Last month, the U.S. Department of Justice and eight states sued Google, claiming it runs its digital ad business to unfairly advantage its own business, to the detriment of its customers and potential rivals. Harry First explains what makes this case so legally and procedurally unusual and the obstacles the government plaintiffs will face.
On January 24, the United States Department of Justice and eight states filed suit against Google, alleging that the way in which the company built and operates its digital display advertising business violates U.S. antitrust law, specifically, Section 1 (restraint of trade) and Section 2 (monopolization) of the Sherman Act. In a 140-page complaint, government enforcers describe the somewhat arcane business for buying and selling the digital ads that appear “magically” when we visit a website. There are 13 billion of those ads published each day—and that doesn’t even include ads in search results, on “closed” systems like Facebook or Snapchat, on mobile apps, or on streaming video. Control of this business is significant.
So, why is the jury out? For one, the governments’ case is not an easy one legally, as I’ll explain below. But there is a more literal reason. In an unusual move, the DOJ is seeking money damages, alleging that Google’s illegal conduct, which led to “supra-competitive fees, manipulated advertising prices, and lower quality advertising matches,” harmed the U.S. government as a website advertiser. When a plaintiff asks for money, it is entitled to have a jury decide its case, so the DOJ has included in its complaint a demand for a trial by jury.
Since 1955, the federal government has had the right to ask for damages to its “business or property” caused by an antitrust violation. Since 1990, it has had the right to seek treble-damages, as has long been the case for private plaintiffs. Inexplicably, the DOJ has almost never sought money damages from antitrust violations, even in price-fixing cases where it could have. In fact, in every monopolization case before this one, the DOJ has asked just for equitable relief, that is, a court order to end the defendant’s anticompetitive conduct and restore competition to the market. These types of cases—a good example is the famous case against Microsoft for maintaining its Windows monopoly—are tried before a judge as the factfinder.
The decision to seek damages and demand a jury trial in a monopolization case is unprecedented. Furthermore, the government plaintiffs also brought their case in an unusual court. To the extent legally possible, a plaintiff generally tries to pick a court that it believes will be most favorable to its case. The plaintiff in an antitrust case is allowed to bring suit in any U.S. district in which the defendant does business. In its two most recent monopolization cases (against Facebook and Google for search) the government plaintiffs chose the District of Columbia. This is where they had sued Microsoft and where they convinced the DC Circuit Court of Appeals to uphold most of their monopolization cases. Suing in DC would give the government the advantage of that court of appeals decision, which is binding on the courts in that district.
But, the government plaintiffs brought this suit just across the Potomac, in the Eastern District of Virginia. Appeals from that trial court’s decision will go to the Fourth Circuit Court of Appeals, not to the DC Circuit. The Microsoft decision will still be important, but it will not be controlling.
Strategic litigation decisions can have a major impact on the ultimate outcome of a case, so these choices are significant, even if their implications are not yet clear. The likeliest reason the plaintiffs chose the Eastern District of Virginia is speed. A major criticism of monopolization litigation, particularly when high tech is involved, is the slowness of the litigation process itself. Courts, it is said, proceed on legal time; technology proceeds on internet time. By the time the litigation process is over, the effort to impose relief to restore competition in fast-moving tech markets is like “trying to shoe a galloping horse,” to quote the trial judge responsible for remedy in Microsoft.
The Eastern District of Virginia is known as a “rocket docket.” Through various local procedural rules, the courts in that district are thought to dispose of their cases more rapidly than other courts. The data bear this out. The median time from filing a case until trial in the Eastern District in 2022 was 18.1 months; in DC it was 55.9 months. Appeals are quicker, too. In the Fourth Circuit the median time for disposing of appeals was 7.8 months; in the DC Circuit it was 11.6 months.
One sign that the choice of venue for the litigation may be more favorable to the plaintiffs is that Google wants to change it. It has already stated its intention to file a motion to move the case to New York, where a raft of antitrust cases involving Google’s ad tech business are pending. The law allows a change of venue “for the convenience of parties and witnesses, in the interest of justice,” but it is hard to see why the transfer would be justified, particularly in light of the policy underlying a recently passed statute designed to bolster the ability of government antitrust enforcers (whether state or federal) to litigate in the forum of their choice. If nothing else, the procedural wrangling over venue will likely slow down litigation.
At some point, though, unless the DOJ withdraws its claim for damages, the case will be tried by a jury. At first blush, a jury trial sounds like a bad idea. The jurors—twelve lay people—will first need to understand how “Google’s vast, opaque ad machine” (to use the Wall Street Journal’s description) is put together. This will involve the technology for aggregating the huge demand for and supply of internet display advertising (and the differences between this advertising and other types of digital ads that might be substitutes). It also requires understanding how the various platforms for that aggregation operate together to allow for the instantaneous matching of the supply of ad space with the demand from advertisers for that space.
Next, the jurors will need to understand the rules that Google has devised for running the auctions on its various platforms (first- and second-price auctions, double bidding, enhanced dynamic allocation, dynamic revenue sharing, to name a few), as well as Google’s decision to acquire several companies between 2008 and 2011 and the effect of those acquisitions on Google’s business and market power.
Finally, the jurors will need to assess whether Google’s conduct unduly restricted competition, perhaps by deceiving advertisers or publishers, overcharging them, or preventing the emergence of effective competition. Google, of course, will have its own take on how its technology has benefited advertisers and website publishers alike.
Does this complexity make the case hopeless? In an article I co-authored with Spencer Waller a decade ago, “Antitrust’s Democracy Deficit,” we argued that antitrust had become overly technical and too fearful of democratic institutions, particularly of juries. We urged more use of juries in antitrust decision-making, both because of the democratic preference for citizen jurors over unelected judges and on the pragmatic political ground that antitrust cases should be understood by the citizens whose interests they are intended to protect, not just by the technocrats who bring them.
Trying this case before a jury will certainly be a challenge, and a bold step by the plaintiffs—as is the request for damages and the choice of the court in which to litigate. Even bolder, though, may be the substance of the governments’ complaint.
First, the kinds of conduct about which the government complains is unusual. This case is not about exclusionary conduct aimed at a particular upstart competitor that threatens the monopolist’s business; it’s not about predatory low pricing (or loyalty discounts) intended to push a particular competitor out of the market; it’s not about refusing to supply essential inputs to a competitor so as to maintain a monopoly. These types of accusations are more common and precedent exists for evaluating them. Rather, the complaint targets Google’s assembly of a system and its alleged manipulation of the rules for running that system that improperly advantaged Google, disadvantaged its customers, and prevented potential rivals from emerging.
Second, the complaint frames the case in a holistic way. “Although each of these acts is anticompetitive in its own right,” the complaint reads, “these interrelated and interdependent actions have had a cumulative and synergistic effect that has harmed competition and the competitive process.”
Courts have generally been skeptical about holistic approaches, derisively labeling them as “monopoly broth.” Instead, they have most often looked at a monopolist’s conduct piece by piece, thereby neglecting the systemic nature of the competition problem. This piecemeal approach may be particularly problematic in this case, where some of the pieces may not fall into recognized categories of exclusionary conduct and others may seem too remote in time to have caused the anticompetitive harm of which the plaintiffs complain. Only through a holistic approach, however, can the jury appreciate the extent of Google’s dominance of this industry and can the court enter an effective remedy.
The governments’ approach to this case is bold—procedurally and legally. Will it be successful? For now, “the jury’s out.”