Principal Deputy Assistant Attorney General Doha Mekki of the Antitrust Division Delivers Remarks at Mercatus Center Second Annual Antitrust Forum: Policy in Transition

 

Thank you for that kind introduction. It’s a pleasure to be with you today.

At the outset, let me congratulate you, Alden, and the staff of the Mercatus Center for putting together today’s program. You’ve brought together thoughtful antitrust lawyers and scholars for what is no doubt an interesting discussion about antitrust enforcement and competition policy.

In preparation for today’s program, I took the opportunity to learn more about the Mercatus Center’s mission and work. Devoted to “market-oriented ideas,” its research focuses on “how markets solve problems.”[1]

You might wonder then how—or even why—someone like me, a federal antitrust enforcer with an admittedly pro-enforcement outlook, finds herself at this sort of confab. But that is the wonderful thing about today’s program. Notwithstanding its worldview, the Center invites people of different viewpoints to share ideas and learning.

For that, I commend you and thank you for bringing all of us together. In times of social, political, and economic polarization, candid conversations like this one are an especially important public good. That is also why I chose this event to give my first significant address as Principal Deputy Assistant Attorney General in the Antitrust Division.

I. Introduction

We are so fortunate to practice in a time when debates about corporate power, antitrust enforcement, and the operation of markets enjoy such prominence in the public consciousness.

These issues and debates, of course, are not new to us. But in the last several years, the discussions have spilled onto kitchen tables across the country, our social media feeds, and of course our news media of choice.

Setting enforcement policy is a tremendous honor; with it comes tremendous responsibility, especially at a time when competition issues feature so prominently in our national dialogue. That is why drawing in a wide swath of viewpoints, including from those who live under the market structures and conditions that our enforcement and policy choices beget, is so important. That was the central thesis behind the agencies’ merger guidelines RFI process. Feedback sharpens our thinking, vets our assumptions, and, ultimately, leads to better considered policy judgments.

In this vein, I acknowledge there is a lot on which we may not agree. But one of my core beliefs and governing principles is that more unites than divides us. And I think that’s particularly true when it comes to the axioms of competition policy.

Let me explain.

In my conversations, I sometimes hear two competing views about the role of corporations. These views loosely affiliate with different ends of the political spectrum, though not exclusively so. Some may view corporations as inherently greedy. Extractive corporations take from consumers and workers alike, entrenching their position by using power for harmful ends. Others don’t see that profit motive as a bad thing. They subscribe to Milton Friedman’s famous line that “the social responsibility of business is to increase its profits.”[2] The invisible hand of the market, they say, transforms that self-promoting corporate conduct into something that benefits all of society.

What strikes me about these competing views, though, is that they share a common prescription: the vast majority of people believe, or at least say they believe, in the value of competition. Whether you are skeptical about the power of government or of corporate power, you likely attach some value to the rivalry between firms, and the benefits that rivalry can produce. If you see corporate motives as suspect, competition keeps corporations in check. And if you believe in less government intervention, you might also believe that competition and antitrust enforcement are alternatives to regulation.

Congress understood the value of competition and enshrined it in the Sherman and Clayton Acts. The Supreme Court, putting these concepts together, explained the necessity of free and unfettered competition.[3] “[T]he unrestrained interaction of competitive forces,” it said, “will yield the best allocation of our economic resources, the lowest prices, the highest quality and the greatest material progress, while at the same time providing an environment conductive to the preservation of our democratic political and social institutions.”[4] “But even were that premise open to question,” the majority wrote, “the policy unequivocally laid down by the Act is competition.”[5]

Maintaining competition is precisely the aim of the Clayton Act as well. Markets that are hobbled by anticompetitive restraints and concentrated by mergers cannot be expected to do what we want them to do. It is for that reason that Section 7 outlaws mergers that “may” operate “substantially to lessen competition” or “tend to create a monopoly.”[6] This is not a controversial mandate. For all the divisions that exist, the vast majority of people, of all stripes, seem to agree that markets should operate free of anticompetitive constraints and illegal market power.

