operating systems, together accounting for 99% of the market.²⁶⁹ Because Apple does not license iOS to other device manufacturers, Android by itself occupies a very large share of the market for licensable mobile operating systems—by some estimates, 99% of that market.²⁷⁰

Some commentators have argued that the market for licensable mobile operating systems is the relevant one for antitrust purposes, based on factors like high switching costs.²⁷¹ Private plaintiffs and (in a separate case that has been settled) a group of state attorneys general have argued that Google has monopoly power in this market based on the company's dominant market share and the presence of substantial entry barriers, such as network effects and research and development costs.²⁷²

Google denies such allegations, arguing that consumers "can and do switch and multi-home among and between mobile and nonmobile ecosystems, including between Android and iOS."²⁷³

Mobile App Distribution

Litigants have also contended that, through its Google Play Store, Google has market power in certain markets related to mobile-app distribution. Some plaintiffs have defined the relevant market as consisting of the distribution of apps to Android users.²⁷⁴ They allege that Google has monopoly power in this market based on the Play Store's market share of more than 90%, strong network effects, high switching costs, and Google's ability to charge a 30% commission on apps purchased through the Play Store.²⁷⁵

Some plaintiffs have also argued in the alternative that Google has market power in a broader market for mobile app distribution—that is, a market not limited to Android users.²⁷⁶ A group of state attorneys general, for example, has argued that Google occupies a sizeable share of this market, enjoys large profit margins, and benefits from formidable entry barriers.²⁷⁷

Google rejects these claims. It contends that consumers can use different platforms to access apps and that "Apple and Google compete vigorously in the mobile operating system environment on multiple dimensions, including innovation, price, privacy, and security."²⁷⁸

In-App Payment Processing

Plaintiffs have further claimed that Google has monopoly power in a market for in-app payment (IAP) processing for Android apps.²⁷⁹ They have based this claim on the Play Store's large share of the market for Android app distribution and Google's requirement that software developers using the Play Store also use Google's IAP processor.²⁸⁰

As discussed, for many transactions, Google charges a 30% commission for IAP processing—a rate that is considerably higher than those charged by other electronic payment processors.²⁸¹

Google has denied possessing monopoly power related to IAP processing.²⁸²

Allegations of Anticompetitive Conduct

Mobile App Distribution

Google has also been accused of engaging in a variety of anticompetitive activities involving app distribution. Some of the allegations are reviewed below.

• Google has allegedly imposed technical barriers that make it difficult for consumers to download Android apps from sources other than the Google Play Store—a practice commonly known as "sideloading."²⁸³ Litigants have claimed that sideloading Android apps entails a complicated process that includes several security warnings discouraging such actions.²⁸⁴ Google has also been accused of making it unnecessarily difficult to update sideloaded apps.²⁸⁵
• Google has allegedly barred software developers from distributing competing app stores through the Play Store.²⁸⁶
• Google has allegedly required mobile device manufacturers that license Android and certain other key Google services to preinstall the Google Play Store on their devices.²⁸⁷ Plaintiffs have argued that this preinstallation requirement harms competition by giving the Play Store an advantage over other app stores.²⁸⁸
• Google has allegedly required device manufacturers that offer the Play Store and other "must-have" Google services to refrain from selling devices that run "Android forks"—modified versions of Android that Google has not approved.²⁸⁹ Plaintiffs argue that these restrictions have stifled the development of alternative versions of Android that would be free from some of the restrictions on app distribution discussed above.²⁹⁰
• Google has allegedly entered into revenue-sharing agreements that deter device manufacturers from developing competing app stores.²⁹¹ The challenged agreements give device manufacturers a share of Google's advertising and Play Store revenue from the devices they sell in exchange for a commitment to refrain from competing against the Play Store.²⁹²

Google has either denied engaging in the relevant conduct or rejected the contention that the alleged conduct is anticompetitive.²⁹³

In-App Payment Processing

Plaintiffs have also accused Google of engaging in anticompetitive conduct in the market for Android IAP processing. They have alleged that Google's requirement that developers using the Play Store also use Google's IAP processor represents an unlawful tying arrangement.²⁹⁴

Digital Advertising

Allegations of Market Power

In addition to its search and app-distribution activities, Google is a major force in digital display advertising markets.

In those markets, online ad publishers—like news websites—sell advertising space through exchanges.²⁹⁵ Those ad exchanges conduct automated auctions in which advertisers can bid for ad space.²⁹⁶

Intermediaries facilitate this process for both publishers and advertisers. Large publishers manage their ad inventory using a type of software known as an ad server, which interfaces with ad exchanges on behalf of publishers.²⁹⁷ On the other side of the market, advertisers employ ad-buying tools, which connect them with ad exchanges and allow them to purchase ad space.²⁹⁸

Google operates in several segments of these markets via an ad exchange, a publisher ad server, and ad-buying tools for advertisers.²⁹⁹

The DOJ and (in a separate lawsuit) a group of state attorneys general (state AGs) have argued that Google has monopoly power in multiple ad-tech markets.

In January 2023, the DOJ filed a complaint alleging that Google has monopoly power in the markets for publisher ad servers,³⁰⁰ ad exchanges,³⁰¹ and advertiser ad networks.³⁰²

In a separate case, a group of state AGs has alleged that Google has monopoly power in the markets for ad exchanges, ad servers, and ad-buying tools for small advertisers.³⁰³ The state AGs also contend that Google has monopoly power or a dangerous probability of acquiring monopoly power in the market for ad-buying tools for large advertisers.³⁰⁴

In September 2022, a federal district court concluded that the state AGs' allegations of monopoly power were sufficiently plausible to survive a motion to dismiss.³⁰⁵ In April 2023, a district court likewise rejected Google's motion to dismiss the DOJ's ad-tech lawsuit.³⁰⁶

Allegations of Anticompetitive Conduct

The DOJ and state AG lawsuits contend that Google has engaged in a range of anticompetitive practices in several digital-advertising markets, allowing it to obtain and cement a dominant position across the ad-tech stack.

The DOJ's lawsuit claims that, in the early 2000s, Google's ad-buying tools occupied a dominant position on the advertiser side of the ad-tech market.³⁰⁷ Then, in 2008, Google acquired a firm called DoubleClick, which operated a leading publisher ad server and a nascent ad exchange.³⁰⁸

After the DoubleClick acquisition, the DOJ contends, Google leveraged its position across the ad-tech chain to benefit its own properties. Among other things, the DOJ alleges that Google made demand from its ad-buying tools available only through its ad exchange.³⁰⁹ Google also allegedly required publishers to use its ad server to receive real-time bids from its ad exchange.³¹⁰ The state AG ad-tech lawsuit makes similar allegations.³¹¹

In September 2022, a federal district court held that the state AGs had plausibly alleged tying claims under Sections 1 and 2 of the Sherman Act based on their assertion that Google had coerced publishers into using its ad server as a condition of receiving live bids from its ad exchange.³¹²

The DOJ and state AG lawsuits also target a program used by Google's ad server that allegedly gave Google's ad exchange advantages over rival exchanges.³¹³ Another set of accusations involves programs under which Google allegedly manipulated bids from its advertiser clients in ways that advantaged its ad exchange and publisher ad server.³¹⁴

Google maintains that its conduct is permissible under antitrust doctrine governing refusals to deal, product design, and tying.³¹⁵

The September 2022 district court decision in the state AG lawsuit concluded that the allegations of anticompetitive harm from these activities were sufficiently plausible to survive a motion to dismiss.³¹⁶ As noted, the district court in the DOJ's lawsuit also denied Google's motion to dismiss that case, which focused on the allegations of monopoly power.³¹⁷

The Google ad-tech lawsuits are complex, and a full discussion of the relevant claims is beyond the scope of this report.³¹⁸ Most of the allegations nevertheless implicate a recurring theme in discussions of antitrust and tech platforms: the leveraging of economic power to obtain and solidify dominance across different markets.³¹⁹

Amazon

Like Google, Amazon has expanded its remit over time. The company began as an online bookseller, but now operates a leading e-commerce marketplace, a major cloud-computing platform, a logistics network, and a television and film studio.³²⁰ The discussion below focuses on the company's e-commerce activities.

Allegations of Market Power

In a pending monopolization lawsuit, the FTC has alleged that Amazon has monopoly power in two markets: the "online superstore" market and the "online marketplace services" market.³²¹ The FTC argues that "online superstores"—which are distinguished based on the breadth and depth of their product offerings—are not reasonably interchangeable with other online stores because consumers' overall shopping costs would increase "dramatically" if they tried to replace online superstores by shopping at several different online stores with more limited offerings.³²² Brick-and-mortar stores are not adequate substitutes for online superstores, the complaint alleges, because of the convenience, wider product selection, and personalized shopping experience offered by online superstores.³²³

The FTC also identifies a separate market for "online marketplace services" offered to sellers.³²⁴ The relevant services include access to a significant number of customers; the ability to create and maintain pages with product information; and the ability to display customer reviews to shoppers.³²⁵ The complaint alleges that "online marketplace services" are not reasonably interchangeable with selling as a vendor to online or offline retail stores, which typically involves wholesale pricing and transfer of title to the relevant products.³²⁶

The FTC's complaint follows similar lawsuits, including an action filed by the D.C. Attorney General (D.C. AG) under D.C. law and another by a putative class of consumers under the Sherman Act.³²⁷ The D.C. AG's complaint alleged that Amazon has monopoly power among online marketplaces,³²⁸ while the consumer lawsuit contends that Amazon has monopoly power in a retail e-commerce market and several e-commerce submarkets for specific products.³²⁹

In 2022, the Superior Court of the District of Columbia dismissed the D.C. AG lawsuit on several grounds, including a failure to plausibly allege monopoly power.³³⁰ The D.C. AG has appealed that decision.³³¹ In contrast, a federal district court has denied Amazon's motion to dismiss the consumer lawsuit, concluding that the plaintiffs plausibly alleged monopoly power, in addition to rule-of-reason claims under Section 1 of the Sherman Act.³³²

Amazon's alleged monopoly power will likely be litigated vigorously. In identifying an "online superstore" market, the FTC appears to be alleging a "cluster" market that consists of a vast array of noncompeting goods.³³³ Courts have recognized cluster markets in other contexts. For example, groups of noncompeting financial and medical services have been deemed to be relevant antitrust markets in cases involving banks and hospitals.³³⁴ In another case, the Supreme Court concluded that the relevant market consisted of a package of centrally monitored alarm services.³³⁵

While some courts have recognized cluster markets based on considerations of administrative convenience (i.e., where distinct markets face similar competitive conditions, obviating the need for separate analyses), the FTC appears to rely on a different theory of clustering grounded in "transactional complementarity."³³⁶ Under this theory, a package of noncompeting goods or services may qualify as a relevant antitrust market if a significant number of consumers would be willing to pay supra-competitive prices for the convenience of receiving the goods or services as a package.³³⁷ Commentators have also argued that economies of scope and network effects may be rationales for clustering noncompeting products in the same market.³³⁸ Whether "online superstores" represent an appropriately defined market for any of these reasons will likely turn on the factual evidence that the FTC can ultimately adduce.

