Antitrust in the Digital Age: A Tale of Two Agencies
During the past few years, antitrust law has gone from an academic concern discussed primarily in the corridors of law schools to a marquee issue in today’s political debate. Driving this shift are the recent enforcement actions against Big Tech. Silicon Valley giants have caught the ire of our primary antitrust enforcers, including the Department of Justice (DOJ), state attorneys general, and the Federal Trade Commission (FTC).
These cases make sense. Tech markets are becoming increasingly concentrated, and the major platform companies are at the center of it. For instance, Google has cornered nearly 90 percent of the search engine and ad tech market, and Apple almost completely dominates the mobile smartphone and app distribution markets. These developments have the DOJ and a slew of state attorney general offices up in arms.
Meta, Microsoft, and Amazon, too, have had some run-ins with antitrust enforcers, specifically the FTC. The FTC has sought to block Meta’s merger with Within Unlimited—a virtual reality company—and Microsoft’s acquisition of gaming company Activision. The agency also has an anti-monopolization case against Amazon regarding its online retail platform.
Although the DOJ and the FTC seem to be addressing the same issue—growing concentration in tech markets—they have taken two very different approaches. The DOJ leverages more traditional and straightforward legal theories—such as a focus on consumer welfare, market efficiencies, and price—to make its cases against Google and Apple. The FTC, by contrast, is challenging those very theories; it has taken its position among the avant-garde of legal theorists who aim to refocus antitrust enforcement to promote competitors.
With the fate of smaller innovators and the advent of artificial intelligence on the line, the question becomes: which approach seems to be working to quell the extreme market concentration in the tech sector without disincentivizing new technologies? Asked another way, is the DOJ’s more Borkian method a practical path forward to dislodge tech chokepoints, or does that task require a Brandeisian hammer?
Bork versus Brandeis
Antitrust law is the primary tool in dealing with market concentration and anticompetitive behavior. In effect, enforcement done right can open the door for new, innovative companies seeking to enter the market. There are, however, two leading schools of thought that seem to be at war with each other: the consumer welfare standard and the neo-Brandeisian approach.
Before delving into the merits of each method, it’s important to understand the basic theories at play. The consumer welfare standard derives from Judge Robert Bork’s book The Antitrust Paradox (1978), an approach courts have been utilizing since the late 1970s. It requires courts to examine whether consumers benefit from a firm’s action, even if it means they lose a competitor as a result.
In general, the consumer welfare standard takes into account prices, of course, but also considers other factors that are harder to measure. As the D.C. Circuit found in the recent AT&T–Time Warner case, the standard extends “beyond higher prices for consumers, including decreased product quality and reduced innovation.” Or, as former FTC Commissioner Christine Wilson explained, the standard considers the effect of “competition on quality . . . in the analysis of vertical restraints” and requires an evaluation of the “effects on innovation.”
The reason why Bork’s consumer welfare standard was so revolutionary was because it applied economic theory as a limiting principle to judicial decisions in rule of reason cases—violations that are not explicitly restricted by antitrust statutes (i.e., per se violations), like price fixing, bid-rigging, horizontal customer allocation, and territorial allocation agreements. In practice, this standard had legal scholars and courts evaluating the market’s efficiencies rather than the size of the overall firm. Hence, the analysis was an economics-intensive enterprise as opposed to evaluating other factors, such as the political or societal implications of any particular firm’s action or size.
Under a consumer welfare standard, competition’s value must be viewed in the context of providing better outcomes for consumers. The consumer welfare standard forces a court to evaluate the following: Which is preferable to the consumer, one company that provides low prices and high-quality services? Or two companies that provide the same low price but offer lower quality services? In this example, both scenarios may yield lower prices, but the consumer welfare standard would say that having the one firm is preferable because it provides a better quality product to the consumer.