But for years enforcers and courts hesitated to fully realize this mandate. If you look around today, my guess is that you can identify plenty of markets that are not yielding the “best allocation of our economic resources” or characterized by the “lowest prices [and] the highest quality.” In my view, that is because we have often enough perceived the possibility that mergers may be illegal, but we have declined to wield the power Congress gave us. In the worst cases, we have usurped the will of Congress, intentionally underenforcing the antitrust laws by administrative fiat. In other cases, we have tried to make predictive judgments about the likelihood of harm—more specifically when anticompetitive effects might not come to pass. In such cases, increasingly we raised the standard that Congress explicitly set forth, requiring something near certainty about the effects of a merger to block it.

Today, I’d like to talk through some examples of times when our reliance on predictions have failed us. I’ll then offer some suggestions for how we might think about an antitrust framework that is fit for purpose.

I won’t hide my thesis. I’ve been at the Antitrust Division for nearly eight years. I’ve seen different approaches to antitrust enforcement and competition policy. What I know is there’s no Delorean with a flux capacitor sitting in the Antitrust Division’s metaphorical driveway. We don’t have Bill and Ted’s phone booth. When we try to predict to near certainty the precise effects of a deal, we can get it wrong.

The question, of course, is what lesson to take from it. Perhaps, you say, if we can’t be certain about the ultimate effects of a merger, we should hang back and let market forces play out.

To that I say: it is easier to assess competition than to predict to a near certainty the competitive effects of a deal in a dynamic world. It is also more consistent with the framework that Congress set forth.

Congress was wise to this quandary. It knew the future of a marketplace is hard to predict, and believed that waiting for conduct to actually ripen into harm was an inadequate way to realize its goals. So, in passing the Clayton Act, it wrote a probabilistic statute that expressly did not require certainty about what happened in the future. Congress prohibited any merger whose effect “may be”—and those words, “may be,” are critical—substantially to lessen competition. Does that mean we prohibit more mergers than we would need to if we had a crystal ball that perfectly predicted the future? Of course. There is nothing radical about that approach. That was Congress’s design. As the Supreme Court explained, the Clayton Act is concerned “with probabilities, not certainties.”[7] Congress’s judgment was that we, in the words of the Supreme Court, need “to brake th[e] force” of “concentration in American business … at its outset and before it gathered momentum.”[8]

I think we have often failed to take this standard at its word. And all too often, the exact harm we were concerned about, but predicted might not materialize, did. That is radical. I can think of no other substantive area of law so central to our economy and society that was nonetheless intentionally underenforced.

We must take Congress’s will seriously. Here’s what that means in practice. When we look out and see various reasonably probable outcomes from a transaction, if one of them is a substantial lessening of competition, we should have a bias towards action. Section 7 doesn’t require certainty. It very well may be the case that no change in competition is also a possible outcome. Even that is not a bar to enforcement. We don’t need to ask what is the most likely outcome. We should not let a temptation to be crystal ball seers cloud our Congressionally-directed mission.

II. Lessons from the past

To illustrate this problem, I’d like to offer three examples.

1. Digital Advertising: Google/DoubleClick and Google/AdMeld

Earlier this week, we filed a landmark monopolization lawsuit against Google. Our complaint alleges in rich detail that, over the last 15 years, Google has unlawfully acquired and maintained monopoly power in digital advertising technologies.

So, what happened 15 years ago? Google acquired DoubleClick, which was then the “leading firm in the third party ad serving markets.”[9] This acquisition was a turning point. As we alleged in our complaint, it “vaulted Google into a commanding position over the tools [website] publishers use to sell advertising” and “set the stage for Google’s later exclusionary conduct across the ad tech industry.”[10]

Our investigation, of course, is not the first time an antitrust enforcer considered whether the DoubleClick acquisition would lead to anticompetitive ends. The FTC investigated the deal in 2007. As the Commission’s statements from the time make clear, they understood the possibility of harm even then. In fact, Commissioner Pamela Jones Harbour dissented from the Commission’s decision to close the investigation because “the combination of Google and DoubleClick has the potential to profoundly alter the 21st century Internet-based economy.”[11] Her clairvoyance was striking.

Nevertheless, a majority of Commissioners predicted that these harms were unlikely. We know differently today.

