Antitrust Remedies After Google

In a conversation with The Regulatory Review, Herbert Hovenkamp, one of the world’s foremost experts in antitrust law, examines the recent Google monopolization ruling and its implications for the future of U.S. competition policy.

During the first Trump Administration, the Department of Justice filed a lawsuit against Google alleging that the company had “unlawfully” maintained a monopoly in the general search advertisements, search services, and text advertising markets. The case proceeded with separate proceedings to determine liability and remedies. The federal trial court decided in 2024 that Google had violated Section 2 of the Sherman Antitrust Act in the general search services market. In particular, the court held that contracts that made Google the default search engine on Android devices and browsers, including Apple’s Safari and Mozilla’s Firefox, were unlawful.

Many commentators speculated whether the court would force Google to sell off its Chrome web browser at the remedies stage. But, as Hovenkamp argues, breaking off Chrome from Google would not have addressed the anticompetitive conduct at the heart of the case.

Instead, the trial court’s remedies decision in September mandated a termination of Google’s exclusionary contracts with browser developers and devices. The decision also imposed requirements for Google to share its data with competitors. Hovenkamp emphasizes that Google’s data index was the “biggest asset” supporting its market dominance. The court agreed, stating that “making data available to competitors would narrow the scale gap created by Google’s exclusive distribution agreements.”

In his interview with The Regulatory Review, Hovenkamp also considers the impact of artificial intelligence (AI) on the court’s decision. He notes that, while the court did not discuss AI during the liabilities stage of the court’s proceedings, it played a “significant” role in the court’s determination of remedies. As new AI tools advance and are increasingly incorporated into search engines, the technology will continue to create uncertainty about the future of the search engine market.

A leading voice in antitrust law, Herbert Hovenkamp is the James G. Dinan University Professor at the University of Pennsylvania Carey Law School. He is a fellow of the American Academy of Arts and Sciences. Hovenkamp, alongside the late Philip E. Areeda, is the coauthor of the authoritative treatise, Antitrust Law: An Analysis of Antitrust Principles and Their Application. U.S. Supreme Court Justice Stephen Breyer once remarked that “most practitioners would prefer to have two paragraphs” from this teatise on their side than “four Supreme Court Justices.” In 2008, Hovenkamp won the Justice Department’s John Sherman Award for his lifetime contributions to antitrust law.

The Regulatory Review is pleased to share the following interview with Herbert Hovenkamp.

The Regulatory Review: The Justice Department sued Google in 2020, alleging that Google violated the Sherman Antitrust Act through its search platform. Last year, the court found Google liable. How did it reach this conclusion?

Hovenkamp: The court held that Google’s 90+ percent market share was attributable to two factors: the superiority of its search engine and its agreements with implementers—such as Apple—to make Google Search the “default” search engine on devices. Since consumers overwhelmingly stick with the default, this served to deny market access to rivals. For purposes of an injunction, the court held, it was sufficient that the default agreements were a likely contributing factor to Google’s market dominance.

TRR: The trial court decided recently against remedying Google’s violation by forcing Google to sell off Chrome, its popular web browser. Did this decision surprise you?

Hovenkamp: No, for two reasons.

First, Chrome is a browser, not a search engine, so divesting Chrome from Google would not break into Google’s monopoly at all. Instead, a breakup would have left the monopoly intact. Indeed, “breaking up” the search monopoly would be impossible as a matter of engineering.

Second,  the judge concluded that a breakup with Chrome would be incredibly “messy” and would interfere with Google’s functionality.

TRR: The court did require Google to share its search data with competitors. How will sharing search data address the anticompetitive conduct for which the court found Google liable?

Hovenkamp: The biggest asset supporting Google’s market dominance was its very large database of search index files—more than twice as big as its closest rival, Microsoft Bing. Sharing the database could enable rivals to produce larger indexes of their own.

TRR: In that remedies decision, the court devoted over 30 pages to discussing the relationship between search platforms and AI. What role did AI play in the court’s decision against breaking up Google?  

Hovenkamp: A big one. AI was almost completely absent from the original liability trial, but had a significant presence a year later in the remedies trial. Generative AI created considerable uncertainty about the impact of AI on future search engine market share, and AI itself was facilitating the entry of new programs, such as ChatGPT and Perplexity, with search engine features.

TRR: As AI evolves, will the court’s remedies, such as data sharing requirements, become inadequate?

Hovenkamp: Who knows? The case can be re-opened for up to six years. Like most equitable remedies, it’s a bit of a guess about the future. The trick is to share enough data so that the new licensees can become equal rivals, but not so much that they become mere agents reselling Google Search.

TRR: Over two decades ago, the Justice Department brought a lawsuit alleging that Microsoft illegally attempted to monopolize the web browser market. The case ended with a court-approved settlement known as a consent decree. How does a consent decree work in antitrust litigation, and could Google and the Justice Department enter into one following the court’s recent decision?

Hovenkamp: A consent decree is just a negotiated settlement. It operates as a contract and, in many cases, gives the court a right to intervene in order to settle disputes. Yes, it is still possible at this time that the Google search case could end up in a settlement via consent decree. Alternatively, one or both of the parties can appeal to the D.C. Circuit, but the parties could still settle at any stage.

TRR: You have written that devising the right remedy is often more difficult than establishing unlawful conduct. What are some ways to improve the remedies process in antitrust cases?

Hovenkamp: First, the separate hearing for determining liability and remedies, such as the judge held in this case, helps. Second, antitrust enforcers need to stop rushing to “breakup” every time a case involves a dominant firm. Historically, breakups in monopolization cases have not been very successful. They have done better in merger cases. A well-designed behavioral decree can do much better, and I think Judge Amit Mehta, who presided over the case, did a good job this time. There is always a certain amount of guesswork.

TRR: Looking ahead, how will this outcome impact antitrust enforcers’ approach to remedies?

Hovenkamp: Injunctive remedies plus some minimal sharing of assets should be the first place to look. To the extent possible, enforcers should want the dominant firm to give up just enough so that the firm’s rivals can become full, facilities-based competitors developing their own products and not merely serving as resale agents.