The Supreme Court under Warren Burger found this view persuasive. In Brunswick Corp. v. Pueblo Bowl-Mat, Inc., the Court stated affirmatively that “antitrust laws . . . were enacted for ‘the protection of competition, not competitors.’” This position views the legislative history of antitrust statutes as not necessarily concerned with the size of a particular firm or even promoting competition for competition’s sake. Instead, Bork’s understanding of antitrust law was that antitrust enforcement only makes sense if it is to address market behaviors that actually harm consumers.
Bork’s influence continued to surface in cases for the next couple of decades. In Reiter v. Sonotone Corp., the Supreme Court cited The Antitrust Paradox when claiming that the Sherman Act is a “consumer welfare prescription.” In Continental T.V., Inc. v. GTE Sylvania, Inc., the Supreme Court relied heavily on economic analysis when evaluating rule of reason cases. These cases are why economic analysis is now a guardrail for an otherwise unbound rule of reason test.
Nevertheless, how courts administer the consumer welfare standard is far from perfect, especially when addressing tech markets. Although the consumer welfare standard considers many factors, courts too often treat consumer price differences as a preeminent indication of monopolization or competitive harm. More libertarian-leaning scholars have argued the same to ensure that no antitrust enforcement occurs in tech markets. These views have led to an overall laxity in antitrust enforcement, because non-price harms, such as the ones levied against Google’s Search or Apple’s anticompetitive actions in its App Store, are difficult to quantify.
This has led some academics, including the current FTC chair, Lina Khan, to revive a Brandeisian framework. This neo-Brandeisian framework is far more interested in reverting to the trust-busting days of the early twentieth century. Under this framework, the number of firms in the market, and the size of the largest, are by far the most important factors. To put it succinctly, the neo-Brandeisian view is that a large company is always a threat to competition, and the only true remedy is more competitors. Contrast that with Bork’s view, in which a firm’s size does not matter unless it’s acting in a way that harms consumers.
To be clear, the neo-Brandeisian movement is not some new approach to antitrust analysis. Far from it. It’s a revival of how antitrust was often used in its early days in the United States. Neo-Brandeisians often point to the words of Senator John Sherman, the sponsor of the Sherman Act, who explained that the Act’s underlying principle was that “[m]onoplies . . . [are] inconsistent with our form of government. . . . If we will not endure a king as a political power, we should not endure a king over the production, transportation, and sale of any of the necessaries of life.” In no uncertain terms, Senator Sherman argued that the Act represented an American policy that holds “[i]f we will not submit to an emperor, we should not submit to an autocrat of trade. . . .”
The neo-Brandeisian approach has its own drawbacks, however. If big is bad and more competitors are better, where’s the limit? Is there a limit? And if so, is there an objective standard against which that limit should be measured? Without answers to those questions, the neo-Brandeisian approach has created a lot of confusion related to what companies could merge, what constitutes an unfair competitive advantage, and what competitors the law is meant to protect.
Today, we are witnessing these divergent views come to a head when addressing Big Tech. On the one hand, some progressives have argued for a neo-Brandeisian approach that focuses almost exclusively on helping the “little guy.” On the other hand, Big Tech has repeatedly argued that price is the only factor that matters—meaning that “free to the consumer” services are effectively immune from antitrust scrutiny.
Big Tech not only attempts to confine antitrust to a focus on consumer prices, but these companies also litigate heavily around the definition of the market to game the litigation in their favor. In Epic v. Apple, for example, Apple succeeded in narrowly defining its relevant market based on the specific app transactions. In this case, the court found that Apple did not have market dominance in mobile gaming transactions as opposed to its dominance as an app store platform generally. This extremely narrow definition of the app market makes it nearly impossible for a court to find that Apple has significant market dominance over any type of app, which is necessary to find antitrust harm.
Yet that’s not true for all courts, and this ruling appears to be an isolated occurrence. For instance, the Court of Appeals for the D.C. Circuit rejected Big Tech’s narrow view in a direct precedent for the DOJ’s latest challenges against Google and Apple, Microsoft v. U.S.
In the context of app stores, a federal jury recently sided unanimously with Epic in its case against Google for its market dominance of the Play Store. Specifically, Epic was able to show the jury that Google’s contracts with app developers and phone manufactures ensured that Epic could not escape its 30 percent app-store tax.
