Keynote as Prepared for Delivery
Thank you so much Cristina, I’m incredibly excited to be here. You have assembled a remarkable collection of competition enforcers and scholars from around the globe. I am particularly pleased to be joined at this conference by fellow U.S. enforcers at the federal and state levels and the many inspiring enforcers from Europe and across the globe.
Conversations like this really matter. We are talking about one of the great economic challenges of our time. The rise of monopoly power over digital markets threatens our democracy in a way we have not seen in generations.
Self-governance cannot be reconciled with monopoly power over digital markets that are central to almost every aspect of our daily lives. In 1890, the U.S. Congress was concerned that monopoly control over a market was “a kingly prerogative, inconsistent with our form of government.” For that reason, they passed the Sherman Act, the first U.S. federal antitrust law. They recognized that a free society cannot coexist with an all-powerful king presiding over any aspect of politics or the economy.
Market based economies and democracy complement one another because they each give citizens choice and opportunity. As citizens, we demand choice and opportunity as part of our fundamental freedoms.
Today, citizens have too little choice in many digital markets. Citizens have too little choice where they get their information, or who takes and uses their personal data. They have too little choice which algorithm will decide what news is promoted across the culture. They have too little choice how to interact with their social network online. This threatens democracy and puts at risk economic progress and prosperity. It stifles open markets and competition.
The digital revolution has created new business paradigms and new markets, but we must acknowledge new market realities demand new approaches to competition enforcement. Competition policy and law enforcement must adapt to address the problems of this new era, just as U.S. antitrust enforcers did a century ago in reacting to robber barons in coal and steel.
That means addressing the realities of the markets of today. Many businesses no longer fit models of vertical inputs and horizontal competitors like the rail lines and factories that resulted in the development of prevailing competition policy. To effectively protect competition in this era, we need to think carefully about the changes in the commercial realities that underlie the digital economy.
I would like to focus today on one particularly important development — moat-building. Building anticompetitive moats to protect monopoly positions has emerged as an important strategy for digital platforms. The durable monopolies that threaten our competitive markets demonstrate the success of these moat-building strategies.
Investors and executives recognize the value of having a moat around a strong market position. Warren Buffett is famous for encouraging investors to find “economic castles protected by unbreachable ‘moats.’”
Anticompetitive moats protect the monopolists’ core product or service. Given the value of a dominant position, monopoly maintenance has become a strategy for dominant firms to maintain their position.
The strategy does not merely involve leveraging a dominant position — moat-building means erecting barriers to entry that protect the core monopoly itself. That’s an important distinction. We must understand how anticompetitive moat-building strategies go beyond tying or leveraging. Even while leveraging a dominant position, the act of building and deepening a moat ultimately protects the core platform monopoly from entry and disruption. Moats defend the core platform monopoly from every direction. In the digital economy, competition could come from platform participants, potential entrants or disintermediating technology.
Moat building uses exclusionary strategies in combination to ensure the deepest and widest barrier around the economic castle of the core monopoly service. Two novel features of digital markets drive moat-building strategies in ways we have not seen before. First, powerful network and feedback effects across entire connected ecosystems can give rise to enormous monopoly power and alter competitive incentives. Digital platforms rapidly grow in scale and power faster and farther than prior networks, which were limited by the time and effort associated with building physical infrastructure. This creates a very deep moat around powerful digital platforms.
The second novel feature of digital markets is their wide range of business relationships and the extreme power dynamics therein. In smokestack industries, companies have fairly simple relationships with vertically related suppliers and horizontal competitors. In today’s digital ecosystems, however, a dominant platform can be a critical trading partner to thousands, or hundreds of thousands, of smaller businesses. The efforts of independent businesses to participate in platform ecosystems drives the success of the platforms.
With the remainder of my time, I’ll discuss three specific strategies enforcers can use to address moat-building and repair competition in digital markets. I will start with mergers before turning to exclusionary conduct.
First, we can update our merger enforcement tools to better fit these market realities. As enforcers, we need to acknowledge the web of acquisitions that monopolists can use to prevent entry into their core markets. The strategy is simple — buy up any firm that shows even a modest potential to develop into a competitive threat.
As enforcers, if we focus only on acquisitions of firms already set to enter a market, we miss acquisitions that allow digital platforms to strengthen their moats through innovation. We should acknowledge that nascent competitor acquisitions do not have to be purely horizontal or vertical — in digital markets, a nascent competitive threat can offer any novel or differentiated product in an adjacent market. Moats surround the entire castle, which means buying up firms anywhere in the ecosystem.