By alleging separate markets for "online superstores" and "online marketplace services," the FTC's complaint also raises questions regarding the impact of the Supreme Court's 2018 decision in Ohio v. American Express (Amex).³³⁹ In Amex, the Court held that two-sided transaction platforms like credit-card networks represent a single market, meaning a price increase on one side of such a market (there, an increase in merchant fees) cannot by itself demonstrate an anticompetitive exercise of market power.³⁴⁰ Instead, the Court concluded that the plaintiffs in Amex needed to show anticompetitive effects on the credit-card market "as a whole"—for example, that the defendant's conduct increased the price or reduced the number of credit-card transactions.³⁴¹

Amex's implications for the FTC's case against Amazon are unclear. In a 2018 article, the current FTC Chair argued that the Supreme Court's reasoning in Amex appeared to apply to Amazon's marketplace.³⁴² While the article did not elaborate on that assessment, there are similarities between Amazon's platform and credit-card networks. In Amex, the Court justified its single-market conclusion on the ground that credit-card networks cannot make sales "unless both sides of the platform [i.e., merchants and cardholders] simultaneously agree to use their services."³⁴³ As a result, the Court reasoned, credit-card networks cannot set prices for one side of the market without considering the impact of those prices on the other side of the market.³⁴⁴ Similar dynamics may be at work in Amazon's case. By allegedly charging monopoly prices to sellers, Amazon may risk losing participants on that side of its platform, which would decrease the value of its marketplace to consumers. If consumers shop elsewhere as a result, that would further diminish the value of Amazon's platform to sellers, setting off a negative feedback loop.³⁴⁵ The court may thus rely on Amex to reject the FTC's effort to define separate markets for "online superstores" and "online marketplace services."

Some commentators, however, have highlighted possible distinctions between Amazon's marketplace and credit-card networks. While credit-card networks do little besides facilitate transactions, for example, Amazon offers sellers a range of additional services.³⁴⁶ Whether these types of distinctions will allow the FTC to sidestep Amex's single-market rule for two-sided transaction platforms remains to be seen.

Allegations of Anticompetitive Conduct

Anti-Discounting Measures

Several lawsuits have alleged that Amazon has implemented measures to punish sellers on its marketplace for offering lower prices in other transaction venues. The FTC's lawsuit contends that Amazon disqualifies sellers from appearing in its Buy Box—which features a product's price and the "Add to Cart" button, among other information—if Amazon discovers sellers offering their products for a lower price in another online store.³⁴⁷ The complaint further alleges that Amazon has entered into contracts with certain important sellers that prohibit the sellers from discounting their products in other online stores.³⁴⁸ The FTC claims that these anti-discounting measures prevent rival online marketplaces from offering products at lower prices and deprives those rivals of necessary scale.³⁴⁹

The D.C. AG lawsuit and the consumer class action discussed above made similar allegations.³⁵⁰ As discussed, the Superior Court of the District of Columbia has dismissed the D.C. AG's complaint,³⁵¹ and the D.C. AG has appealed that decision.³⁵² The district court in the consumer lawsuit, by contrast, denied Amazon's motion to dismiss, concluding that the plaintiffs had plausibly alleged that Amazon's conduct caused anticompetitive harm.³⁵³

Amazon has rejected the allegation that the relevant policies are anticompetitive, arguing that they reflect a decision to highlight products that are competitively priced.³⁵⁴

Tying of Amazon Prime and Amazon's Fulfillment Service

The FTC's lawsuit also alleges that Amazon maintains its monopolies by coercing sellers to use its fulfillment services (i.e., storing, packaging, and preparing products for shipment).³⁵⁵ Specifically, the FTC contends that Amazon effectively requires sellers to use its fulfillment services as a condition of participating in Amazon Prime—a subscription program that offers customers fast shipping of eligible products, among other benefits.³⁵⁶ Prime eligibility boosts a seller's chances of winning the Buy Box, the FTC alleges, while sellers that forgo Prime eligibility "effectively disappear from Amazon's storefront."³⁵⁷

The FTC argues that Amazon's tying of Prime to its fulfillment services stifles the growth of other online marketplaces in two ways. First, by allegedly tying Prime to its fulfillment services, Amazon effectively requires sellers that want to use both Amazon's marketplace and other online marketplaces to use two separate fulfillment providers.³⁵⁸ This duplication, the FTC contends, creates extra costs that could be avoided by consolidating inventory with one fulfillment provider, which deters sellers from using other online marketplaces.³⁵⁹ Second, Amazon's conduct allegedly prevents independent fulfillment providers from gaining necessary scale, which likewise increases the costs to sellers of utilizing multiple online marketplaces.³⁶⁰

In response, Amazon has said that it allows sellers that participate in Prime to use other fulfillment providers as long as those providers "are able to meet . . . Prime customers' high expectations for fast, reliable delivery."³⁶¹

Use of Third-Party Seller Data

Amazon's dual role as both a marketplace operator and a seller on its own marketplace has also attracted scrutiny. Critics have contended that this integration generates conflicts of interest, which have led Amazon to leverage control of its marketplace to advantage its own products and services in various ways.³⁶²

Some of these allegations involve Amazon's use of data. The HJC Report and European regulators have accused Amazon of using data generated by third-party sellers on its marketplace to identify and imitate popular products for its private-label business.³⁶³

During congressional testimony in July 2020, Amazon's founder and former chief executive said that the company has a policy against using seller-specific data to aid its private-label business.³⁶⁴ He indicated, however, that he could not guarantee that this policy had never been violated.³⁶⁵ Amazon reportedly does not have a policy against using aggregated seller data to assist its retail business.³⁶⁶

Commentators have disputed the competitive effects of a platform's use of user data to enter new markets. Some have argued that Amazon's entry into new markets forces other sellers to lower their prices—an outcome that antitrust traditionally encourages.³⁶⁷ Others contend that the alleged copying may have longer-term anticompetitive effects by chilling incentives to innovate.³⁶⁸

Self-Preferencing

Amazon's dual role as a marketplace operator and private-label seller has led to a range of other concerns about self-preferencing. For example, a 2016 ProPublica investigation concluded that Amazon designed the ranking algorithm for its marketplace to favor its own offerings and products offered by sellers that use its fulfillment services.³⁶⁹ The HJC Report alleged that Amazon has engaged in other forms of self-preferencing, such as refusing to allow certain competitors to advertise on Amazon's platform.³⁷⁰

Predatory Pricing

Amazon has also been accused of engaging in predatory pricing at various points in its history.³⁷¹ These allegations have been directed against several aspects of Amazon's business, including its sale of e-books;³⁷² its sale of diapers and ultimate acquisition of the parent company of Diapers.com;³⁷³ and Amazon Prime.³⁷⁴

In previous academic work, the current FTC Chair has argued that Amazon exemplifies the rationality of predatory pricing in markets characterized by strong network effects and extreme scale economies, contrary to the assumptions that underpin current doctrine.³⁷⁵

Other commentators have challenged these allegations.³⁷⁶ In response to the claims involving Diapers.com, some have noted that Amazon has not been accused of occupying a monopolistic share of the market for online diaper sales or diaper sales generally.³⁷⁷ Others have argued that the HJC Report failed to produce sufficient evidence to conclude that Amazon prices Prime memberships below cost.³⁷⁸ Commentators have also questioned whether Amazon's critics are relying on a coherent concept of predation, contending that some of the relevant literature appears to reject the idea that raising prices at some point is a necessary part of a predatory strategy.³⁷⁹

Apple

Apple is the most valuable company in the world.³⁸⁰ The firm designs, manufactures, and sells iPhone smartphones, Mac personal computers, iPad tablets, and several wearables and accessories, in addition to offering a range of related services.³⁸¹ The discussion below focuses on issues related to the company's mobile operating system and App Store.

Allegations of Market Power

As discussed, Apple's iOS and Google's Android are the two dominant operating systems for mobile devices in the United States and globally.³⁸² More than half of the mobile devices in the United States run a version or derivation of iOS.³⁸³ Apple's App Store is the only method by which software developers can distribute apps on iOS devices; Apple does not allow iOS users to download other app stores or sideload apps.³⁸⁴ Like Google, Apple requires developers to use its IAP processor as a condition of accessing its App Store and has charged 30% commissions for that service.³⁸⁵

In 2020, Epic Games—the developer of the video game Fortnite—challenged these restrictions under Sections 1 and 2 of the Sherman Act. Epic alleged single-brand aftermarkets for iOS app distribution and iOS in-app payment processing in which Apple possessed monopoly power.³⁸⁶

Apple denied Epic's allegations, arguing that the relevant market consists of all video game transactions, including transactions involving gaming consoles, personal computers, and streaming services.³⁸⁷

In September 2021, a federal district court arrived at a conclusion that fell between the two parties' positions. Instead of a single-brand aftermarket for iOS app distribution or a market for video-game distribution generally, the court held that Apple competes in a market for digital mobile gaming transactions, which includes transactions on iOS and Android mobile devices.³⁸⁸

The district court rejected Epic's narrower proposed aftermarkets for several reasons, including Epic's failure to establish consumer unawareness of Apple's restrictions or produce evidence of the magnitude of the relevant switching costs.³⁸⁹

The court ultimately concluded that Apple possesses market power—but not monopoly power—in the market for digital mobile gaming transactions.³⁹⁰

Epic appealed that decision. In April 2023, the Ninth Circuit took issue with some aspects of the district court's market-definition analysis, but affirmed the court's rejection of Epic's proposed aftermarkets and the holding that Apple does not possess monopoly power.³⁹¹

