Justice News

Deputy Assistant Attorney General Michael Murray Delivers Remarks at the Santa Clara University School of Law
Santa Clara, CA
United States
Friday, March 1, 2019

Thank you for that kind welcome.  It is a pleasure to be here with you at Santa Clara University Law School.    

I am here today to talk about the Department of Justice’s antitrust initiatives in the labor space.  As you may know, and as a panel later today will discuss, the Department has been active in this area for many years, perhaps most prominently several years ago with a series of settlements concerning no-poach agreements and technology companies.  Today, though, I would like to discuss the Department’s current activities, including several recent amicus filings that have prompted discussion in the bench, the bar, and industry.  My goal is to set these filings in both historical and legal context, to provide a better understanding of the Department’s views in this important area of antitrust practice. 

Before turning to that task, however, I would like to briefly discuss the related initiative that gave rise to those filings. There have been quite a few stories in the press recently about the Department’s renewed focus on its amicus program, which consist of the Department’s discretionary filings in private antitrust cases.  Commentators are saying that the Division is “get[ting] off the sidelines” and “seeking to make its voice heard.”  The reaction to our filings has ranged from encouragement to hostility—encouragement from those we are supporting and hostility from those we are not supporting.  It turns out the age-old adage of “where you stand depends on where you sit” is fully applicable to antitrust.

This reaction to our amicus program is half accurate.  It is true that we are making an active amicus program a priority.  Specifically, as our Assistant Attorney General, Makan Delrahim, told Congress, we are trying to expand our amicus program in order to “more proactively and more effectively promote the use of antitrust and competition principles across the judiciary.”  To that end, we filed significantly more briefs in 2018 than 2017 (9 versus 1), and if the current rate holds we will double that again in 2019, to approximately 20 briefs.  

But it has been our legal obligation since at least 1975 to file amicus briefs in just this way.  In that year, Attorney General Levi codified a delegation of authority to the Assistant Attorney General of the Antitrust Division.  The text of it, which still exists today, is that the Assistant Attorney General of the Antitrust Division is “assigned” and “shall conduct” and “handle” “[g]eneral enforcement . . . of the Federal antitrust laws and other laws relating to the protection of competition and the prohibition of restraints of trade and monopolization, including . . . participation as amicus curiae in private antitrust litigation.”  So although our focus is renewed, it is well-grounded.  It should be no surprise that we are doing what we are tasked to do.  

That the amicus program is discretionary, though, raises a significant question: how do we decide which cases to pursue?  The answer is that we learn of significant antitrust cases in the same way that scholars do:  we search for them in the reporters and the dockets, we read about them in the news, and parties and organizations bring them to our attention.  It is worth pausing on that last point: it is increasingly common for parties and organizations to come to us with a pitch about a case that they believe raises significant issues.  That makes a lot of sense, because it is likely that we are already reviewing the case, and parties reasonably believe that it is useful for them to try to persuade us of their view of the case.  In this way, our amicus program is becoming somewhat similar to the amicus program of the Solicitor General, in which parties routinely advocate in person and in writing that the Solicitor General file an amicus brief on their side in the Supreme Court.  This is now a routine part of our practice. 

We sort through these cases in the same way that we sort through which affirmative cases to bring in our criminal and civil programs:  we exercise prosecutorial discretion.  We are guided by the oath that we take as federal employees to “support and defend the Constitution” and “faithfully discharge the duties of [our] office.”  And we balance resource and policy constraints with the interest of the United States in a proper administration of the law.  Indeed, the very fact that we are more active on the amicus front appears to redound to the benefit of the interest of the United States, by discouraging private parties from making the more extreme versions of their arguments, due to the possibility of a contrary United States amicus brief that could undermine their credibility. 

A review of our filings indicates that we are willing to file on a wide variety of topics, from immunities and exemptions to the topic today, labor, to which I will now turn. 

We have recently offered several filings to courts considering the proper standard for assessing market allocations in a labor market.  