This was the exact opposite result of that in Epic v. Apple. How did this happen? It goes back to market definition. In the Apple case, the judge decided that the market was “video game transactions”—a definition that makes an antitrust finding near impossible. In the Google case, the jury defined the market as all app transactions on Android. And the jury considered more evidence on how Google’s “secret revenue sharing deals” with mobile phone manufacturers and competing game platforms contributed to edging out Epic’s gaming store as a real competitor. This framing also gave the jury a clearer picture as to how Google’s abuses impacted innovation within the mobile ecosystem as a whole.
Still, the divergence between Epic’s Apple and Google cases is hard to square because these cases offer little guidance on how to define the app store markets. What’s more, Google has filed a motion for a judge to either grant a new trial or overturn the jury decision, which could completely upend this victory.
These precedents are precisely why the DOJ’s approach is needed and may be more likely to win out over the FTC’s. The DOJ is on the verge of filling in a key gap in consumer welfare analysis, specifically adding more clarity to non-priced offenses (e.g., harms to innovation).
DOJ’s Winning Strategy
The DOJ has brought forward three major cases against Big Tech. Two are against Google; one for its monopolization of the search engine market, initiated by the Trump administration, and one for the ad tech marketplace, which was initiated under the Biden administration. The other is against Apple for its monopolization of the mobile app market. These are straightforward, evidence-based cases that have relied on years of investigation into each firm.
In these cases, the DOJ consistently relies on the consumer welfare standard, but the key argument is that these monopolizations harm innovation—a non-price-based harm. As former Assistant Attorney General Makan Delrahim rightly pointed out in a speech, “courts interpreting the Sherman and Clayton Acts have recognized harms to competition in the form of lower output, decreased innovation, and reductions in quality and consumer choice.” He went on to say in the same speech that cases like Microsoft and AT&T–Time Warner provided a path forward for antitrust enforcers to evaluate non-priced harms, such as harms to competition. He explained that the D.C. Circuit in AT&T–Time Warner “recognized that harm to competition extends ‘beyond higher prices for consumers, including decreased product quality and reduced innovation.’” The D.C. Circuit’s “legal analysis will help,” Delrahim concluded, “when [the DOJ] bring[s] [its] next case alleging non-price effects as competitive harm.”
And it looks like the DOJ is putting in the work. If successful in the Google and Apple cases, the result will be far more clarity on non-priced harms and a much-needed update to how we evaluate consumer welfare in the digital age, all without throwing the baby out with the bathwater.
In each of these cases, DOJ is relying on recent precedents, such as Microsoft and AT&T–Time Warner, and traditional legal frameworks to make its case. The DOJ makes strong demonstrations that both Apple and Google have extraordinary market power in their respective markets, which recalls its enforcement against Microsoft in the 1990s.
Take its case against Google concerning the monopolization of search. The DOJ’s case against Google’s tactics mirrors the one the agency made against Microsoft. For instance, Google has arranged for service providers—browsers and device manufacturers—to make its Search the default engine. Google pays Apple over $26.3 billion per year to make Google Search the default search engine for Safari—a direct competitor to Google’s browser, Chrome. Google requires device manufacturers that use its Android operating system to preinstall certain apps that use Google Search as their default search engine.
The DOJ has made a compelling case that Google’s abuse of market power has adversely impacted innovation based on Google’s track record. Google has used its extraordinary market power to kill or substantially slow the development of more efficient search technologies, like vertical search, that compete directly with Google’s tech. When the vertical search company Foundem gained more users, Google changed its Search algorithm to digitally bury Foundem.com by placing it as far as 170 pages down, irrespective of how many people directly searched for that product—even if users searched for the company’s exact name.
