Last week, the United States Department of Justice (DoJ) filed its proposed remedies following a federal judge’s finding that Google monopolized the market for search services. DoJ is asking for broad and deep interventions to address the liability determined in Judge Amit Mehta’s opinion, and they’re worth analyzing in context. DoJ cites the Microsoft case from the late ‘90s and early 2000s to support its proposition that it can prevent even the unlikeliest of potential anticompetitive conduct with its proposed measures. This is a bit ironic, however, since the Microsoft example illustrates most vividly that DoJ’s proposed remedies in the present case are not a great fit. And we should know. ACT | The App Association, founded in 1998 as the Association for Competitive Technology (ACT), advocated vigorously against the more far-reaching proposed remedies (many of them from competitors) against Microsoft because of how they would affect small businesses’ ability to compete in the relevant markets and access key technologies.
Today, Google faces similarly severe proposals to eliminate its discretion over access to its resources and to compete in adjacent markets. But in this case, the proposals come from the putative adult in the room, DoJ, as it goes all in on behalf of competitors rather than small businesses and consumers. For example, DoJ proposes that Google provide unfettered access to “indexes, data, feeds, and models used for Google search, including those used in AI-assisted search features, and (2) Google search results, features, and ads, including the underlying ranking signals, especially on mobile.” Here, DoJ warns that genuine privacy concerns must be distinguished from pretextual arguments to “deny scale to rivals.” This passage appears to suggest that all businesses relying on Google’s offerings would rather they be free and commoditized instead of differentiated and high quality. It also reveals DoJ’s willful disregard for privacy as a critical aspect of competition for consumers in digital markets and a key feature that small businesses rely on digital platforms to provide.
While DoJ may earnestly seek better competitive prospects for other companies in this lawsuit, disintegrating and proscribing future investment by companies that own managed online marketplaces would harm small business prospects in the app economy. Small businesses must be able to tap the global markets, seller and developer services, and deep wells of consumer trust that large managed marketplaces provide. They further demand that the artificial intelligence (AI) tools and other building blocks for their own offerings benefit from significant investment and competition from those best positioned to do so in capital-intensive input markets. With the benefit of hindsight, viewing the DoJ’s proposed remedies through the historical context of the Microsoft case helps us all evaluate how its grand plan would undermine the economic dynamism it seeks to protect.
This is not to say the online marketplaces are perfect or always going to do the right thing. The point is that there are several reasons to be skeptical of DoJ’s proposed remedies: 1) the remedies are based on flawed forecasts of the future; 2) DoJ does not seem to have learned from past experience in similar cases; and 3) DoJ likely underestimates the unintended consequences such significant interventions would precipitate.
Predicting the Future. In an especially worrisome—and outwardly unself-conscious—passage, the remedy framework threatens that “[a]n effective remedy requires administration as well as protections against circumvention and retaliation, including through novel paths to preserving dominance in the monopolized markets. This is especially true in dynamic industries like the markets at issue here” (emphasis added). Ultimately, it is unlikely that DoJ misses the internal inconsistency of saying the government should be primed to step in because a market is dynamic. Dynamism is one of the main reasons for the government to leave market activity alone—even if market participants are so untrustworthy as to have profits in mind rather than DoJ’s preferred industrial organization. That DoJ is proposing this administrative approach as a way of enforcing potentially spinning off “Chrome, Play, and Android” is strong evidence that it is reading dynamism the wrong way.
The structural proposal to split the Google Play store off from the rest of the company, for example, seems to be a sledgehammer-style solution to the original complaint’s allegation that Google ties Google Play to its search widget when dealing with handset manufacturers. The obvious benefits of having preinstalled apps “simply work” for consumers when they turn on a smart device for the first time aside, you do not need to break an entire division off of a company to prevent bundling the Play store with other offerings. Going with that nuclear option seems rooted partially in a desire to exact revenge on Google for doing so much better than rivals but also to serve a highly speculative notion that continued control of Google Play will result in Google finding brand new ways of advancing its search offering in anticompetitive, rather than procompetitive ways. Thin threads like these should not be a basis for chopping up a company, especially when the integrated nature of its offerings is a primary source of their value.
In order to follow through on DoJ’s proposed “flexible” administrative approach to the behavioral remedies, Judge Mehta would have to identify a business practice as fitting into the category of carving a “novel path to preserving dominance,” perhaps in an adjacent market that is not searchable. How would the court, guided by DoJ, do this? In practice, the remedy would likely push the court to view any of Google’s significant investments in adjacent markets as presumptively anticompetitive and prohibited, creating an exceptionally myopic, anti-consumer, and anti-small business outcome. Adjacent markets to search, such as the development of generative AI, are resource-intensive and, therefore, will naturally exhibit concentration when economic conditions are healthy (as they are now). Ensuring that competitors, be they well-resourced like Google or smaller startups, have a strong incentive to move into new markets should be among the highest priorities for antitrust lawyers and economists. Otherwise, the incumbents in those new markets will not face credible challenges, and competition will not yield the benefits it should. Small businesses are perhaps the most important consumers of various vertical levels of generative AI, from the apps themselves down to computing power, data, and foundation models. Their ability to compete with larger rivals in their own markets will depend at least in part on robust investment in the various vertical links in the chain by firms with the know-how and resources to do so in generative AI. Presumptively eliminating a substantial portion of this investment would undermine small companies’ competitive prospects. In proposing its “flexible” behavioral remedies, DoJ’s heedless foray into prophesy could bring about the possible AI dystopia it tries to ascribe to Google’s success in search.