In the United States, I believe our Congress drafted the antitrust laws to address, among other things, the threat of dominant players buying nascent competitive threats. When debating the last revisions to the U.S. merger law after World War II, this concern was clear. The House of Representatives Report on the amendments explained this. It expressed concern with control “achieved not in a single acquisition but as a result of a series of acquisitions,” and “intended to permit intervention in such a cumulative process.”
Likewise, the Senate report explained that “the intent here…is to cope with monopolistic tendencies in their incipiency and well before they have attained such effects.”
A faithful application of this Congressional intent and the plain text of the statute to modern market realities demands aggressive enforcement against acquisitions by firms that already possess a dominant position.
We hope to advance this thinking through the review of our merger guidelines, now underway in partnership with the Federal Trade Commission. The project aims to incorporate wide public input in order to strengthen enforcement against illegal mergers. One of the questions we have raised is whether enforcers should do more to address acquisitions that, in the words of our statute, “tend to create a monopoly.” We have also asked about nascent competitor acquisitions generally. I look forward to reviewing the comments and to developing guidance that strengthens our enforcement to better protect competition in the face of acquisitive moat-building strategies.
Second, we must also assess moat-building strategies when considering unilateral conduct. That means examining the entire course of conduct of a monopolist across its entire ecosystem.
The moat analogy is particularly helpful here. A moat does not work if it’s only built in front of the castle — the attackers just go around the side. For the same reason you can’t just build it to the left or the right. The entire point is to build a complete defensive barrier. In the same way, enforcers cannot just look at one aspect of a firm’s conduct, leaving the others to separate consideration. You have to look all around the moat to see it for what it is.
This demands that we assess all of a monopolist’s exclusionary conduct in light of the overall strategy to maintain dominance. Reviewing moat-building conduct in a vacuum or in distinct parts risks misunderstanding the basic commercial realities at play. The anticompetitive effect of one aspect of the strategy is magnified by the other parts.
So my second suggestion is that we consider as a whole the course of exclusionary conduct by dominant platforms.
My third and final suggestion is particularly timely. Modern digital platforms have a powerful ability to discriminate in ways that harm competition and exclude entry. I talked earlier about the web of business relationships we see in the modern digital economy. When a dominant digital platform serves as a critical trading partner for thousands of businesses and millions of users, it has enormous power to influence their behavior.
The mere threat of being discriminated against by the platform becomes existential to other businesses. If your business becomes disfavored, your sales or prospects for growth could dry up overnight. This makes discrimination, and the threat of discrimination, an important and effective exclusionary tool.
Self-preferencing works much the same way. The threat of facing self-preferencing can coerce firms to cooperate and not compete with the platform for fear they would be the next target. Self-preferencing can expand the moat by discouraging potentially competitive businesses on the platform from innovating in ways that would compete with the platform or disintermediating the platform.
These discriminatory strategies can come at great cost to competition. They protect the dominance of the core platform businesses, helping to maintain monopoly positions by harming competition. At the same time, independent businesses that face the threat of discrimination have less reason to invest in their own products and services in the first place. Why invest in an innovation if it’s subject not only to copying, but to discriminatory preferencing by a critical sales channel? The resulting impacts on innovation sap dynamism from our digital economy, and discourage investment in areas that might threaten dominant platforms.
That is why I’m delighted to announce that the Department of Justice has formally weighed in with the United States Congress in support of legislation intended to clarify the illegality of discriminatory tactics by dominant digital platforms. Our House and Senate are considering legislation led by Congressmen David Cicilline and Ken Buck, and by Senators Amy Klobuchar and Chuck Grassley.
This important bipartisan legislation would clarify the illegality of anticompetitive and exclusionary discrimination by dominant platforms. It would help enforcers prevent these harms. In a letter sent just this week, the Department expressed its strong support for the legislation and encouraged Congress to work to finalize it and pass it into law. We are eager to provide full-throated support for this worthy and important endeavor.
Let me conclude by reiterating how hopeful I am. Since joining the Antitrust Division as AAG, I have been inspired to find so many hard-working and brilliant people working to solve the monopoly problems our economies face. I’ve found that same enthusiasm from enforcers and legislators all around the world, many of whom are here today.