Allegations of Anticompetitive Conduct

Mobile App Distribution and IAP Processing

As discussed, the Epic Games lawsuit challenged Apple's requirement that iOS app developers distribute their apps through Apple's App Store and use Apple's payment processor for in-app purchases.³⁹² The lawsuit also targeted anti-steering provisions in Apple's developer agreements, which prohibit developers from using certain communications methods—such as in-app links—to inform users about out-of-app payment options.³⁹³

In Epic Games, the district court concluded that the first two restrictions did not violate the Sherman Act, but that Apple's anti-steering provisions violated California competition law.³⁹⁴ In rejecting Epic's federal antitrust claims, the district court concluded that Apple had proffered valid procompetitive justifications for its App Store and payment-processor requirements based on their promotion of security, privacy, and the monetization of intellectual property.³⁹⁵ The district court further concluded that Epic had not shown that those procompetitive benefits could be achieved through less restrictive means.³⁹⁶

On appeal, the Ninth Circuit affirmed the district court's decision.³⁹⁷ In affirming the district court's rejection of Epic's Section 1 tying claim, the appellate court concluded that the modified per se rule against tying does not apply to ties involving platform software products, following the D.C. Circuit's reasoning in Microsoft.³⁹⁸ While the Ninth Circuit disagreed with some aspects of the district court's rule-of-reason analysis, it affirmed the finding that Epic failed to show that Apple could achieve the relevant procompetitive benefits through less restrictive means.³⁹⁹

Self-Preferencing

The HJC Report alleged that Apple has taken a variety of steps to preference its own apps and harm rival app developers.⁴⁰⁰ Among other things, the report accused Apple of injuring competition by preinstalling its own apps on iPhones;⁴⁰¹ denying third-party apps access to certain APIs and device functionalities that are available to its own apps;⁴⁰² favoring its own apps in search results on its App Store;⁴⁰³ and removing rival apps from the App Store.⁴⁰⁴

Apple has denied giving preferential treatment to its own apps in search rankings.⁴⁰⁵ The company has claimed that it has removed specific apps from its App Store based on violations of its privacy policies.⁴⁰⁶

Use of Competitively Sensitive Information

Like Amazon, Apple has faced allegations that it uses its access to data generated by dependent businesses to identify and imitate popular offerings.⁴⁰⁷ In particular, software developers have accused Apple of using competitively sensitive information about popular apps to build competing apps and integrate certain functionalities into iOS.⁴⁰⁸

Apple has responded to such allegations by stating that it does not violate other companies' intellectual property rights.⁴⁰⁹

Big Tech Mergers and Acquisitions

Some of the allegations discussed above involve Big Tech mergers and acquisitions. As noted, the FTC is currently challenging Facebook's acquisitions of Instagram and WhatsApp,⁴¹⁰ while Google's acquisition of DoubleClick is a key part of the DOJ's monopolization lawsuit targeting the company's ad-tech practices.⁴¹¹

Some policymakers have expressed broader concerns about Big Tech mergers.⁴¹² The companies have been active dealmakers: between 2000 and 2019, the four firms engaged in hundreds of mergers and acquisitions.⁴¹³ Many of the transactions fell below the numerical thresholds that trigger pre-merger review by the antitrust agencies.⁴¹⁴

These deals have prompted some commentators to worry that the Big Tech firms are cementing their dominant positions by acquiring promising potential competitors.⁴¹⁵ Transactions involving "nascent" competitors have been a particular point of concern.⁴¹⁶ While the concept of a nascent competitor has been defined in different ways, it generally refers to an innovative firm that represents a serious yet uncertain future threat to an incumbent.⁴¹⁷

These issues are discussed in greater detail in "Mergers & Acquisitions" infra.⁴¹⁸

Antitrust Reform and Big Tech: General Issues

The issues discussed above have prompted calls for reform. Some proposals would supplement the antitrust laws with sectoral competition regulations directed at large technology platforms.⁴¹⁹ Others would work within the existing antitrust framework by adjusting burdens of proof and changing certain doctrinal rules.⁴²⁰

While the relevant options are varied, they all implicate the threshold question of whether tech platform markets have unique features that warrant special treatment under competition law. The proposals that would supplement antitrust with a new regulatory regime raise additional questions regarding scope and administration.

This section of the report discusses these general issues in the debate over antitrust reform directed at Big Tech firms.

Are Tech Platforms Special?

As discussed, outside of a narrow set of per se offenses, antitrust is a fact-specific enterprise. Generally, courts employ a case-by-case approach to evaluate claims of anticompetitive behavior.⁴²¹ Because liability typically depends on case-specific facts rather than the application of bright-line rules, antitrust investigations and litigation are often time-consuming and expensive.⁴²² The open-ended nature of the relevant legal standards can also make it difficult to predict whether certain conduct violates the law, which may undermine enforcement by allowing large firms to profit from anticompetitive strategies and treat potential lawsuits as a cost of doing business.⁴²³

Advocates of reform have argued that these features of antitrust adjudication make it ill-suited to deal with tech platform markets characterized by a unique confluence of structural characteristics, such as strong network effects, economies of scale, economies of scope derived from user data, and consumer tendencies to single-home.⁴²⁴

According to some, these characteristics cause certain platform markets to tip in favor of a single dominant firm.⁴²⁵ After an initial period of competition, one company may gain an edge that becomes self-reinforcing. For example, a platform with a large user base and associated data advantages may be the most attractive to new users, generating a positive feedback loop that allows it to grow even larger and thereby become even more attractive.⁴²⁶ Prospective entrants may then face difficulties achieving the scale necessary to compete with the dominant incumbent.⁴²⁷

Big Tech firms may also derive benefits from their roles as gatekeepers for key digital ecosystems, like mobile operating systems, app stores, online marketplaces, and social networks.⁴²⁸ By controlling access to these ecosystems and setting the rules within them, tech platforms can allegedly preserve their dominant positions and leverage those positions to obtain advantages in related markets.⁴²⁹

Some analysts contend that antitrust adjudication is too slow to adequately police markets characterized by these winner-take-all dynamics.⁴³⁰ By the time a market has tipped, they suggest, remedies for anticompetitive conduct may be unable to restore meaningful competition.⁴³¹ Occasionally, this line of argument involves the claim that a particular digital platform is a natural monopoly, meaning its cost structure is such that market demand can be served most efficiently by a single firm.⁴³² In such industries—public utilities are prominent examples—competition is unable to discipline market participants, which has traditionally led policymakers to favor direct regulations of price, entry, and quality of service.⁴³³

Other commentators have rejected the claim that antitrust is unable to grapple with competition issues involving large digital platforms. Some dispute that Big Tech markets have all decisively tipped in favor of a single firm.⁴³⁴ Rather, they contend that the tech giants compete in diverse markets characterized by different competitive dynamics.⁴³⁵ While some of those markets may be susceptible to tipping, others arguably retain a competitive fringe or exhibit competition among rivals of comparable size.⁴³⁶ This variety is said to emerge from several characteristics that distinguish many online platforms from traditional natural monopolies, including product differentiation, multi-homing by consumers, and low switching costs.⁴³⁷ Defenders of the current antitrust regime have thus emphasized the heterogeneity of platform markets, which they contend militates against categorical treatment of Big Tech firms and in favor of the existing fact-specific approach.⁴³⁸

Some observers have gone further in their rejection of sectoral regulation, arguing that many tech markets have features that support a cautious approach to intervention under the existing antitrust laws. This perspective emphasizes the distinction between static price competition and the dynamic rivalry prevalent in tech markets, which involves efforts to develop new products, services, and business models.

Commentators have argued that dynamic competition has non-interventionist implications for several areas of antitrust doctrine.

First, in dynamic industries, traditional market definition may overstate the market power of leading firms.⁴³⁹ In particular, static market shares may exaggerate a platform's market power by failing to account for the threat of displacement by differentiated or innovative rivals.⁴⁴⁰ Examples of this type of displacement include Facebook supplanting MySpace as the leading social network and Google unseating Yahoo! and AltaVista to become the dominant search engine. In both cases, the ousted incumbents were widely perceived as invulnerable—an assumption that proved incorrect.⁴⁴¹ Some observers have highlighted these episodes in arguing that Big Tech platforms face constant competitive pressure despite occupying large static market shares.⁴⁴²

Second, technological innovation often involves combining products or features that were previously available only as separate offerings. Although this type of product integration often benefits consumers, it is also potentially vulnerable to antitrust challenge under tying law.⁴⁴³ To avoid chilling innovation, some courts and commentators have endorsed exceptions to the modified per se rule against tying for platform software products.⁴⁴⁴ Other observers have gone further and advocated a rule of per se legality for the introduction of new products.⁴⁴⁵

Third, dynamic competition may raise complications in assessing claims of price predation. The alleged difficulty arises from the fact that technology firms often offer low or zero prices to encourage widespread adoption of new products. In winner-take-all markets with network effects, all firms may charge low prices—which may entail short-term losses—with an eye toward recoupment once they surpass rivals.⁴⁴⁶ Some commentators have questioned the appropriateness of traditional cost and recoupment tests for predation claims in markets characterized by this type of competition.⁴⁴⁷

Others have argued that the two-sided nature of certain platform markets raises additional difficulties for predation analysis. The potential trouble involves indirect network effects, whereby a platform becomes more valuable to users on one side (e.g., advertisers, merchants) as it gains more users on the other side (e.g., users of a search engine, e-commerce customers).⁴⁴⁸ In such markets, evaluating a firm's pricing on only one side of a platform may be misleading. For example, a firm may offer below-cost prices to one side of a platform to attract more users on the other side, where it charges above-cost prices.⁴⁴⁹ Looking only to one side of a two-sided market may thus yield inaccurate conclusions about predation.⁴⁵⁰

Fourth, some have raised more general concerns about the role of antitrust in tech markets. This strand of the literature builds on the error-cost framework originally developed by Frank Easterbook,⁴⁵¹ who relied on decision theory to argue that antitrust rules should err on the side of permissiveness.⁴⁵² Specifically, Easterbrook—who is now a federal judge—reasoned that monopoly profits eventually induce the entry of new firms, mitigating the costs of judicial decisions permitting anticompetitive conduct (false negatives or Type II errors).⁴⁵³ In contrast, market forces cannot correct decisions condemning procompetitive conduct, leading Easterbrook to conclude that the costs of those mistakes (false positives or Type I errors) are more durable.⁴⁵⁴