The reactions to our position have been swift and fierce.  The first filing, in the Western District of Pennsylvania, took the position that no-poach agreements among competitors are per se unlawful unless they are ancillary to a separate legitimate transaction or collaboration.  This position has been characterized by certain defense-friendly lawyers as an unsupported assertion based on lawless guidance that creates a “trap,” but hailed by plaintiff-friendly attorneys as pro-worker plaintiff.  The second filing, in Washington state, explained that no-poach agreements in the franchise context are likely subject to the rule of reason.  This position has been called a “welcome clarification” by defense-friendly lawyers but has been criticized by plaintiff-friendly lawyers as pro-business defendant.  We have not been able to please all of the people all of the time.

But of course that is fine with us, because we are not in the business of satisfying a particular constituency.  Our Assistant Attorney General has promised that “[w]e will be the officious inter-meddlers in random cases,” that “[w]e’re not going to take anybody’s side, but the side of what we believe the . . . law should be.”  I am reminded of the famous line from T.S. Eliot’s Murder in the Cathedral.  Archbishop Thomas Becket is foreseeing his eventual murder for standing on principle against the king.  He then is confronted by four tempters who propose pathways forward.  The first three advise him to do the wrong thing to avoid death, but the last urges him to do the right thing, to accept death, but not because it is right to stand on principle but rather because it would make him famous.  Becket responds that “The last temptation is the greatest treason:  To do the right deed for the wrong reason.”  In our amicus program, as our Assistant Attorney General said, we try to take the right position for the right reason, not because of the likely reaction.  In this way, the amicus program parallels the mission of the academy.  The good faculty member writes articles to advance the truth, for the motivation of advancing the truth, and not to be popular, or to get tenure, or to be liked by students, or to get on television.

What all this means is that the best way to understand our position in the labor antitrust space is not through a simplistic pro-labor or pro-business paradigm.  Rather, our position is best understood through the lens of the rule of law and as informed by history and doctrine. 

I will first briefly discuss the historical relationship between antitrust law and labor markets.  I do this for two reasons: first, it is critical to understanding the Department’s position; second, I am not aware of any authority—law review, trade publication, or treatise—that provides a detailed survey tailored for this purpose.

The historical relationship between antitrust law and labor markets is long and complicated.  It dates at least to the 1890s.  In discussing what became the Sherman Act on the floor of the Senate, Senator Sherman himself pointed to the power of monopoly to “command[] the price of labor.”  And, of course, employers used antitrust law to fight labor unionization before the passage of the Clayton Act.  Subsequently, a complex body of statutory and nonstatutory exemptions arose to address the intersection of antitrust law and labor markets.  Over time, antitrust litigation in labor markets has become “rare,” at least relative to other antitrust litigation.  But that is not because the laws do not apply.

Quite the contrary: federal antitrust enforcers long have been active in this space, and in particularly relevant ways for at least the past twenty-five years.  In the early 1990s, the FTC challenged nursing homes’ practice of boycotting registries for temporary nursing services and the Council of Fashion Designers’ attempt to reduce the fees and other terms of compensation for models.  The United States, for its part, sued a Utah society for human resources professionals for sharing wage information that caused the matching of wages.  In 2007, the Department also sued an Arizona Hospital Association for setting a uniform rate for per diem nurses. 

And, of course, in this decade, the Department pursued civil enforcement actions against several major technology companies that entered into naked no-poach agreements regarding their competitors’ employees.  At that time, significantly, the Department alleged that the behavior constituted a per se violation.

This enforcement focus on the labor markets has become even more acute in the past few years.  In October 2016, the Division and the FTC issued their Antitrust Guidance for Human Resources Professionals.  As the Principal Deputy Assistant Attorney General recently summarized, “[t]he Guidelines cautioned that naked agreements among employers not to recruit certain employees, or not to compete on employee compensation, are per se illegal and may thereafter be prosecuted criminally.”  