For example, the majority identified the risk that Google would “bundle or tie DoubleClick’s ad serving technology” to enhance its own position, or “leverage DoubleClick’s leading position in third party ad serving to its advantage in the ad intermediation market.”[12] But despite acknowledging this risk, the Commission did not take action because it was not certain this harm would come about. Now compare that prediction with what actually happened, as alleged in our complaint. After acquiring DoubleClick, “Google forged an exclusive link between Google Ads”—its homegrown, market-leading product—“and DFP through the AdX ad exchange”—the two products it acquired from DoubleClick.[13] As alleged in our complaint, and as Commissioner Harbour predicted, this move had “a profound effect on the evolution of digital advertising.”[14] It “tilted the industry in Google’s favor, driving publishers to adopt and stay on Google’s DFP publisher ad server in order to have access to Google Ads’ advertiser demand.”[15] And it “cut off the possibility that Google Ads’ advertising spending could sustain, or encourage the entry of, a rival ad exchange or publisher ad server.”[16]

The FTC explained its reticence to take action. It predicted that the ad intermediation market was “unlikely to ‘tip’”[17] and that the third party ad serving markets were “competitive and … likely to become even more so in the future.”[18] Not so. Again, compare these statements with what actually happened, as alleged in our complaint. Because of Google’s conduct, DFP became the “only realistic publisher ad server option.”[19] Its share, according to Google’s own documents, grew to “a remarkable 90%,” which gave Google a “durable monopoly over the publisher ad server market.”[20]

I could go on. The majority’s statement and the dissent read as a roadmap of the conduct Google actually undertook. Remarkably, it was clear to everyone what Google might do but, still, a majority of Commissioners wanted more certainty.

I do not mean to call out the FTC. As citizens, we owe the FTC a huge debt of gratitude for its work to protect consumers. It is a tremendous agency whose staff and leadership over the years have engaged thoughtfully and deliberatively about the challenges of the moment. Its challenges are hardly exclusive.

We at the Antitrust Division have fallen prey to the same thinking—with the exact same company. Just a few years after the DoubleClick acquisition, Google announced another step in its path toward dominance: the acquisition of AdMeld. AdMeld competed with Google to provide services to web publishers that facilitated the sale of display advertising space. We at the Division, like the FTC, identified a reasonable possibility of antitrust harm. But we thought this risk would not turn into reality. We predicted that “multi-homing,” a practice in which publishers would rely on multiple display advertising platforms, would “lessen[] the risk that the market will tip to a single dominant platform.”[21] We were wrong. “Google’s increasing scale and dominance across the ad tech stack, coupled with its subsequent exclusionary conduct, destroyed the ability of advertisers and publishers to effectively multi-home among alternative ad exchanges.”[22] As alleged in this week’s complaint, Google “folded [AdMeld’s] functionality into Google’s existing products, and then shut down its operations with non-Google ad exchanges and advertiser tools.”[23] And what happened? “As a result, the market tipped and AdX became the dominant ad exchange.”[24] As we allege in the complaint, “Google’s AdX is the largest ad exchange in the market; it is approximately four times larger than the next largest ad exchange . . . and has been for at least several years.”[25]

2. Entertainment: AT&T/Time-Warner

Even in full-blown trials, we see courts encounter the same challenges.

In 2017, the Division sued to block the merger of AT&T and Time Warner. At the time, the case was the first litigated vertical merger challenge in more than 40 years. Our theory of harm focused on the value of Time Warner’s entertainment content. We alleged that if AT&T were able to pull Time Warner content from AT&T’s content distribution rivals—called a “blackout”—it would drive those rivals’ customers to AT&T. We alleged that this was an incentive built into the structure of the market.

The Court did not challenge the legal viability of this theory. Nor did it dispute that there was some risk of this anticompetitive harm occurring. But, instead of blocking the merger, it relied on testimony from the companies’ executives to make a prediction: There “is likely never going to be[] an actual long-term blackout of Turner content,” it said, referring to a division of Time Warner.[26]

A blackout of Time Warner content is exactly what happened. Less than six months after AT&T consummated the transaction, it pulled HBO from all Dish networks, starting a multi-year blackout of Time Warner content.[27] This was the first blackout in HBO’s 43-year history. And it was decidedly “long-term.” Lasting nearly three years, it became “one of the longest-running carriage disputes in TV history.”[28]