Several commentators have suggested that Easterbrook's argument has particular force in technology markets.⁴⁵⁵ They have appealed to what is often called the "inhospitality tradition" of 1960s antitrust—an era in which courts were highly skeptical of nonstandard agreements and business conduct.⁴⁵⁶ In one narrative, once courts realized that many challenged practices had benign or procompetitive explanations, the law underwent a needed course correction in which context-specific inquiries into economic effects replaced the formalism of per se rules as the dominant mode of antitrust analysis.⁴⁵⁷ Some argue that this history has implications for the optimal level of intervention in tech markets, where generalist judges may mistake novel products and business strategies for anticompetitive conduct.⁴⁵⁸

Others have come to the opposite conclusion. John Newman—formerly the Deputy Director of the FTC's Bureau of Competition—has argued that false negatives in tech markets are far more common and costly than false positives.⁴⁵⁹ Newman contends that the structure of many digital markets insulates incumbents from competitive threats; that digital markets provide incumbents with unique anticompetitive strategies; and that challenged conduct in digital markets typically has few redeeming benefits.⁴⁶⁰ These features, he maintains, justify more vigilant antitrust scrutiny of tech markets, contrary to what he characterizes as the "orthodox" view of error costs.⁴⁶¹

Revisiting the Goals of Antitrust: The Neo-Brandeisian Movement

The optimal level of antitrust intervention—in tech markets and more generally—depends on the underlying goals of antitrust law.⁴⁶² As discussed, the last 40 years have been marked by a general (though not complete) consensus that antitrust should be limited to promoting some conception of economic welfare.⁴⁶³ While there are lingering disputes within the welfarist approach,⁴⁶⁴ modern doctrine generally eschews "non-economic" considerations like equity, the protection of small businesses, and the promotion of democratic values.⁴⁶⁵

In the past decade, a group of scholars and activists commonly referred to as "Neo-Brandeisians" has mounted a critique of this consensus.⁴⁶⁶ Members of this movement have argued that the existing antitrust regime has failed to preserve competition, resulting in rising economic concentration, growing wealth inequality, and a political system captured by corporate interests.⁴⁶⁷ Their influence has not been limited to the academy; some self-described Neo-Brandeisians—like FTC Chair Lina Khan and former White House advisor Tim Wu—have occupied policymaking positions within the federal government.⁴⁶⁸

Two separate arguments are evident throughout Neo-Brandeisian scholarship.

First, Neo-Brandeisians deny that antitrust should serve only economic goals like consumer welfare.⁴⁶⁹ Instead, they endorse a broader normative vision in which antitrust deconcentrates markets, promotes fairness, and disperses economic and political power.⁴⁷⁰

Second, Neo-Brandeisians have argued that the consumer-welfare standard has failed even when judged on its own terms. They allege that, in applying an exclusively economic approach to antitrust, courts have embraced simplistic theories that downplay the harms from concentration and the likelihood of exclusion, leading to uncompetitive markets with higher prices and degraded quality.⁴⁷¹ Some Neo-Brandeisians have argued that these putative deficiencies are especially severe vis-à-vis large tech platforms, many of which are able to forgo immediate profits to establish long-term dominance and then leverage that dominance across business lines.⁴⁷²

This second set of arguments is not unique to Neo-Brandeisians. As discussed, the Post-Chicago tradition challenged the laissez-faire prescriptions of Chicago School academics from within the welfarist paradigm.⁴⁷³ Many commentators also continue to criticize current doctrine as unduly permissive on economic grounds.⁴⁷⁴ The distinctiveness of the Neo-Brandeisian critique thus lies in its call for an expansion of the range of antitrust goals—not its repudiation of conservative economic theories.⁴⁷⁵

The key difficulty facing this project involves operational specifics. One of the central criticisms of mid-20th century antitrust was that it allegedly offered no principled standard for weighing many of the conflicting goals that courts had read into the antitrust statutes.⁴⁷⁶ Some business conduct, for example, may benefit consumers while harming a firm's smaller rivals.⁴⁷⁷ An antitrust regime that embraces both consumer welfare and small-business protection may thus have difficulty trading off those values in a coherent fashion. Other constituencies—workers, labor unions, potential entrants, existing competitors—may also have divergent interests.⁴⁷⁸ Adding abstract principles like fairness and democracy to the calculus could create even further uncertainty.⁴⁷⁹

The Neo-Brandeisian response appears to involve a preference for bright-line rules over the type of fact-intensive analysis employed by current law.⁴⁸⁰ Neo-Brandeisians may thus conceptualize their favored goals as operating at the level of rule formulation, but not as factors that courts must balance in individual cases.

It is unclear, however, whether Neo-Brandeisians intend to reduce all of antitrust doctrine to bright-line rules.⁴⁸¹ Moreover, as discussed below, some competition legislation directed at tech platforms would stop short of adopting categorical prohibitions. Instead, it would proscribe specified conduct only upon a showing of harm to "competition," or ban specified conduct while offering an affirmative defense to platforms that show the relevant activities do not harm "competition."⁴⁸² If this language is not intended to denote a welfarist conception of competitive harm, the ambiguities surrounding alternative understandings of "competition" resurface.

Several commentators—including some Neo-Brandeisians—have supported a "protection of the competitive process" test as an alternative to the consumer-welfare standard.⁴⁸³ The details of this approach, however, remain hazy.⁴⁸⁴ Some advocates have equated it with the promotion of deconcentrated market structures.⁴⁸⁵ Jonathan Kanter—the Assistant Attorney General for the DOJ's Antitrust Division—has similarly described the "competitive process" as protecting "the guarantee that everyone participating in the open market—consumers, farmers, workers, or anyone else—has 'the free opportunity to select among alternative offers.'"⁴⁸⁶

As legal standards, these formulations have been criticized for failing to offer meaningful guidance to courts or litigants. It seems unlikely, for example, that they are intended to mean that any conduct that reduces the number of competitors in a market should be illegal; that rule would outlaw many instances of procompetitive price cutting and product improvement, in addition to small mergers in unconcentrated markets.⁴⁸⁷ The proposals instead appear to endorse an approach that ensures there are "enough" competitors, though it is not clear whether Neo-Brandeisians would endorse the use of some further analytical criterion in evaluating whether a given market has "enough" firms.⁴⁸⁸

Thus far, Neo-Brandeisian theories do not appear to have made considerable headway with the courts.⁴⁸⁹ Some commentators, however, have argued that the movement's influence is evident in several developments at the antitrust agencies, including the FTC's 2023 proposal to ban non-compete clauses in employment contracts,⁴⁹⁰ the DOJ's effort to revive its long-dormant authority to pursue criminal monopolization charges,⁴⁹¹ and the 2023 draft merger guidelines.⁴⁹² Whether Neo-Brandeisian ideas will have a broader impact on antitrust doctrine or legislative action remains to be seen.

Scoping Reform Proposals

As discussed, some legislative proposals would create special competition rules for large technology platforms.

Two general models have emerged. One model involves special rules for digital platforms that offer specified services and meet certain quantitative and qualitative criteria intended to capture platforms with bottleneck power over business users.⁴⁹³ Because proposals in this category involve the designation of covered platforms by regulators, this report refers to this strategy as the "designated-platform approach."⁴⁹⁴

The second model is narrower. While the designated-platform approach would apply the same set of rules to covered firms in a range of markets (e.g., social networking, e-commerce, online search), some legislation would apply only to individual markets.⁴⁹⁵ This report refers to this strategy as the "market-specific approach."

The subsections below review these two models for sector-specific competition rules.

The Designated-Platform Approach

Policymakers in the United States and EU have explored the designated-platform approach. The EU has adopted legislation called the Digital Markets Act, which applies special regulations to designated "gatekeepers."⁴⁹⁶ Firms are to be designated as "gatekeepers" if they offer certain "core platform services"—including search engines, app stores, operating systems, advertising services, social networking, and online marketplaces—and meet certain quantitative and qualitative criteria.⁴⁹⁷

In the United States, several bills in the 117th Congress would have adopted a broadly similar approach.⁴⁹⁸ One of those bills has been reintroduced in the 118th Congress,⁴⁹⁹ and other legislation in the 118th Congress employs a comparable strategy.⁵⁰⁰

The proposals would empower a regulator to designate a platform offering any of the relevant services as a covered platform based on (1) quantitative thresholds involving market capitalization, annual sales, and active users, and (2) the platform's status as a "critical trading partner."⁵⁰¹

Different bills would impose different types of competition rules on designated platforms, including rules involving discriminatory conduct against business users,⁵⁰² vertical integration,⁵⁰³ and mergers.⁵⁰⁴

Depending on the interpretation of the "critical trading partner" standard, the designation criteria may encompass the platforms discussed earlier in this report:

• Facebook, Instagram, and WhatsApp (which are controlled by Meta Platforms);
• Google Search, Android, the Google Play Store, and some of Google's ad-tech services;
• Amazon Marketplace; and
• Apple's iOS and App Store.⁵⁰⁵

Certain Microsoft properties and TikTok—a short-form video app controlled by the Chinese firm ByteDance—may also fall within the bills' coverage.⁵⁰⁶

The designation criteria employed in these proposals raise several issues. Some commentators have criticized the use of market capitalization and annual sales as factors that would determine a firm's regulatory status.⁵⁰⁷ Those criteria, they contend, have little relevance for a platform operator's ability to harm competition, which instead depends on a firm's market power.⁵⁰⁸ As discussed, courts typically assess claims of market power by evaluating a firm's size within a relevant antitrust market—not its absolute size.⁵⁰⁹ Critics of the designated-platform bills thus argue that the proposals employ arbitrary designation criteria intended to single out a small handful of companies.⁵¹⁰

The bills seek to address some of these concerns about arbitrariness with the additional requirement that covered platforms include only "critical trading partners"—a term defined to mean persons with the ability to "restrict or impede" a business user's access to customers or tools or services needed to effectively serve customers.⁵¹¹ This phrase would represent a novel addition to the antitrust lexicon.

The use of the new "critical trading partner" language instead of the more familiar concept of market power may be a response to some of the more demanding elements of market-power doctrine. Market definition—which is required if a plaintiff seeks to establish market power via indirect evidence—often involves a costly and time-consuming battle of economic experts.⁵¹² The "critical trading partner" terminology may be motivated in part by a desire to ease these burdens.