The DOJ uses the same framework when making its case against Apple. Much like Microsoft in the browser market, Apple both provides the device and the software that developers and consumers require to access apps and mobile features. Indeed, the DOJ dedicates a significant portion of its brief to illustrating how Apple’s alleged abuses parallel those in the Microsoft case. For instance, the DOJ demonstrates how Apple’s limits on cross-platform applications lock in consumers and deter innovation. The DOJ even makes direct comparisons to show how Apple’s actions in the present day are akin to Microsoft’s degradation of Apple’s QuickTime in the 1990s. This is relevant because both Microsoft and Apple imposed such measures to increase the switching costs for consumers and make it harder for them to move away from their respective products.
Indeed, the DOJ, through multiyear investigations and data, shows that what Apple does today may be far worse than what Microsoft did to provoke antitrust enforcement in the past. The DOJ has identified concrete examples of Apple’s harm to competition and, by extension, innovation. In particular, Apple disallows so-called super apps that provide a user with broad functionality in a single app; thus forcing the use of Apple’s native apps. Apple also blocks out cloud-based gaming to force gamers and game developers to pay its App Store tax. It limits or degrades messaging apps to make switching to Android devices extremely difficult for consumers. And it doesn’t stop there. Apple forces pairing and limits third-party functionality for wearables like smartwatches and prevents third-party digital wallets on its devices.
To put it another way, when you buy an iPhone, it isn’t yours; according to Apple, it’s Apple’s. And Apple—not you—gets to make the decisions on how it can be used and with whose services.
The DOJ’s case makes a clear connection between this anticompetitive behavior and diminished consumer welfare. For instance, when Apple decides to change or enforce its developer guidelines to kill free apps, like Parler or BlueMail, or will not allow a subscriber to download gaming apps or streaming apps from another app store with a more favorable rate, consumers are stuck with lower quality products. Even if you wanted to text Android users with the same encryption as when you use iMessage, you can’t, because Apple has blocked the app—Beeper Mini—that allows you to do that. Apple’s walled-garden approach makes switching costs to Android extremely expensive for the consumer because, if a consumer disagrees with Apple’s policy to deplatform an app, they cannot just replace their phone. They would have to change out their MacBook, iPad, and resubscribe to every app they downloaded.
The DOJ has already seen success in applying this expanded consumer welfare framework to prevent bad mergers. For instance, the DOJ sued to block JetBlue’s $3.8 billion acquisition of Spirit Airlines. For weeks, the DOJ confronted airline executives with documents showing JetBlue and Spirit knew that the merger would raise prices on more than one hundred nonstop and connecting routes where JetBlue and Spirit currently compete for customers. Witness after witness testified that airlines lower their prices when Spirit flies a route. The agency’s work was rewarded as the court agreed with its analysis and blocked the acquisition outright.
The DOJ’s decision to bring these cases shows that some of Big Tech’s abuses can be addressed through tried-and-true legal theories. As Assistant Attorney General Delrahim alluded to in the previously quoted remarks, it appears that non-price frameworks that are difficult to quantify, like cost to innovation, are increasingly being incorporated into traditional consumer welfare analyses.
These cases brought by the DOJ will allow judges to pick up the baton to further articulate the metes and bounds of non-priced harms within a consumer welfare framework. The results will be useful in providing firms with a context on what contractual arrangements will be viewed as an affront to innovation or competition. It will also signal to larger tech firms what guardrails must be in place when contracting with suppliers.
These cases may even outline what restrictions need to be in place for a merger between a Big Tech company and smaller developers to pass muster under antitrust laws. For example, when Microsoft sought to merge with Activision, it clearly learned from its run in with the DOJ in the 1990s. To ensure that it did not run afoul of antitrust scrutiny, Microsoft voluntarily offered to enter into a consent decree with the FTC, which included a ten-year agreement to continue making Call of Duty available for PlayStation and its PlayStation Plus subscription service—the primary competitor in that relevant market. That provision is a clear response to the non-priced concerns raised in both Microsoft and AT&T–Time-Warner. Microsoft recognizing the parameters of what constitutes an effect on innovation in a consumer welfare context is one of the reasons why it has been so successful in court versus the FTC’s aggressive enforcement against its merger with Activision.
