Skip to main content

Assistant Attorney General Makan Delrahim Delivers Remarks at the Antitrust Division's Second Roundtable on Competition and Deregulation


Washington, DC
United States

Good morning. I am Makan Delrahim, Assistant Attorney General for the Antitrust Division. I welcome you to the Department of Justice, the only agency in government named after a moral imperative. 

Today, we are pleased to continue the Division’s series of roundtables on competition and deregulation. I am joined at the table by several of my colleagues from the Antitrust Division here at DOJ. 

I would like to thank our distinguished participants in this roundtable discussion and their organizations for their participation today. It is an honor to have with us so many thought leaders in the areas of antitrust, competition and regulation. Let me also thank everyone who has made a written submission; it will be part of the record. I am encouraged by the strong turnout of individuals in the audience interested in these topics, and by the comments the public has submitted.

Today’s roundtable will focus on antitrust consent decrees, and on deregulation through different approaches to consent decrees. Just as this roundtable is in the middle of our three-part series, today’s topic is central to our goals to streamline antitrust enforcement. I believe antitrust enforcers should approach using consent decrees consistent with a view of the Antitrust Division as a law enforcement agency, not a regulatory one. As Justice Robert Jackson once posited, “[w]e should not spend great sums to obtain decrees which are economically unenforceable and, when carried out in form, are often only lessons in futility.” 

Former Assistant Attorney General of the Antitrust Division, William Baxter noted in 1982 that “[d]ecree provisions that were perfectly sensible and desirable when entered can be unreasonable today if they have been successful in promoting competition where there previously was none. . . . Where competition has been restored in the relevant market, the continued effectiveness of such provisions serves only to restrain competition, not to promote it.” He also stated that, “[O]ur understanding of industrial organization and the dynamics of competition has improved markedly in recent decades. Many older decrees reflect legal positions that were based upon mistaken economic theories.”

I am happy to report that several initiatives are now underway at the Division: An emphasis on structural relief in remedying anticompetitive mergers as the Supreme Court has emphasized repeatedly; improvements to consent decrees to make them more enforceable and less regulatory; the establishment of the Office of Decree Enforcement; and an initiative to terminate legacy antitrust judgments.

First, we have announced a renewed emphasis on seeking structural relief when possible, as opposed to regulatory behavioral conditions, to remedy anticompetitive mergers. Doing so is consistent with the Division’s broader emphasis on antitrust as law enforcement, not regulation.

I am deeply skeptical that Congress, in enacting Section 7 of the Clayton Act, envisioned a regime in which the Antitrust Division or a federal court would become the overseer of a company with thousands of employees, earning billions of dollars in annual revenues, and second-guessing market competition or future consumer or business behavior. In Section 7, Congress did not call for illegal mergers to be regulated, it called for them to be prohibited.  

As Justice Gorsuch reminded us in an opinion earlier this week, we should “Start where the statute does.” The text of Section 7 is clear. It deems unlawful any “acquisition” whose “effect . . . may be substantially to lessen competition.”  A violation of Section 7 arises in many cases from the acquisition of a company as a whole, but in some instances, the violation comes from the acquisition of one or more parts or subsidiaries. In either case, The Clayton Act specifically contemplates proceedings initiated by the Department of Justice praying that “such violation shall be enjoined or otherwise prohibited.” The text specifically contemplates an injunction or other form of prohibition of the violation. If Congress envisioned that anticompetitive mergers should be permitted so long as they are regulated by the Department of Justice or other agencies of the government, it certainly could have said so. 

It is therefore unsurprising that many courts, including the Supreme Court, consider divestitures or other structural relief to be the preferred method of remedying problematic acquisitions. According to the Supreme Court, “[c]omplete divestiture is particularly appropriate where asset or stock acquisitions violate the antitrust laws,” and such relief “is simple, relatively easy to administer, and sure.”