Ignoring the Past. In November 2001, DoJ and Microsoft reached a settlement on the final antitrust remedies (the revised proposed final judgment or RPFJ) that would bind Microsoft in the coming years. Pursuant to the Tunney Act, DoJ opened up its RPFJ to public comment, unleashing a flood of “helpful” suggestions from Microsoft’s competitors to more completely eliminate Microsoft’s ability to compete in their markets. DoJ’s succinct rejection of these entreaties stands in stark contrast to the Department’s current posture: “The most persistent complaint is that the fencing-in and restorative provisions are not absolute prohibitions on competitive activity by Microsoft or absolute requirements that Microsoft surrender its technology for the benefit of competitors. . . . Protecting competitors from legitimate competition from Microsoft is not a goal of public antitrust enforcement” (emphasis added).
We fully agreed with DoJ’s dismissal of competitor demands to turn Microsoft into a public utility and eliminate its competitive overtures in other markets. In a Senate Judiciary Committee statement for the record, specifically taking issue with an alternative proposed settlement from a handful of state attorneys general, ACT argued that “requiring that Windows ‘must carry’ Java does nothing for consumers who can download it with one click and only serves to thwart competition by giving Sun Microsystems a special government-mandated monopoly with which other Middleware companies will have to compete. . . . [R]equiring Microsoft to port its Office product to Linux is tantamount to making it a ‘ward of the state.’ There are already several office productivity suites available to Linux users, and some are even free.” We opposed open-ended mandates for Microsoft to deal with rivals because they would eliminate Microsoft’s investment and innovation incentive, a necessary precondition for its continued contributions in adjacent markets that have demonstrably benefited small businesses that use their enterprise platforms, tools, and services.
In the present Google search case, DoJ’s proposed open access regime also evinces an intent that the remedy be a form of forced dealing with rivals. Such a requirement would typically flow from a type of liability that does not apply in this case and is a remedy antitrust law generally does not dispense. And for good reason. Forcing rivals to deal with one another, all else being equal, tends to lead to restricted output, higher prices, or both—the opposite of what antitrust law and policy seek. Even worse, DoJ goes further than to mandate that Google “deal with” rivals. It recommends that Google be required to “provide support for educational-awareness campaigns that would enhance the ability of users to choose the general search engine that suits them best.” Such advertising campaigns for entire industries are generally illegal under antitrust law because they are collusion, but some sectors have obtained statutory exceptions that come with “check-off” programs (e.g., “Beef: it’s What’s for Dinner,” from the cattle industry). Requiring that Google not only make its most valuable search assets fully available to DuckDuckGo but also to pay for DuckDuckGo’s advertising is an indefensible protection for a large competitor that puts small businesses and consumers last. Small business users of Google search tools often find them most effective because they enable accurate and powerful advertising services. Moreover, now that generative AI tools are competing directly with traditional search, the market will benefit from continued investment in ad networks that complement or are embedded in them. Stripping core search assets for parts and effectively removing Google’s ongoing incentive to outdo its rivals would diminish the end product for small businesses using those tools.
What can the Past and Present Teach us About the Future? As our 2001 hearing statement noted, Microsoft’s competitors stoked fears that it would enter adjacent markets like “instant messaging and digital media.” To prevent this, their proposed remedies mirrored the general approach of DoJ’s Google plan, to give the court carte blanche flexibility to smack down any future play into an adjacent market. With the benefit of hindsight, DoJ was absolutely right to take our advice in 2002 and reject calls to give the court free rein to stop competition from Microsoft. The fears of Microsoft’s dominance in instant messaging, digital media, “personal digital assistants (PDAs), cell phones, set-top boxes/game consoles, web terminals and powerful servers that connect them all” were unfounded. To the extent these markets exist in the forms they were contemplated 20 years ago, Microsoft has generally not been a meaningful competitor, with the exception of gaming consoles. Even in that market, Microsoft is not dominant and is even striving for access to mobile platforms to sustain its gaming business.
Most importantly, the decision not to mandate unfettered open access and eliminate competition in adjacent markets was a good thing for small businesses. Microsoft continued to innovate on its core operating system and enterprise software while also building a robust and dynamic set of cloud services that small businesses everywhere rely on, especially those in software and connected device markets. All of this tells us that the future is hard to predict, especially in dynamic industry sectors. Small businesses in the app economy stand to gain immensely from Google’s continued investment and resource allocation to emerging markets like those around generative AI. The court stands at a critical phase as it considers DoJ’s proposed remedies and should appropriately interpret the dynamism DoJ references to mean it should err on the side of consumers benefiting from evolving market forces, rather than eliminating competition from Google.