These Guidelines are notable in many ways.  They are directed to HR professionals, which is not our usual target audience for policy pronouncements.  Perhaps as a result, the Guidelines do not really argue for their viewpoint: they do not contain case citations or in-depth analysis of the law, but rather a few references to previous enforcement actions and several broad pronouncements in lay English.  Perhaps unsurprisingly in light of the audience and the message, critics have viewed those pronouncements as ambiguous in important respects.  For example, critics claim that the Guidelines are unclear as to whether settlements among companies litigating trade secrets cases are running afoul of the antitrust laws.  

For these reasons and others, the Guidelines remind me of a famous Seinfeld episode.  Jerry arrives at a car rental agency to rent a car, having previously reserved a mid-size.  The clerk tells him that they’re out of mid-sizes.  Jerry responds:  “But the reservation keeps the car here. That’s why you have the reservation.”  The clerk says: “I know why we have reservations.”  Jerry replies: “I don’t think you do. If you did, I’d have a car. See, you know how to take the reservation, you just don’t know how to hold the reservation and that’s really the most important part of the reservation, the holding. Anybody can just take them.”  In my view, the Guidelines are like the reservation: anyone can put out Guidelines.  The key is making them workable, explaining them, and justifying them, so that they can actually guide people.

The Division’s current leadership has endeavored to do just that.  About a year ago, Principal Deputy Andrew Finch announced that for unlawful per se “agreements that began after the date of th[e] announcement [of the Guidelines], or that began before but continued after that announcement, the Division expects to pursue criminal charges.” Around the same time, our Assistant Attorney General explained that there are open criminal investigations on this topic.

The Department’s first enforcement action, in the case known as Wabtec, followed shortly thereafter.  There, the Department successfully sued two of the largest railway producers in the country for agreeing not to poach each other’s employees, in the process alleging that the defendants’ actions constituted a per se violation of the antitrust laws.  The Department did not pursue criminal charges in that case; the agreement did not continue past the date of the announcement of the Guidelines.

Where does that leave us?  I think the following summary is fair: the Division has a longstanding enforcement interest in the labor space, including the no-poach space, and has taken the position, consistent with case law, that naked no poach agreements are per se violations of the antitrust laws.  

That brings me to our recent filings.  This is really a story of, as the Deputy Attorney General recently remarked, quoting the Book of Ecclesiastes, “there is nothing new under the sun.” For in light of this history, it is no surprise that our recent filings have taken the position that naked no poach agreements among competitors are per se violations while no poach agreements that are not naked or not among competitors are subject to the rule of reason.  

But I will spell out the proper thinking about these cases.  I do so because the Guidelines lack this detail.  The corresponding caveat is that the Division is not so presumptuous to believe that it has all of the answers in this space, but I do believe that these are the right questions to ask in a systematic way.

The first relevant question is whether the entities that allegedly entered into a no-poach agreement are capable of the “concerted action” required by Section 1. Copperweld and American Needle provide the test to be applied to this question.  In the typical case, applying these precedents is not a difficult inquiry.  Separate, unrelated corporations competing in the market for widgets are clearly “independent centers of decisionmaking.”  But, in other cases, the inquiry can be more complicated, because at its core it is fact-specific, based on what the Court called “a functional consideration of how the parties involved in the alleged anticompetitive conduct actually operate.”  For example, in the franchise context, at least one circuit court—your home circuit—has held that franchisors and franchisees are a single entity exempt from Section 1.  But that decision—which pre-dates American Needle—“require[d]” and turned on “an examination of the particular facts of [the] case.”  Indeed, a district court in the same circuit after American Needle reached the opposite conclusion, based on the allegations and facts in its case.  In sum, whether separate companies—be they companies in a franchisor/franchisee relationship or another arrangement—are capable of concerted action is a complicated question of fact. 