The Division’s preference for structural relief extends beyond legitimate concerns of administrability and efficiency. As enforcers of the antitrust laws, we have a duty to ensure that the risk of a failed remedy—and thus harm to consumers—falls on the parties to the unlawful transaction, not on the American consumer and taxpayer. This approach is consistent with the text of the Clayton Act, which makes illegal a merger that “may” substantially lessen competition. Congress recognized that a risk of harm is what renders a transaction unlawful—or, as courts say, a “reasonable probability” of anticompetitive effects.

Where such a risk exists, the role of enforcers or a court should be to eliminate the risk entirely or place that risk on the parties—not to design elaborate remedies that purport to reduce that risk while still permitting the merged company to retain control over the source of harm. Divestiture of the source of anticompetitive harm substantially eliminates the risk of harm. Behavioral conditions merely lower the risk, assuming they are effective. At worst, behavioral conditions can fail to curb a company’s natural economic incentives to act in ways that harm competition.

Most behavioral decrees cure neither the incentive nor the ability of the merged company to exert enhanced leverage gained through an anticompetitive merger. That is particularly true where such conditions require a third party victim to be willing to come forward to challenge the actions of a necessary business partner, potentially scorching their business relationship. Rational businesses may well decide that paying a higher supercompetitive price is better than spending millions in legal fees and risking an unfavorable ruling. That is not an outcome dictated by the free market. 

We take seriously our responsibility to protect the American consumer against the risk of harm, as the stakes can be enormous. Some mergers can profoundly transform an entire industry, or have a domino effect by precipitating other acquisitions until the competitive landscape has changed. Once this has occurred, it is often too late, and it is nearly impossible to unring the bell. Behavioral conditions merely muffle the bell’s ring. Divestiture, by comparison, preserves separate control, and leaves open the opportunity for independent innovation and collaboration through arms’ length transactions.

Turning to the Division’s second recent initiative, the Division is incorporating a set of improvements to its consent decrees that will make them more enforceable and less regulatory. The Division has included these provisions in all recent settlements, and will include them going forward in settlements of merger and civil non-merger actions.  

These include provisions that the Division may establish a violation of a consent decree by a preponderance of the evidence, as opposed to clear and convincing evidence. They also include provisions allowing the United States to apply for an extension of a decree’s term if the court finds a violation of the decree. The terms require defendants to agree to reimburse the taxpayers for attorneys’ fees, expert fees, and costs incurred in connection with any consent decree enforcement effort. In addition, the Division can, after a certain number of years, terminate the decree upon notice to the court and defendants. 

A third development at the Division with respect to consent decrees is the creation of the Office of Decree Enforcement, which I announced at the University of Chicago last week. The office will help ensure compliance by parties to current consent decrees and our enforcement of those decrees. 

Yesterday, we announced our initiative to terminate outdated antitrust judgments. Nearly 1,300 “legacy” judgments remain on the books of the Antitrust Division, and many remain open on the dockets of courts around the country. Some of these judgments are nearly 100 years old. The vast majority of them no longer protect competition because of changes in industry conditions, changes in economics, changes in law, or for other reasons. The perpetual consent decrees call to mind the famous line from the Eagles song, “Hotel California”: “You can check out any time you like, but you can never leave.” 

In the earliest consent decrees, in the first two decades of the antitrust laws, there was little guidance on what constituted a violation of the relatively new Anti-Trust Act. As a result, antitrust settlements played a significant role in providing guidance to businesses on the rules of the road. Our earliest consent decrees were almost always perpetual. Why? Simply because that had been the default historically, stretching back at least to the twelfth century in agreed-upon judgments, inscribed upon vellum much like this perpetual decree from 1599. 

It’s not that enforcers failed to envision that markets would change, or that they assumed that antitrust recidivism was extraordinarily high. That’s just how settlement decrees had been written for centuries. AAG Jackson challenged the aluminum monopoly in 1937—at about the same time the Department was purchasing substantial amounts of aluminum to provide the impressive detailing and sculptures in this building, which I hope you noticed on the way in. Jackson recognized that the decree the Department entered into with Alcoa 25 years earlier, in 1912, perpetually enjoining specific illegal behavior, had not foreclosed the possibility that Alcoa could find other ways to inhibit competition. And, it required the Supreme Court, in 1938, to determine that the 1912 perpetual decree did not prevent the Department from bringing future antitrust cases against Alcoa -- which it did.