Some commentators have also lodged theoretical objections to the centrality of market definition in contemporary antitrust.⁵¹³ Among other things, they have highlighted the limitations of binary market analysis when it comes to differentiated products.⁵¹⁴ Products that fall within a relevant antitrust market, for example, all count as equally effective substitutes for the product at issue; the market-definition paradigm does not consider different rates of substitution among products within a relevant market.⁵¹⁵ Similarly, firms deemed to fall outside a relevant market are treated as if they exert no competitive pressure on a defendant.⁵¹⁶

Reality is often more nuanced. In markets with differentiated products—like many technology markets—there may be a range of firms that compete with a defendant to various degrees. Singling out a specific market boundary along this type of continuum may thus yield inaccurate assessments of market power.⁵¹⁷

The "critical trading partner" requirement thus appears to respond to dissatisfaction with existing law. However, the requirement's precise relationship with current doctrine is unclear. The core concern of market definition—the availability of reasonable substitutes—seems relevant to whether a platform has the ability to "restrict or impede" a business user's access to customers or necessary tools. As a result, some of the considerations that figure in market definition would potentially play a role in evaluations of the "critical trading partner" requirement. The exact ways in which this inquiry may differ from traditional market definition accordingly remain uncertain.

The literature also reflects different views of the requirement's stringency. Some commentators have argued that the relevant bills are "carefully targeted" because they would apply only to "critical trading partners."⁵¹⁸ Others have contended that the additional criterion is unlikely to exclude firms that meet the bills' quantitative thresholds.⁵¹⁹ The analytical framework governing assessments of the "critical trading partner" standard would thus have to be fleshed out in practice, if Congress were to enact legislation employing that concept.⁵²⁰

The use of the "critical trading partner" language instead of a market-power requirement is not inherent to the designated-platform approach. In empowering a regulator to designate platforms for special competition regulation, Congress could consider limiting designations to firms that possess significant market power.

In the 117th Congress, S. 1074 would have taken that approach.⁵²¹ The bill would have imposed special merger rules on "dominant digital firms"—a term defined to mean companies that provide online services and possess "dominant market power" in any market related to such services.⁵²² Under the legislation, the FTC would have been empowered to designate companies as "dominant digital firms" based on their possession of "dominant market power" and several other factors, including network effects, use of exclusivity agreements, and vertical integration.⁵²³

Besides these issues involving designation criteria, the designated-platform approach implicates the broader question of whether the Big Tech firms (and any other designated firms) are sufficiently similar to warrant uniform regulatory treatment. As discussed, some commentators have argued that certain platform markets share structural similarities that justify a consistent regulatory response, while others have emphasized the differences between those markets. For proponents of new competition regulations, that issue may be the central question that determines the choice between the designated-platform approach and market-specific regulation.

The Market-Specific Approach

Some proposals for sectoral competition regulation rely on a more targeted strategy than the designated-platform approach. Instead of applying the same set of rules to designated firms operating across a range of different tech markets, policymakers could adopt regulations tailored to individual markets. In recent years, lawmakers have introduced bills targeting two industries: app stores and digital advertising.⁵²⁴

In the 117th Congress, the Open App Markets Act (OAMA) would have established competition rules for large app stores.⁵²⁵ Among other things, the legislation would have prohibited operators of covered app stores from tying their app stores to their payment processors; preferencing their own apps in search results; and using nonpublic information derived from third-party apps to compete with those apps.⁵²⁶ The bill's requirements are discussed in greater detail in "Ex Ante Conduct Rules" infra.

In the 118th Congress, the Advertising Middlemen Endangering Rigorous Internet Competition Accountability (AMERICA) Act would impose structural-separation requirements and conduct rules on certain digital-advertising platforms.⁵²⁷ The legislation would prohibit firms with more than $20 billion in annual digital-advertising revenue from owning platforms that operate in more than one of the key nodes in the ad-tech supply chain (ad exchanges, sell-side brokerages, and buy-side brokerages).⁵²⁸ It would also require firms with more than $5 billion in annual digital-advertising revenue to abide by customer-protection rules involving best execution and transparency.⁵²⁹

Enforcement

Reform proposals have taken different approaches to issues of enforcement.

In the 118th Congress, the Digital Consumer Protection Commission Act (DCPCA) would establish a new federal agency tasked with regulating large online platforms.⁵³⁰ The new agency would have a broad mandate covering competition, transparency, privacy, and national security issues.⁵³¹

On the competition front, the DCPCA would charge the new agency with enforcing prohibitions of abuses of dominance⁵³² and conflicts of interest,⁵³³ in addition to granting the agency the authority to block platform mergers under a public-interest standard.⁵³⁴

The bill's abuse-of-dominance provision would make it unlawful for a covered platform operator to abuse its dominance or "otherwise engage in conduct that harms competition or creates or helps maintain an unfair method of competition, a monopoly, or a monopsony, regardless of any alleged procompetitive benefit or efficiencies."⁵³⁵ While the legislation identifies several practices as presumptive violations of this prohibition, it would also give the new agency rulemaking authority to identify additional practices as presumptive violations.⁵³⁶ Platforms could rebut a presumptive violation with clear and convincing evidence that their conduct "did not result in any harm to the relevant aggrieved party."⁵³⁷

Other proposals adopt a different approach. As discussed, several bills in the 117th Congress (one of which has been reintroduced in the 118th Congress) would have empowered the DOJ and FTC to designate firms based on certain quantitative and qualitative criteria.⁵³⁸ Designated firms would then be subject to special competition rules. The bills would have charged the DOJ and FTC with enforcing the relevant prohibitions in federal court, but would not have given the agencies rulemaking authority to expand or clarify their scope.

Reform Proposals

This final section of the report discusses the substance of various proposals to reform the competition laws governing Big Tech platforms. The proposals fall into five categories: (1) ex ante conduct rules, (2) structural separation and line-of-business restrictions, (3) special merger rules, (4) interoperability and data-portability mandates, and (5) changes to general antitrust doctrine.

Ex Ante Conduct Rules

As discussed, some commentators have advocated the adoption of prophylactic conduct rules for Big Tech platforms, which would supplement general antitrust law. This subsection reviews several of these proposals for ex ante competition regulation.

Self-Preferencing

The ability of large digital platforms to preference their own offerings is a recurring concern in debates over antitrust reform. As discussed, several of the Big Tech firms have been accused of engaging in various forms of self-preferencing. Google has allegedly favored its own verticals in general search results; its own app store and apps through its control of Android; and its own ad-tech businesses through its presence in multiple segments of the ad-tech market.⁵³⁹ Apple has likewise been accused of preferencing its own apps and app store,⁵⁴⁰ while Amazon has allegedly privileged its own private-label products and products that use its fulfillment services.⁵⁴¹

The primary concern with this type of conduct involves monopoly leveraging.⁵⁴² As discussed, leveraging theories of harm can take two forms. Offensive leveraging occurs when a firm attempts to use monopoly power in a primary market to extract additional profits from a secondary market.⁵⁴³ By contrast, defensive leveraging involves the use of monopoly power to gain an advantage in a secondary market so as to preserve a primary market monopoly—for example, by eliminating competitive threats that might emerge from the secondary market.⁵⁴⁴

Defensive leveraging may be a viable theory of harm under existing monopolization law.⁵⁴⁵ Offensive-leveraging claims, however, cannot succeed under Section 2 absent evidence that a defendant had a dangerous probability of monopolizing a secondary market; mere harm to competition in the secondary market is not sufficient.⁵⁴⁶

For some of the self-preferencing allegations against Big Tech firms, these limitations may preclude antitrust claims.⁵⁴⁷ It may be unlikely, for example, that Amazon will achieve monopoly power over most of the products that it sells on its marketplace. As a result, it would be difficult to challenge the preferential display of those products under an offensive-leveraging theory.⁵⁴⁸ This type of alleged favoritism may also be a weak foundation for a defensive-leveraging or monopoly-maintenance case; it is not clear that Amazon's elevation of allegedly inferior products would help it maintain a putative e-commerce monopoly.

Similarly, the case law governing refusals to deal may serve as an impediment to antitrust claims challenging platform self-preferencing. A platform operator's favorable treatment of its own verticals relative to rivals that use its platform is typically less harmful to rivals than an outright refusal of access.⁵⁴⁹ Because antitrust imposes access duties only in a narrow set of circumstances, courts would likely find many forms of self-preferencing to be permissible if such conduct is evaluated as a refusal to deal.⁵⁵⁰

In the 118th Congress, the American Innovation and Choice Online Act (AICOA) would respond to these doctrinal difficulties by prohibiting covered platform operators from preferencing their own products and services "in a manner that would materially harm competition."⁵⁵¹

Given the ubiquity of self-preferencing by vertically integrated firms, the meaning of the "materially harm competition" standard is key to assessing the prohibition's scope. However, many argue the meaning of that language is not clear.⁵⁵² The bill does not by its terms clarify whether the "materially harm competition" standard embodies a consumer-welfare test or one of the alternative standards for assessing competitive harm urged by proponents of antitrust reform.⁵⁵³ As a result, it is unclear whether the AICOA would permit defendants to justify challenged conduct on the ground that it benefits consumers.

If the AICOA becomes law, this may be a dispositive issue in many litigated cases. A wide range of platform self-preferencing may harm a firm's rivals while also offering consumer benefits. For example, when Google displays a Google Maps result in response to a search query, it may disadvantage rival map services, but benefit consumers.⁵⁵⁴ Apple's preinstallation of its own apps on iPhones, Microsoft's inclusion of certain apps with its Windows operating system, and Amazon's free provision of its video-streaming service to Amazon Prime members may have similar effects.⁵⁵⁵

It is not clear how the "materially harm competition" standard would apply to such practices. In cases that do not involve per se offenses, Sherman Act defendants typically have the opportunity to defend challenged conduct on the ground that it benefits consumers.⁵⁵⁶ To the extent that the "materially harm competition" standard is intended to incorporate prevailing concepts of competitive harm from the antitrust case law, then, consumer-welfare arguments would likely be cognizable.