This should have come as no surprise. AT&T owned DirecTV, Dish’s primary rival for satellite TV. And the blackout, in the words of the Wall Street Journal, created an incentive for Dish “viewers … to drift to rivals like AT&T’s DirecTV.”[29] Dish itself blamed the power created by the merger for AT&T’s conduct: “Plain and simple, the merger created for AT&T immense power,” it claimed, and “AT&T is implementing a new strategy to shut off its recently acquired content from other distributors.”[30]

One other aspect of this case is worth mentioning: the predictions of efficiencies advanced by the parties and entertained by the court. After rejecting the Division’s theory of harm, the court predicted that the merger would yield “most, if not all” of the $1.5 billion in cost-savings suggested by the parties.[31] The merger, the companies and court seemed to say, would create a hyper-efficient body capable of delivering superior product to consumers.

Of course, we know how the story ended. Just three years after the merger was consummated, the companies reversed course. AT&T sold Time Warner, leading at least one national newspaper to wonder whether “this $100 billion deal [was] the worst merger ever.”[32]

I would be remiss if I did not pause to reflect on the sentiment at the time this lawsuit was filed. The wisdom of the merger challenge was the subject of significant speculation and writing. Suffice it to say, I hope and believe history will vindicate the Antitrust Division, its tremendous staff, and my former boss Assistant Attorney General Delrahim who had the courage to bring the challenge.

3. Beer: Miller/Coors

The limits of our predictive power surface in other industries too. Consider beer.

In 2008, the Division investigated a joint venture between Miller and Coors. We found that the two companies were close competitors—competition that would be lost after the transaction. We nonetheless allowed the deal to go forward on the theory that “the joint venture is likely to produce substantial and credible savings that will significantly reduce the companies’ costs of producing and distributing beer,” which would, in turn, “have a beneficial effect on prices.”[33]

Did this “beneficial effect on prices” materialize? Not according to a 2017 merger retrospective by economists Nathan Miller and Matthew Weinberg. The authors found precisely the opposite. They document an “abrupt[]” increase in prices right after the deal was consummated.^[34]  Importantly, these price increases were not only the result of unilateral effects—they also were evident in MillerCoors’ major competitor, Anheuser-Busch, suggesting coordinated effects.[35]

III. The Path Forward

So where does all of this leave us? I believe there are five lessons we can draw.

1. Congress says what it means and means what it says.

We need to enforce the statute as Congress wrote it. Decades of Supreme Court precedent speak to the need to arrest concentration and trends towards concentration in their incipiency. In many cases, including those I’ve described today, agencies and courts accurately identified a risk to competition, but did not act because they weren’t sufficiently certain the harm would come to pass.

But that is not the standard. Congress wrote a statute that applies whenever a merger “may” substantially lessen competition. It’s an incipiency statute. Congress was concerned that, if we wait until the harm has actually come about, it would be too late.

We cannot wait for evidence that harm is sure to exist. It’s the rare case that we will see documentary evidence laying out a post-merger price hike or a plan to shutter innovation or diminish quality. That puts a premium on the value of structural evidence. When a change in market structure suggests that a firm will have an incentive to reduce competition, that should be enough. Congress did not write a statute asking whether a substantial lessening of competition was “certain” or the “most likely” outcome. As long as that harm “may” occur, the merger violates the law and should be enjoined.

2. Accept that firms will exercise their market power.

In every merger review, our core inquiry is: how does competition present itself in this industry? Once we see that a merger may lead to, or increase, a firm’s market power, only in very rare circumstances should we think that a firm will not exercise that power. Too often, judgments that a firm will obtain market power, but somehow not exercise it, have turned out to be wrong.

Here’s what that means in more concrete terms. In the vast majority of circumstances, promises that companies will not exercise their market power, including for reasons of professed reputational concerns, should not be a viable defense to an otherwise anticompetitive merger. Reputation evidence, such as it is, is often too fleeting, too fickle, and too malleable to protect the public from illegal mergers and abuses of market power.

The same is true of executive testimony and corporate promises to continue competing or freeze prices. Like reputation evidence, it puts the public at the mercy of the merged firm. Simply put, there is no guarantee that an executive’s view today will remain the company’s plan tomorrow.

3. The structural presumption is an important proxy for market power.

While competition analysis is more dynamic than static, information about market structure can help screen off the risks of unilateral effects and the likelihood of future oligopolistic behavior. The structural presumption, like other bright lines and rules, also helps businesses conform their conduct to the law with more confidence about how the agencies will view a proposed merger or conduct. Again, I come back to my thesis. It is easier to measure competition than to predict competitive effects in a dynamic world, and that is more consistent with the framework that Congress set forth.