Periodic reforms of the antitrust laws throughout the 20th Century noted the undesirability of perpetual decrees. This is because a decree can be rendered obsolete not only by changes in our understanding of the way markets work but also by changes in the law. The Division long followed the policy of seeking perpetual decrees in cases against per se violations of the antitrust laws. However, certain conduct previously pursued as per se illegal became properly analyzed under the “rule of reason.” Firms bound by perpetual decrees based on an outmoded per se view of a restraint are unfairly prevented from adopting policies that, if adopted by rival firms and analyzed under the “rule of reason,” may be perfectly lawful and procompetitive. Such an imposition obviously denies consumers the benefits of market competition. 

It is important to remember that consent decrees arise from a conclusion that a transaction or behavior was illegal. At the same time, a cornerstone of the American legal system is that settlement is favored, and indeed most lawsuits are settled before trial. In this regard, antitrust is no different than many other areas of law. Parties often desire to abide by the law, and they seek ways to pursue their objectives within the bounds of what the law allows. 

In this way, consent decrees involve putting into effect a settlement, where the parties bring their conduct or transaction to conform to the law, in the context of an enforcement action. The antitrust agencies find the right tools to address the issues presented; however, the tool picked for a company in 1920 may not be the same tool that’s right for the job today. So when a decree from 1920 is still in effect, binding companies that are successors of—or successors of successors of, and so forth—the companies in 1920, the consent decree can distort incentives in marketplace dynamics and present complex and expensive compliance challenges.

We remember from history reading of the Gilded Age steel trust, the railroad trust, the oil trust—corporate forms that gave us the distinctly American word “antitrust.” That said, we still have in effect a consent decree concerning what the Department, quite literally, described in 1913, the bicycle “coaster brake trust.” The various successor companies today make cabinets, asbestos, power drills, and filters, likely not in competition with each other, yet are still bound by a decree crafted for the sake of competition in the “coaster brake” market. 

To facilitate the termination of these judgments, the Division is reviewing all of its legacy judgments to identify those that no longer serve to protect competition. The Division has assigned each judgment to a Division attorney, who will examine court papers, internal case files, and publicly available information to determine whether each judgment continues to serve competition. The Division has already identified many judgments that it likely will seek to terminate unilaterally after a public comment period. It will begin its efforts by proposing to terminate a set of judgments entered by the federal district courts in Washington, D.C. and Alexandria, Virginia.  

Today’s discussion will consider perspectives on these issues and others. And it is important that we who work in the antitrust agencies are aware of these perspectives, for a consent decree by its own terms, can be entered only with the consent of the parties.

Our first session this morning will focus on behavioral consent decree provisions. We will look back – exploring how consent decrees have been used, the trends, and lessons learned from various approaches. Then we will look ahead, and examine how enforcement should employ consent decrees going forward.  

In the second session, we will focus on the usefulness of perpetual consent decrees, and on terminating outdated ones.  More specifically, we will consider what effects perpetual decrees have on market competition, how companies subject to perpetual decrees adjust their conduct, and what role industry reliance on a decree should play with respect to the decision to eliminate or retain a perpetual decree. This session will also consider specific perpetual decrees, perhaps most prominently the ASCAP and BMI decrees. In addition, we will welcome our panelists’ insights on issues the Division should consider in carrying out the Judgment Review Project.

And now for the logistics. I will ask each of our panelists to provide a brief opening statement immediately after I introduce them. At the conclusion of the opening statements, we will begin the first session and discuss behavioral decree provisions. We will then have a brief 10 minute break, and return with our second session, discussing perpetual consent decrees. At the conclusion of session two, we will have a wrap-up in which each panelist will have the opportunity to make a brief closing statement.

Now, let me turn to our panelists. Thank you once again for your willingness to participate today and to share your views on these important issues. We appreciate your time and your views.

Updated April 26, 2018