In interpreting other industry-specific competition statutes, however, some courts and commentators have taken the view that "harm to competition" encompasses types of harm beyond those proscribed by the antitrust laws.⁵⁵⁷ Additionally, some of the AICOA's proponents have rejected suggestions to amend the bill to adopt a consumer-welfare test.⁵⁵⁸

An interpretation that eschewed consumer-welfare justifications would also be consistent with the normative vision articulated by many advocates of antitrust reform. As discussed, the role that consumer welfare is meant to play in non-welfarist conceptions of "competition" is not clear.⁵⁵⁹ Much of the reformist literature, though, appears to reject the idea that courts and enforcers should balance different antitrust goals against one another.⁵⁶⁰ This context, along with the bill's omission of other traditional antitrust concepts like market power, may cut against the argument that consumer-welfare arguments would be cognizable under the "materially harm competition" standard.⁵⁶¹

The DCPCA appears to be more explicit about this issue. That legislation would make it presumptively unlawful for covered platforms to preference their own products and services, "regardless of any alleged procompetitive benefits or efficiencies."⁵⁶² Defendants could rebut an allegation of unlawful self-preferencing only by establishing by clear and convincing evidence that their conduct "did not result in any harm to the relevant aggrieved party."⁵⁶³

Tying

The AICOA, OAMA, and DCPCA also contain provisions prohibiting tying in certain circumstances.

The AICOA would bar covered platform operators from tying access to or preferred placement on their platforms to the purchase or use of other products or services that are not "part of or intrinsic to" a platform.⁵⁶⁴ The legislation would offer an affirmative defense, however, for conduct that does not result in "material harm to competition."⁵⁶⁵

The OAMA included a narrower tying provision. The bill would have prohibited covered firms from conditioning access to their app stores on the use of their payment processors for in-app transactions.⁵⁶⁶

The DCPCA's tying provision is the broadest of the three bills. It would presumptively prohibit covered platforms from engaging in tying, while offering platforms the ability to rebut an alleged violation with clear and convincing evidence that their conduct "did not result in any harm to the relevant aggrieved party."⁵⁶⁷

All three bills go beyond current tying law. As discussed, the modified per se rule against tying allows a plaintiff to prevail by showing that

• 1. The defendant offered two distinct products;
• 2. The defendant conditioned the sale of one product (the tying product) on the purchase of the other product (the tied product);
• 3. The defendant possessed sufficient economic power in the tying product market to coerce purchasers into acceptance of the tied product; and
• 4. The defendant's conduct affected a "not insubstantial" amount of interstate commerce in the tied product.⁵⁶⁸

Additionally, some courts have required plaintiffs to demonstrate that a tying arrangement had anticompetitive effects in the tied product market,⁵⁶⁹ while others have entertained business justifications for challenged ties.⁵⁷⁰ Two cases—the Ninth Circuit's recent decision in Epic Games v. Apple and the D.C. Circuit's 2001 Microsoft decision—have also held that the rule of reason, rather than the modified per se rule, applies to ties involving platform software.⁵⁷¹

In addition to ties involving app stores and payment processors, the tying provisions in the AICOA and DCPCA may implicate several of the practices discussed earlier in this report, including some of Google's conduct in ad-tech markets and the link between favorable placement on Amazon's marketplace and use of Amazon's fulfillment services.⁵⁷²

Interoperability and Data Access

Other proposals involve the ability of business users to interoperate with and access data they generate on covered platforms.

The AICOA includes an interoperability provision that would prohibit covered platform operators from restricting or impeding the ability of business users to interoperate with features that are available to the operator's own products or services, except where such access would lead to a "significant cybersecurity risk."⁵⁷³ Among other conduct, the prohibition may be directed at Facebook's alleged refusal to allow certain app developers to access Facebook Platform and Apple's alleged refusal to allow developers to access some APIs and device functionalities that are available to Apple's apps.⁵⁷⁴

The AICOA's data-access provision would prohibit covered platform operators from restricting or impeding a business user from accessing or transferring data generated by the user's activities on a covered platform.⁵⁷⁵

The OAMA also contained interoperability requirements. The bill would have required covered companies to allow users of their operating systems to install third-party apps and app stores through means other than the covered companies' app stores.⁵⁷⁶ It also would have mandated that covered firms provide developers with access to operating-system interfaces, development information, and hardware and software features on terms that are functionally equivalent to those that covered firms offer to their own apps.⁵⁷⁷

The DCPCA contains both interoperability and data-portability mandates for covered platforms. The bill's interoperability provision would require covered platform operators to

• allow business users to interoperate with hardware and software features that are available to services on a covered platform provided by the platform operator;
• provide business users with continuous and real-time access to data generated by the use of certain platform services; and
• provide business users with the necessary tools to access and analyze data on a covered platform without a transfer from the platform.⁵⁷⁸

The DCPCA's data-portability provision would require covered platform operators to maintain a set of interfaces that provide users with "effective portability" of the data they generate using certain platform services.⁵⁷⁹

Other, potentially broader interoperability and data-portability legislation is discussed in greater detail in "Interoperability & Data Portability" infra.

Use of Nonpublic User Data

As discussed, some Big Tech firms have been accused of using their access to user data to identify and imitate popular offerings. Amazon, for example, has allegedly used nonpublic data to find profitable opportunities for its own private-label business.⁵⁸⁰ Apple has similarly been accused of using competitively sensitive information to replicate fast-growing apps and integrate certain functionalities into iOS.⁵⁸¹

Some proposals would prohibit this conduct. The AICOA includes a provision that would bar covered companies from using nonpublic data from dependent businesses to support their own offerings.⁵⁸² The OAMA included a similar prohibition.⁵⁸³

These types of measures have been debated. Some commentators have argued that a platform operator's imitation of rival products typically increases static efficiency by stimulating competition and lowering prices.⁵⁸⁴ Others have argued that a ban on the use of nonpublic data by platform operators would boost dynamic efficiency by protecting the incentives of other businesses to innovate.⁵⁸⁵

Most-Favored-Nation Policies

Another general category of proposals involves platform restrictions on the activities of business users in other transaction venues.

In the 117th Congress, the reported House version of the AICOA would have prohibited covered platform operators from restricting a business user's pricing of its products or services or its communications on a covered platform regarding other transaction options.⁵⁸⁶

The DCPCA contains a similar prohibition. The bill would make it presumptively unlawful for a covered platform to interfere with or restrict a business user's pricing of its products or services, whether or not those products or services are offered on the platform.⁵⁸⁷

Some of the pricing restrictions targeted by these bills include most-favored-nation clauses (MFNs), which prohibit a platform's business users from offering lower prices on rival platforms.⁵⁸⁸ Platform MFNs may make it difficult for rivals to compete with a dominant platform by charging lower commissions, because such clauses prevent business users from passing along those savings to consumers.⁵⁸⁹

The primary procompetitive benefit proffered in defense of MFNs involves concerns about free-riding. The basic worry is that, absent an MFN, consumers will use a highly functional platform to search for and compare products, but then make their purchases on a different low-cost platform.⁵⁹⁰ Under those conditions, platforms may lack incentives to invest in expensive site features like an attractive design or effective comparison tools, even though those features benefit consumers.⁵⁹¹

The literature has distinguished between "narrow" platform MFNs (which restrict a seller's prices only on the seller's own website) and "wide" platform MFNs (which restrict a seller's prices on all other platforms).⁵⁹² Some theoretical analyses have concluded that narrow MFNs are more likely to be procompetitive than wide MFNs.⁵⁹³

Both narrow and wide platform MFNs would fall within the scope of the DCPCA and the reported House version of the AICOA in the 117th Congress. Under the relevant version of the AICOA, challenged restrictions would have escaped liability if a platform operator established by clear and convincing evidence that its conduct would (1) not harm "the competitive process by restricting or impeding legitimate activity by business users," or (2) increase "consumer welfare."⁵⁹⁴ Under the DCPCA, by contrast, covered platforms would need to show by clear and convincing evidence that their conduct "did not result in any harm to the relevant aggrieved party."⁵⁹⁵

App Preinstallation

The AICOA also includes a provision prohibiting covered firms from restricting or impeding the uninstallation of preinstalled apps, unless such conduct is necessary for the security or functioning of the platform or to prevent data from being transferred to a foreign adversary.⁵⁹⁶ The OAMA contained a similar prohibition.⁵⁹⁷

These prohibitions appear to be directed at concerns that Google and Apple have leveraged control of their mobile operating systems to favor their own apps and app stores.⁵⁹⁸ Though the bills do not explicitly prohibit the preinstallation of a covered firm's proprietary apps, commentators have debated whether such preinstallation would run afoul of the AICOA's general self-preferencing prohibition.⁵⁹⁹

Structural Separation and Line-of-Business Restrictions

Several of the proposals discussed above respond to concerns that Big Tech firms face conflicts of interest when they operate both a digital platform and vertically related businesses that compete with platform users. The proposals seek to address those concerns by prohibiting specific categories of allegedly problematic conduct. One possible downside of this approach involves enforcement costs.

The worry is twofold. First, conduct rules require regulators to continuously monitor the behavior of covered firms.⁶⁰⁰ Second, the availability of affirmative defenses means that rule enforcement may entail some of the same issues of cost and timeliness that have led to dissatisfaction with the existing antitrust framework.⁶⁰¹

Based on these potential difficulties, some commentators have argued that structural restrictions have important advantages over behavioral rules.⁶⁰² Such restrictions can take two general forms. Structural regulation could involve total separation, meaning firms would be prohibited from owning both a covered platform and a business that operates on that platform.⁶⁰³ Alternatively, regulations could mandate partial or functional separation, whereby firms would be required to house a covered platform and vertically related businesses in separate legal entities.⁶⁰⁴

There is precedent for these types of structural regulations, including in the railroad, banking, and telecommunications industries.⁶⁰⁵

The DCPCA would impose a separation requirement on certain tech platforms. The legislation would make it unlawful for a covered platform operator to "maintain, or engage in any action that creates, a platform conflict of interest."⁶⁰⁶ It would empower a newly created agency to order divestitures and other necessary remedies to eliminate prohibited conflicts of interest.⁶⁰⁷ The new agency would also have the authority to engage in rulemaking to implement this prohibition.⁶⁰⁸

In the 117th Congress, H.R. 3825, the Ending Platform Monopolies Act, contained a more detailed separation requirement.⁶⁰⁹ The bill would have prohibited covered platform operators from owning, controlling, or having a beneficial interest in a "line of business" that

• utilizes the covered platform for the sale or provision of products or services;
• offers a product or service that the covered platform requires business users to purchase or utilize as a condition of accessing or receiving preferred placement on the platform; or
• gives rise to a "conflict of interest."⁶¹⁰