It follows that the DOJ leveraging Microsoft and AT&T–Time-Warner’s non-priced frameworks to advance antitrust enforcement can help responsibly rein in Big Tech, promote competitive mergers in digital markets, and buffer against overenforcement from too zealous regulators.
The Khan Effect
FTC Chair Lina Khan has made her feelings about the consumer welfare standard no secret. In her January 2017 article in the Yale Law Journal titled “Amazon’s Antitrust Paradox,” she argues that the consumer welfare focus is too narrow and ignores the broader statutory mandate. She writes:
[T]he undue focus on consumer welfare is misguided. It betrays legislative history, which reveals that Congress passed antitrust laws to promote a host of political economic ends—including our interests as workers, producers, entrepreneurs, and citizens. . . . Antitrust law and competition policy should promote not welfare but competitive markets. By refocusing attention back on process and structure, this approach would be faithful to the legislative history of major antitrust laws. It would also promote actual competition—unlike the present framework, which is overseeing concentrations of power that risk precluding real competition.
That’s an antitrust view right out of the early twentieth century—more competition is better no matter the effect on consumers.
The question we need to ask is how effective this approach is in practice. If we look at the FTC’s track record on that front, it’s not especially successful. So far, Khan has focused her efforts on blocking mergers with a potential-competitors theory, but to little avail. When the FTC challenged Meta’s merger with Within Unlimited—a fitness VR company—using the potential-competitors theory, the court threw it out because the FTC had not brought enough evidence that the merger would lessen competition for either actual or potential competitors.
When the FTC attempted to block Microsoft’s merger with Activision, the district court judge found that the FTC had failed to proffer any real evidence that Microsoft’s merger would lead to competitive harm. There, the FTC’s theory was that Microsoft would withhold Activision’s games (like Call of Duty) from competitors (like the Sony PlayStation). But there was no evidence that Microsoft would preclude Sony from accessing Activision’s primary game franchises. Worse for the FTC’s case, the court identified evidence that Microsoft had no intention of making any of Activision’s titles exclusive to its Xbox. The judge found that the FTC had not identified “any instance in which an established multiplayer, multi-platform game with cross-play, that is, a game that shares Call of Duty’s characteristics, has been withdrawn from millions of gamers and made exclusive.”
The problem with these results, for those concerned about market concentration, goes beyond this specific case. Because of the FTC’s poorly structured briefs, we now have case law that leaves fewer tools, such as potential-competition theory, to fight that concentration.
In fairness, the FTC has had some successes. In its case against Facebook for monopolizing the personal social media market, a judge determined that the FTC had developed appropriate metrics for measuring market power and has moved the case forward. The judge also allowed the court to consider metrics such as fewer data and privacy protections, less ad choice, and more ads to show competitive harm, as such factors could indicate a low quality of service.
But this minor success only occurred after Facebook won its initial motion to dismiss because the FTC did not provide a valid market definition. What’s more, the court has only allowed the case to move forward; it did not rule in the FTC’s favor. This mistake by the FTC set it back four years—the FTC initially filed in 2020. Additionally, the relief the agency is asking for may not even be feasible as it would require Facebook to unwind two mergers it completed more than a decade ago with WhatsApp and Instagram.
The FTC’s monopolization case against Amazon, too, is on an even more precarious footing because it hinges on whether Amazon’s use of a now-retired algorithm (“Project Nessie”) amounted to Amazon imposing a form of monopoly rents. As in its case against Facebook, the FTC is seeking a significant divestiture, but that too would have to be based on whether a court agrees to the FTC’s extremely narrow market definition, which it calls the online superstore market. This means that the judge would have to discount Walmart, Target, and a whole host of brick and mortars as interchangeable providers. Though not impossible or without precedent, it nonetheless will be an uphill battle.
These cases are simply not as clear cut as the ones brought by the DOJ. Furthermore, it is not even clear that the FTC’s proposed remedies would address the consumer harms it identifies. If the FTC wins its case against Amazon, then Amazon would have to discontinue a tool it has already abandoned (Project Nessie). How does that help anyone? If the FTC somehow wins its case against Meta, would Instagram even be able to operate on its own? If it still had to contract with Meta to access some functions, share resources, or get access to its ad network, would the competitive landscape really be any different than it is today? Or would the virtual reality fitness market really be any different (or even be relevant) had the FTC blocked Meta’s merger with Within Unlimited?
