DOJ v. Google: Six Weak Spots in Judge Mehta’s Decision

In a much-anticipated decision, Judge Amit P. Mehta of the U.S. District Court for the District of Columbia found that Google illegally monopolized the general search market through a series of agreements with browsers, Android OEMs, and wireless carriers to make Google the default search engine. The opinion marks the first major court decision finding that a “Big Tech” company engaged in monopolization since the historic U.S. v. Microsoft case, upon which Judge Mehta heavily relies in ruling against Google. Despite being heralded as a landmark achievement for Biden administration antitrust enforcers, the victory may be short lived. While the next phase is a remedies trial, Google has already stated its intention to appeal and will likely be able to highlight at least six significant weak spots in Judge Mehta’s opinion.

#1: Market Definition

In concluding that Google has monopoly power, Judge Mehta defined the key relevant market as “general search services.” While this market definition may seem to make sense—although Google does face significant competition from many specialized search providers like Amazon and Yelp—Judge Mehta’s reasoning really doesn’t. Specifically, after essentially admitting there was no good quantitative evidence of substitutability supporting the government’s theory, Judge Mehta relied on what are known as the “Brown Shoe factors,” an open-ended qualitative market definition test examining “practical indicia,” to conclude that Google had a monopoly in general search. However, not only has the Brown Shoe test long been criticized as highly subjective and best used as a supplement to quantitative economic evidence, but it is from a 60 year-old merger case. Indeed, in his opinion, Judge Mehta appears to cite no Supreme Court or D.C. Circuit case where the Brown Shoe factors have been used in a monopolization context.

#2: Barriers to Entry

On the facts, Judge Mehta found that Google’s high share in this general search services market was protected by barriers to entry and specifically that “artificial intelligence (AI) has not sufficiently eroded barriers to entry—at least not yet.” And yet, while elsewhere in the opinion Judge Mehta does concede, as he must, that “AI technologies have the potential to transform search” and the advent of “generative AI is perhaps the clearest example of competition advancing search quality,” he misses the fundamental Schumpeterian nature of this competition. Just because AI has not “supplanted” search does not mean that it isn’t providing a competitive constraint to search. To give a simple but clear analogy: that the need for audio disks has not been “eliminated” does not mean that digital music hasn’t created meaningful, effective competition. In other words, not only do firms like OpenAI already have Google in their sights in search, but AI is transforming how we access information right before our very eyes.

#3: Consumer Harm

Not only did Judge Mehta have to find that Google had a monopoly, but also that it violated the antitrust laws by engaging in exclusionary conduct to maintain it. And, under modern antitrust law, that means proving that Google’s various default agreements harmed consumers. However, in an otherwise meticulously well-organized opinion, all the key arguments Judge Mehta identified as to why Google’s conduct was anticompetitive—foreclosure of a substantial share of the market, depriving rivals of scale, and reduced incentives to invest and innovative—are much more about harm to competitors than harm to consumers. Indeed, the latter two findings regarding reduced scale and incentives to invest are ultimately just second order effects of the first finding regarding foreclosure, which modern courts, including Microsoft, have made clear serves as a “screening function.” That is, more is needed to show harm to consumers as distinct from mere harm to competitors.

#4: Behavioral Economics

In finding that Google deprived rivals of scale, Judge Mehta endorsed both behavioral economics theories as well as their application by the DOJ to the Google case regarding the so-called “power of defaults,” or the so-called limitations consumers face in clicking away to another search engine. But this is no ordinary factual finding. The modern antitrust enterprise is rooted in neoclassical economics, including assumptions of rationality that the Supreme Court has utilized in foundational antitrust cases like Matsushita Elec. Industrial Co. v. Zenith Radio Corp., which requires that circumstantial evidence of a conspiracy must attest to a “rational economic motive” that tends to exclude the possibility of independent behavior. Indeed, in other contexts, like securities fraud, for nearly 40 years the Court has adhered to the efficient markets hypothesis, which presumes that a misleading statement is relied upon by an ordinary investor who holds the security. As such, Judge Mehta’s decision to deviate from this general understanding of market rationality should not go unnoticed by the D.C. Circuit.

#5: Causation

In addition to showing some type of anticompetitive conduct and consumer harm, Judge Mehta had to find a causal connection between them—in other words, that the former is what has resulted in the latter. To do this, Judge Mehta abandoned the presumptive “but-for” causation test used in cases like FTC v. Rambus and instead relied on Microsoft, which applied a more lenient “reasonably capable” test to take into account the fact that Netscape was only a potential competitor to Microsoft’s operating system. And while there is a reason why a lower standard might make sense in the circumstances where a potential competitor is being excluded (the harms are more forward looking by nature), these are not the facts of the Google case. Indeed, Google’s principal search rivals are not “nascent threats,” but have long competed in search. And, by not applying the but-for causation test, Judge Mehta made it easier for the DOJ to get around one of Google’s core arguments: that even without or “but-for” its default agreements, consumers would still use Google search because it is, as Judge Mehta admits, “the industry’s highest quality search engine.”

#6: Procompetitive Justifications

Finally, Judge Mehta found that none of the many procompetitive justifications Google proffered for its conduct were compelling enough to shift the burden back to the DOJ to show that the purported anticompetitive harms outweighed these benefits. This holding highlights the one-sided nature of the opinion: even in Microsoft, the existence of some procompetitive benefits were acknowledged. And, here again, Judge Mehta appears to have misapplied the law, albeit implicitly. Specifically, Judge Mehta claims to have not evaluated whether Google’s conduct was the least restrictive means to achieve procompetitive benefits in discarding Google’s justifications. However, that’s exactly what he seems to have done in rejecting, for example, Google’s argument that its default agreements improve user experience in holding that even assuming Google “has established the value of a default placement to competition and consumers, it has not shown that exclusive defaults across all key access points have such utility.” But showing that exclusivity is in some way the least restrictive means to achieve these benefits is not required by the Microsoft standard Judge Mehta claimed to apply, meaning that at least some of Google’s procompetitive justifications should have held up even by his own factual findings.

To sum up, this case is far from over, and Google will have several strong arguments it can make on appeal to challenge Judge Mehta’s decision. However, before the appeal, the matter will next move to a remedies trial where it will be determined how Google’s supposed anticompetitive conduct should be addressed. And even here, there is already talk of wild antitrust remedies that are untethered to the findings in Judge Mehta’s opinion. For example, in a New York Times op-ed, former Biden antitrust official Tim Wu suggested that Google could be broken up by forcing the divestiture of its Chrome browser and its Android operating system. But doing this would be a clear abuse of judicial discretion and an affront to the rule of law. To be sure, while there was debate about whether Microsoft should have to separate its operating system from its browser—a remedy ultimately deemed unnecessary—here there are no allegations that Google is leveraging any supposed market power in either its browser or Android operating system to benefit its search monopoly (e.g., Google requiring Chrome or Android users to use its search product). At bottom, and in the words of Judge Mehta, Google is what it is today because it has “innovated consistently, and made shrewd business decisions.” Hopefully courts will ensure that this antitrust case doesn’t end up hindering Google from continuing to do just that.