The bill would have provided that "conflicts of interest" arise when a platform operator's ownership or control of another "line of business" creates the incentive and ability for its platform to (1) advantage the platform operator's products or services over those of competitors, or (2) exclude or disadvantage the products or services of competitors.⁶¹¹

These types of proposals have generated debate. Critics of separation requirements have argued that a platform's entry into new markets typically benefits consumers.⁶¹² For example, by selling its own private-label products on its marketplace, Amazon may offer consumers low-cost alternatives to established brands.⁶¹³ Integration into related business lines may also create efficiencies.⁶¹⁴ Apple and Google, for instance, may be well-positioned to produce apps and app stores for their respective operating systems, as well as related devices like earphones and smart watches.⁶¹⁵

Separation requirements may also face line-drawing difficulties. The boundary between a covered platform and separate services is not always clear.⁶¹⁶ For example, Apple produces many apps and functionalities—including a voice assistant (Siri), a camera app, and a payment system (Apple Pay)—that are integrated with its iOS operating system to various degrees.⁶¹⁷ Whether these services would qualify as "lines of business" that are distinct from iOS may be uncertain; H.R. 3825 did not define that term. Because tech platforms regularly add new functionalities to their primary services, some observers have argued that an absence of clarity surrounding permissible activities may deter innovation and thereby harm consumers.⁶¹⁸

Proponents of separation requirements have acknowledged these criticisms. In response, they have argued that the innovation benefits of an equal playing field would likely outweigh any losses in static efficiency that result from the elimination of a platform operator's downward pricing pressure in adjacent markets.⁶¹⁹ In addition, advocates of separation rules contend that any decreases in platform innovation caused by such rules must be weighed against likely increases in innovation by platform users.⁶²⁰

Mergers & Acquisitions

Substantive Merger Law

Other proposals target mergers involving large tech platforms. The DCPCA would create a new agency charged with reviewing certain mergers that involve covered platform operators.⁶²¹ For transactions subject to the HSR Act's pre-merger filing requirement, the bill would require covered platform operators to submit their filings to the new agency, in addition to the antitrust agencies. To proceed with such transactions, covered platform operators would need to prove by clear and convincing evidence that the transactions would "serve the public interest."⁶²²

The bill lists 13 factors that are relevant to this inquiry, including a transaction's effects on competition, national security, privacy, and local communities.⁶²³ It also identifies certain fact patterns in which a covered transaction "would not" serve the public interest, in addition to fact patterns that "may" support a determination that a transaction would not serve the public interest. The bill prescribes that a covered transaction "would not" serve the public interest if

• the acquired firm offers services or products that overlap or compete with, or that are functionally equivalent to, those offered by the covered platform operator;
• the acquired firm is a "critical trading partner in the supply chains or business ecosystems of the parties"; or
• the transaction would create a "platform conflict of interest."⁶²⁴

The legislation identifies the following facts that "may" support a determination that a covered transaction would not serve the public interest:

• the transaction would result in a post-transaction market share of greater than 33% in any relevant market (including labor markets);
• the transaction would (1) result in a Herfindahl-Hirschman Index (HHI) greater than 1,800 in any relevant market, and (2) increase the HHI by more than 100;⁶²⁵ or
• the transaction would result in an aggregation of data or access to data that "harms the competitive process or creates or helps maintain a monopoly, a monopsony, market power, or unfair methods of competition."⁶²⁶

In addition to requiring covered platform operators to prove by clear and convincing evidence that HSR-reportable mergers would serve the public interest, the DCPCA would empower the newly created platform regulator to block non-reportable transactions involving covered platform operators upon a determination that they would not serve the public interest.⁶²⁷

The DCPCA would also give the new regulator the authority to review consummated mergers involving covered platform operators and unwind those transactions upon a determination that they "materially harmed the public interest."⁶²⁸ The bill would allow the regulator to evaluate whether a consummated transaction "materially harmed the public interest" by evaluating the factors identified above, while also providing that such harm is present if

• the covered platform operator acquired a "critical trading partner";
• the transaction resulted in a post-transaction market share of greater than 50% in any relevant market (including labor markets); or
• the transaction (1) resulted in an HHI greater than 2,500 in any relevant market, and (2) increased the HHI by more than 200.⁶²⁹

The DCPCA thus envisions a broad overhaul of the legal regime governing mergers involving dominant platforms. In place of the existing framework in which courts inquire primarily into a transaction's economic effects, the bill would adopt per se rules against platform mergers that fall into certain categories and empower a new specialist regulator to block platform mergers under a public-interest standard.

Legislation in the 117th Congress likewise would have adopted certain per se rules for mergers involving dominant platforms, but would not have established a public-interest standard for those mergers or created a new regulator. H.R. 3826, the Platform Competition and Opportunity Act, would have prohibited covered platform operators from acquiring other firms unless they could demonstrate by clear and convincing evidence that a target does not

• compete with the platform operator;
• constitute "nascent or potential competition" for the platform operator;
• enhance or increase the platform operator's market position with respect to products or services offered on or directly related to a covered platform; or
• enhance or increase the platform operator's ability to maintain its market position with respect to products or services offered on or directly related to a covered platform.⁶³⁰

The reported version of the bill included an amendment that would have exempted transactions of less than $50 million from the prohibition.⁶³¹

For transactions of $50 million or greater, then, the bill would have prohibited covered platforms from engaging in horizontal mergers; mergers involving "nascent or potential" competitors; and vertical and conglomerate mergers that enhance, increase, or help maintain their market positions with respect to products or services "offered on or directly related" to a covered platform.

As drafted, the bill raised three issues. The first involved the legislation's prohibition of acquisitions involving "potential" competitors.⁶³² As discussed, antitrust doctrine has recognized two theories of harm in potential-competition cases: the elimination of perceived potential competition and the elimination of actual potential competition.⁶³³ Courts have identified prerequisites for both theories.⁶³⁴

The relationship between those prerequisites and H.R. 3826's requirement that a covered platform show that a target firm is not a "potential" competitor may have generated complex legal questions if the bill had become law. For example, the bill could have been read to allow platform operators to make such a showing by negating an element of both types of potential-competition claims.⁶³⁵ However, that is not the only interpretive option; the details surrounding the relevant burden would have had to be fleshed out in practice. That the Supreme Court has recognized only the perceived-potential-competition theory might have complicated this inquiry.

The second issue concerned the bill's prohibition of Big Tech acquisitions involving "nascent" competitors.⁶³⁶ Commentators have offered different definitions of the concept of nascent competition.⁶³⁷ In general, however, the term has been used to refer to firms and technologies with uncertain prospects that nevertheless pose serious threats to an incumbent.⁶³⁸

Despite posing such threats, acquisitions of nascent competitors may be difficult to challenge under existing law. As discussed, to prevail under an actual-potential-competition theory, a plaintiff must establish a "substantial likelihood" that a target firm would deconcentrate the relevant market or produce other procompetitive benefits.⁶³⁹ In cases involving unproven or developing technology, that burden could prove problematic for a plaintiff. H.R. 3826 was a response to this doctrinal difficulty.⁶⁴⁰

While the bill thus sought to address an issue that has generated considerable attention, the analytical framework that would govern inquiries into nascent competition remains unsettled. There is little case law addressing issues of nascent competition in the merger context.⁶⁴¹ Accordingly, H.R. 3826 would have leaned on the courts to develop standards for evaluating whether a firm constituted a nascent competitor of a covered platform.

The third issue raised by H.R. 3826 involved the breadth of the provisions prohibiting mergers that "enhance or increase" a platform operator's market position or ability to maintain its market position.⁶⁴²

By their terms, these prohibitions did not distinguish between procompetitive mergers and anticompetitive mergers. As drafted, the bill thus appeared to prohibit mergers that "enhance or increase" a covered platform's market position by improving the quality of its products or services, even when the target company is not a competitor, potential competitor, or nascent competitor of the platform. As a result, H.R. 3826 may have limited Big Tech platforms to in-house development or licensing of complementary technologies; acquisitions of firms that could enhance a platform's core offerings would have likely been off-limits.

S. 1074—another bill in the 117th Congress—would have taken a similarly strict approach toward Big Tech mergers.⁶⁴³ Among other things, S. 1074 would have prohibited companies designated as "dominant digital firms" from engaging in acquisitions valued at more than $1 million.⁶⁴⁴

Other proposals are more limited. Several commentators, for example, have advocated a requirement that Big Tech firms bear the burden of proving that their mergers would not harm competition.⁶⁴⁵

Abstracting from specific policy options, the debate over special merger rules for Big Tech firms has focused on two general concerns.

First, opponents of such rules have argued that Big Tech mergers are typically benign or procompetitive.⁶⁴⁶ Acquisitions of complementary technologies, for example, may reduce the transaction costs associated with licensing arrangements or allow for more efficient integration with a platform's offerings.⁶⁴⁷ Mergers may also stimulate competition among Big Tech firms by giving them an attractive means of entering or expanding within each other's core markets.⁶⁴⁸

Second, some have argued that limitations on Big Tech mergers may reduce startup investment by eliminating a popular exit route for venture investors and other entrepreneurs.⁶⁴⁹

Proponents of special merger rules for tech platforms have responded that the procompetitive benefits of tech mergers are often overstated.⁶⁵⁰ Merger limitations targeting a handful of prospective acquirers may also leave startup investors with enough viable exit options to mitigate concerns about dampened investment. Additionally, some commentators have suggested that reducing investment in innovations that end up in the hands of dominant incumbents is the intended outcome of the relevant proposals.⁶⁵¹

The Merger Review Process

Before moving on from mergers, one final topic warrants mention: the HSR pre-merger review process. As discussed, under the HSR Act, parties to mergers that exceed certain thresholds must report their transactions to the DOJ and FTC and abide by specified waiting periods before closing.⁶⁵² This process gives the agencies the opportunity to review proposed mergers for antitrust concerns and seek relief before deals are consummated.