The Path Forward for AI and Emerging Technologies
It’s becoming increasingly evident that the DOJ’s Borkian scalpel is more effective than the FTC’s Brandeisian hammer to address concerns surrounding Big Tech concentration. Moreover, the DOJ’s cases can have a real impact when it comes to evaluating competitive and consumer harms for nascent technologies like AI. For instance, Google is a significant player in the AI race. Not only does it have a strong presence at key data entry points of the internet’s ecosystem, it pays Apple, a competitor, billions to be the default search engine on the latter’s browsers. Furthermore, Google and Apple are teaming up again to integrate Google Gemini—its AI functions—in all of Apple’s devices. Given that Google and Apple make up over 99.6 percent of the smartphone market, this will mean smartphone AI will essentially be defined by the two companies.
The DOJ’s search case and, to a lesser extent, ad-tech case can create additional breathing room for competitors in not just the search engine market, but also for AI. Why? Because it would prevent those companies reaching anticompetitive arrangements, like setting Google Search or Gemini as a default, by establishing that such actions may ultimately harm the consumer.
The Apple case can also substantially limit concentration because, similar to the remedy in Microsoft, the obvious solution would be to require Apple’s operating system to be interoperable. This means that, even if Apple wanted to make such deals with Google’s Gemini, consumers would not be limited to Apple’s offering only. They would be free to download third-party apps and even app stores to compete directly with Apple’s or, rather, Google’s AI applications.
At the same time, the FTC has also opened investigations into AI. Recently, the FTC issued a Section 6(b) order that requires some larger companies to disclose various generative AI investments and partnerships. Section 6(b) orders allow the FTC “to conduct wide-ranging studies that do not have a specific law enforcement purpose.” The Commission is requesting that Alphabet, Amazon, Anthropic PBC, Microsoft, and OpenAI disclose what strategies they are deploying to develop their AI systems, which include who they are partnering with and what direct investments they are making or receiving in developing those systems. The focus of this inquiry, however, remains far too limited, including only the largest AI companies even though the AI ecosystem is far vaster than that, encompassing hundreds of companies, such as Cohere and C3 AI. It’s far from clear that the investigation will yield any insight into AI’s competitive landscape given the order’s focus.
Another aspect to consider is the FTC’s track record in pursuing this type of investigation. For instance, almost three years ago, the Commission launched an investigation into the privacy practices of nine social media and video streaming companies—including TikTok, Facebook, Twitter, YouTube, and Amazon—using the same authority. The investigation was wide-ranging, demanding information about what data is collected, how it is collected, how it is used, and how it affects American children. And the investigation carried the force of law—the Big Tech targets were required to respond within forty-five days. And yet, thirty-nine months later, the Commission has not released the results of its work. Not for Google. Not for Amazon. Not for Facebook. Not for anyone. Given that data is the fuel for all of AI and it covers almost all of the same companies, why won’t the FTC release that report first before entering into another fishing expedition? If it wants to rein in Big Tech, this also is a great way for the FTC to deliver.
Starting a new investigation before concluding the previous one makes little sense given that the FTC’s investigation appears to be focused on purely antitrust matters. If so, then why not let DOJ handle it as it not only has the requisite expertise, but also a far better success rate? The DOJ not only has the tools, but has already brought three cases that address AI issues.
In short, if we have learned anything from the competing antitrust approaches of the Biden administration, it appears that an expanded consumer welfare standard, incorporating non-priced harms, is likely to win out. Neo-Brandeisian critiques may be more theoretically exciting for some antitrust advocates, and perhaps have helped to raise public awareness. The FTC has also taken meaningful action in other areas, such as recently banning noncompete agreements. But the DOJ’s enhanced consumer welfare approach seems poised to deliver more concrete successes in addressing concentration in Big Tech and beyond.