4. Be skeptical of efficiencies.

When we think a merger may substantially lessen competition, the optimal remedy is to block it. Sometimes, I’ve heard a suggestion that a merger should be permitted, even though it decreases competition, because it creates efficiencies for the merged firm. This is not, however, the scheme Congress envisioned. It is competition that Congress wanted to preserve. Not efficiency. The Clayton Act does not ask whether a merger creates a more or less efficient firm—it asks about the effect of the merger on competition. The Supreme Court has never recognized efficiencies as a defense to an otherwise illegal merger. And in the exceedingly rare instances when low courts have considered them, they have explained that there is a high burden for demonstrating that efficiencies increase competition.

​​​​​​​5. We must be willing to litigate.

Too often, instead of litigating to stop an anticompetitive merger, we have relied on consent decrees to try to fix the problem. We have relied on a prediction about what schemes would address, in the future, the structural issues created by a merger. Again, we have too often been wrong. The FTC’s own study on remedies, released in 2017, found that when the Commission relied on divestitures of limited packages of assets in horizontal, non-consummated mergers, the scheme failed to restore competition almost one third of the time.[36] Other studies released at the time suggested an even greater failure rate.

Litigation is intrinsically risky. Enforcing a probabilistic statue comes with an inherent risk that a court will see probabilities differently. Perhaps our desire for certainty in merger enforcement was born out of this fear. But if we only challenge the cases in which we’re nearly positive that anticompetitive harm will come to pass, we’re not enforcing the law as Congress wrote it.

If we take our mandate seriously, we need to acknowledge the possibility that sometimes a court might not agree with us—and yet go to court anyway.

*          *          *

At the Division, we are putting these principles to work. We are taking action when we think mergers imperil competition. We are not relying on untested efficiencies to approve mergers, nor are we coming up with predictions for why firms might not exercise the market power a merger gives them. When we see a reasonable probability that a merger may harm competition, our bias is to protect competition.

And we are having success. In the recent Penguin Random House/Simon & Schuster merger, our investigation showed that the acquisition would have enabled Penguin, which was already the largest book publisher in the world, to exert outsized influence over which books are published in the United States and how much authors are paid for their work.[37] So we took the parties to court. And we won. This victory was important because it was principally a merger challenge about monopsony power.

In other matters, parties have abandoned mergers after we expressed competition concerns. Cargotec and Konecranes proposed a merger that we believed threatened to harm competition in the sale of shipping container handling equipment, which is critical to the global supply chain. The parties ultimately abandoned the deal.[38] When Verzatec, a producer and seller of pebbled fiberglass reinforced plastic wall panels sought to acquire its biggest competitor, Crane, we filed suit under both Section 7 and Section 2.[39] There, too, the parties walked away. And, similarly, when two of the world’s four suppliers of insulated container boxes and refrigerated shipping containers sought to merge, the Division conducted a thorough investigation, prompting the parties to abandon.[40] In all of these cases, we observed the risk of harm to competition and focused on taking action.

Of course, not all cases will turn out in our favor. The fact that we have not won every case we have brought is, to me, a feature, and not a bug, of our approach to enforcing the law Congress wrote. The desire for certainty is at least one root cause of our underenforcement in recent years.

Here’s what I’d like to leave with you today. We share a common goal in protecting competitive markets. We share a belief that companies should be able to compete on the merits and the best ones should win. And I suspect we also agree that the world, and our economy, are complex, and we are enforcement cannot be conditioned on reliably predicting the effects of deals with certainty.

If we do indeed share these premises, I’d like you to consider the prescription I’ve given about the way forward: it is easier to measure competition than to predict competitive effects in a dynamic world, and that is more consistent with the framework that Congress set forth.

From my experience, when we demand certainty, we raise the standards for enforcement well beyond what the statute contemplates and we underenforce the law. Congress wrote a statute concerned with probabilities and focused on stopping incipient threats.      

As the saying goes, the time is late and I must be going now.

Let me close by saying what an honor it is to be with you. I congratulate you again, Alden, on today’s event.

Thank you.