Some commentators have expressed concerns about the number of Big Tech mergers that fall below the relevant thresholds and thus avoid HSR review.⁶⁵³ In September 2021, the FTC released a report indicating that the four Big Tech firms discussed in this report and Microsoft together engaged in 819 non-reportable deals between 2010 and 2019.⁶⁵⁴

In response to worries about these transactions, some have supported a blanket HSR filing requirement for acquisitions by dominant tech platforms.⁶⁵⁵ Opponents of such a rule have argued that it would be burdensome and offer few benefits for regulators.⁶⁵⁶

Interoperability & Data Portability

Network effects and switching costs are frequent themes in discussions of Big Tech.⁶⁵⁷ Some reform proposals seek to address these structural features of certain platform markets by imposing interoperability and data-portability obligations on designated-platform operators.⁶⁵⁸

Interoperability refers to the ability of distinct services to work together and communicate with one another.⁶⁵⁹ Interoperability can develop organically—as with email—or as a result of a legal mandate.⁶⁶⁰ Examples in the latter category include the 1996 Telecommunications Act's requirement that local exchange carriers interconnect with other providers.⁶⁶¹ The DOJ's 2002 monopolization settlement with Microsoft also included an interoperability provision prohibiting Microsoft from excluding other firms' web browsers from its Windows operating system.⁶⁶²

These types of measures seek to lower the entry barriers associated with networked industries by shifting network effects from individual firms to the market as a whole, making them available to nascent and potential competitors of a dominant incumbent.⁶⁶³

Data portability, by contrast, refers to a consumer's right to move his or her data from one platform to another.⁶⁶⁴ Telecommunications law again offers an example by granting phone users the right to retain their phone numbers when they change carriers.⁶⁶⁵ Such requirements decrease the switching costs that might otherwise discourage consumers from taking their business to a more attractive provider.⁶⁶⁶

In the 117th Congress, H.R. 3849 would have imposed interoperability and data-portability duties on designated digital platforms.⁶⁶⁷ The bill would have directed the FTC to develop standards implementing those duties for individual covered platforms.⁶⁶⁸ In promulgating standards under the legislation, the FTC would have been advised by technical committees that included representatives of a platform's competitors, competition and privacy-advocacy organizations, the National Institute of Standards and Technology, and covered platforms.⁶⁶⁹

The obligations contemplated by H.R. 3849 were potentially broader than those in the AICOA, which were discussed earlier in this report.⁶⁷⁰ The AICOA's interoperability provision would prohibit a covered platform operator from restricting the ability of business users to interoperate with features that are available to the operator's own products or services.⁶⁷¹ Accordingly, the prohibition would have been limited to a platform operator's unequal treatment of firms that utilize its platform.⁶⁷²

In contrast, H.R. 3849 would have granted the FTC rulemaking authority to impose potentially broader, platform-specific interoperability obligations.⁶⁷³ For a social network like Facebook, an interoperability rule might have included duties to allow users of other networks to "friend" Facebook users and transmit posted content from Facebook to other networks.⁶⁷⁴ Supporters of interoperability have argued that these types of obligations would catalyze competition by allowing users of upstart social networks to benefit from Facebook's scale.⁶⁷⁵

H.R. 3849's data-portability provision was also potentially broader than the parallel requirement in the AICOA. While the AICOA's requirement would apply only to a platform's business users,⁶⁷⁶ H.R. 3849's data-portability obligation would have encompassed individuals who use a covered platform.⁶⁷⁷

A rule implementing this duty might have required a social network like Facebook to keep a user's messages, photos, and other content in an accessible format that could be transferred to other platforms.⁶⁷⁸ Although this type of requirement may have partially overlapped with the ongoing transferability contemplated by H.R. 3849's interoperability mandate, it could also have included categories of data not subject to continuous real-time interoperability for technical or other reasons.⁶⁷⁹ Data-portability rules may likewise require Amazon to allow retailers on its marketplace to port their customer reviews to rival e-commerce platforms and Apple to permit iPhone users to transfer their message histories to an Android device.⁶⁸⁰

Objections to interoperability and data-portability mandates take several forms. Some have highlighted the complexity of interoperability requirements, which may pose challenges of implementation and enforcement.⁶⁸¹ Others have focused on possible privacy and data-security risks that might accompany both interoperability and data-portability rules.⁶⁸²

H.R. 3849 attempted to address complexity concerns by directing the FTC to establish technical committees to assist with rule development.⁶⁸³ The bill sought to mitigate privacy and data-security risks by imposing data-security requirements on firms that interoperate with or receive ported data from a covered platform.⁶⁸⁴

Another category of criticism directed at interoperability requirements involves innovation concerns. Some have worried that interoperability may result in homogenized markets as an incumbent's rivals coalesce around a single set of standards.⁶⁸⁵ Compelled interoperability also potentially implicates the free-rider problems that motivate narrow duty-to-deal doctrine: by requiring firms to share the fruits of their innovation with competitors, policymakers may dampen incentives to invest in new products.⁶⁸⁶ Defenders of interoperability have acknowledged this risk, but maintain that interoperating Big Tech platforms would still face incentives to innovate to prevent rivals from gaining a competitive edge.⁶⁸⁷

Changes to General Antitrust

While the proposals discussed above would entail special competition rules for large tech platforms, other options involve changes to general antitrust law. Because general antitrust reform is a vast topic, this report does not attempt an exhaustive overview of the relevant proposals. Instead, it briefly reviews selected bills involving exclusionary conduct and merger law.

Exclusionary Conduct

S. 225, the Competition and Antitrust Law Enforcement Reform Act (117^th Cong.)

In the 117th Congress, S. 225 would have made several changes to the legal framework governing exclusionary-conduct claims.⁶⁸⁸ The bill would have amended the Clayton Act to prohibit "exclusionary conduct that presents an appreciable risk of harming competition."⁶⁸⁹ "Exclusionary conduct" would have been defined to mean conduct that (1) "materially disadvantages" an actual or potential competitor, or (2) "tends to foreclose or limit" the ability of an actual or potential competitor to compete.⁶⁹⁰

S. 225 would have adopted a presumption that exclusionary conduct presents "an appreciable risk of harming competition" if it is undertaken by a firm with a market share of greater than 50% or that otherwise has "significant market power" in the relevant market.⁶⁹¹ That presumption could be rebutted, however, if a defendant established by a preponderance of the evidence that

• 1. "distinct procompetitive benefits of the exclusionary conduct in the relevant market eliminate the risk of harming competition presented by the exclusionary conduct";
• 2. another firm has "entered or expanded their presence in the market with the effect of eliminating the risk of harming competition posed by the exclusionary conduct"; or
• 3. "the exclusionary conduct does not present an appreciable risk of harming competition."⁶⁹²

The bill would have provided that several of the conduct-specific tests that courts have adopted in Sherman Act cases would not apply to exclusionary-conduct claims under the amended Clayton Act. Among other things, exclusionary-conduct plaintiffs would not have to show

• that a defendant terminated a prior course of dealing,⁶⁹³ which some courts have held is a prerequisite for refusal-to-deal liability under the Sherman Act;⁶⁹⁴
• that the defendant priced its products below its costs or is likely to recoup losses from below-cost pricing,⁶⁹⁵ which are both requirements for predatory-pricing claims under the Sherman Act;⁶⁹⁶ or
• that the conduct of a multi-sided platform presents an appreciable risk of harming competition on more than one side of the platform,⁶⁹⁷ contrary to the rule the Supreme Court adopted for two-sided transaction platforms in Ohio v. American Express.⁶⁹⁸

S. 225 also would have provided that market definition is not necessary to prove an antitrust violation, except in cases where the applicable statute includes the phrase "relevant market," "market concentration," or "market share."⁶⁹⁹

S. 1074, the Trust-Busting for the Twenty-First Century Act (117th Cong.)

S. 1074—another bill in the 117th Congress—also would have made changes to the standards governing exclusionary-conduct claims.⁷⁰⁰ The legislation would have provided that, in litigation under Section 1 or Section 2 of the Sherman Act, a defendant that relies upon procompetitive effects to justify its conduct must establish by clear and convincing evidence that

• 1. the relevant procompetitive effects "clearly outweigh" any anticompetitive effects; and
• 2. the defendant "could not obtain substantially similar procompetitive effects through commercially reasonable alternatives that would involve materially lower competitive risks."⁷⁰¹

Like S. 225, the bill would have provided that market definition is not required to prove a violation of Section 1 or Section 2.⁷⁰²

Mergers

S. 225, the Competition and Antitrust Law Enforcement Reform Act (117th Cong.)

In addition to the exclusionary-conduct provisions discussed above, S. 225 would have modified several aspects of merger law. The bill would have amended Section 7 of the Clayton Act to prohibit mergers that "create an appreciable risk of materially lessening" competition⁷⁰³—a change from the current language that prohibits mergers that may "substantially" lessen competition.⁷⁰⁴ The term "materially" was defined to mean "more than a de minimis amount."⁷⁰⁵

S. 225 also would have shifted the relevant burden of proof to the merging parties in certain circumstances.⁷⁰⁶ For example, merging parties would have had the burden of proving that their transactions would not "create an appreciable risk of materially lessening" competition in cases where

• a merger would lead to a "significant increase in market concentration";
• a firm with a market share greater than 50% or that possesses "significant market power" acquires a competitor or a company that has a "reasonable probability" of becoming a competitor;
• a transaction is valued at more than $5 billion; or
• the acquiring firm has assets, net revenue, or a market capitalization exceeding $100 billion and the transaction is valued at $50 million or more.⁷⁰⁷

S. 1074, the Trust-Busting for the Twenty-First Century Act (117th Cong.)

S. 1074 also included merger restrictions.⁷⁰⁸ The bill would have prohibited firms with market capitalizations exceeding $100 billion from engaging in mergers whose effect "may be to lessen competition in any way."⁷⁰⁹ It also would have explicitly provided that market definition is not necessary to block a merger and that mergers shall not be presumed to be legal on the grounds that the parties are not direct competitors.⁷¹⁰

S. 3847, the Prohibiting Anticompetitive Mergers Act (117th Cong.)

In the 117th Congress, S. 3847 would have taken a similarly skeptical approach to large mergers. The legislation would have prohibited mergers valued at more than $5 billion, mergers that result in a market share of over 33% for sellers or 25% for employers, and mergers that would result in specified levels of market concentration.⁷¹¹

S. 3847 also would have made changes to the merger-review process.⁷¹² Among other things, the bill would have extended the initial HSR waiting period from 30 days to 120 days and allowed the antitrust agencies to block mergers without obtaining a court order.⁷¹³

In addition, the bill would have directed the DOJ and FTC to review mergers consummated after January 1, 2000, if they would have qualified as "prohibited mergers" under the categories mentioned above.⁷¹⁴ It would have further required the agencies to pursue remedies to restore competition or address the anticompetitive effects of those mergers in specified circumstances.⁷¹⁵