Assuming the relevant parties are capable of concerted action, the next item on the agenda is determining the proper standard of law—the per se rule or the rule of reason.  The next question, consequently, is whether the entities that allegedly entered into a no-poach agreement are competitors in the labor market, that is, whether there is a horizontal relationship among them with respect to the alleged no-poach agreement.  I emphasize here that companies can be competitors in the labor market but not competitors in product or service markets.  Companies in different industries can compete in the same market for employees.  But if they are not competitors in the labor market but instead are, for example, vertically related in their industry, then any agreement among them is subject to the rule of reason.  That is black-letter law as a general matter—most recently reaffirmed in Amex—and longstanding law in the franchise context as well since at least Continental v. GTE.  Franchisors and franchisees, of course, are primarily in a vertical relationship in their industry and generally not competitors with respect to the labor market. Consequently, agreements among them likely are subject to the rule of reason.  Indeed, this conclusion is not particularly controversial: the American Antitrust Institute recently proclaimed that no-poach agreements between a franchisor and franchisee are vertical restraints that are subject to the rule of reason.

If, however, the entities are not vertically related but rather horizontally related as competitors in the labor market, then they have entered into a classic market allocation.  As we all know and as Leegin reaffirmed, agreements among competitors to divide markets are per se unlawful, absent a significant caveat I will return to in a moment.  That per se rule extends to customer allocation schemes for new or existing customers, geographic divisions, and even seemingly small divisions of markets.  

An agreement to allocate employees is no different than one to allocate customers: one eliminates competition for employees, another for customers.  As the eBay court put it, no-poach agreements are “a ‘classic’ horizontal market division.”  The leading treatise on antitrust law—Herbert Hovenkamp’s Antitrust Law—takes the same position.     

I should pause to note that competitors, of course, must actually have entered into an agreement, as opposed to merely engaged in parallel conduct.  Gone also are the days when a hub-and-spoke-conspiracy could be established without a establishing a rim around the spokes.  But, of course, if there is a rim, then there is little point in using the hub-and-spoke analogy: that is really just a horizontal agreement among competitors that also happens to include a non-competitor.   

The next question that must be asked concerns the ancillary restraints doctrine.  Even a horizontal restraint is not subject to the per se rule if it is “subordinate and collateral to a separate, legitimate transaction,” and reasonably necessary to “make the main transaction more effective in accomplishing its purpose.”  Whether a particular restraint meets this standard depends on the facts of the case.  For example, in a franchise case, the ancillary restraints question turns on the relationship between the no-poach agreement and the franchise system, particularly its promotion of interbrand competition.

All told, then, there are two ways for a no-poach agreement to be subject to the rule of reason and not the per se rule: verticality and ancillarity.  But a final question still remains: should such a restraint be analyzed under the full rule of reason or the quick look doctrine?  The quick look doctrine is a rarely-applicable version of the rule of reason analysis that is appropriate only when “an observer with even a rudimentary understanding of economics could conclude that the arrangement in question would have an anticompetitive effect.”  The doctrine recently has been criticized and its exact contours have been called amorphous by Professor Hovenkamp.  But one thing that is clear is that the quick look doctrine is not supposed to be an intermediate third category of analysis, as if it were some kind of compromise between the per se rule and the rule of reason, or invoked whenever a court thinks, for example, that a particular restraint is too hard to classify.  To the extent that recent court decisions embody that reasoning, they deserve further scrutiny.

The foregoing questions are those that should be asked when considering no-poach agreements.  I have not attempted to answer all these questions, even in hypotheticals, because the answers depend on the facts, and they vary.  In addition, delving into the particular facts of every no-poach case is not in the interest of the United States.  What is in our interest is that courts administer the antitrust laws properly with the rigor and nuance that the foregoing questions would entail.

I have covered a great deal of ground in this speech.  I explained what we are trying to do with our amicus program: that it is the interest of the United States that courts properly apply the antitrust laws.  So we watch cases, taking on the role of tireless advocate for the right answer, whether it benefits plaintiffs, defendants, workers, employers, or others.  I also have explained a nuanced and complex framework for the no-poach space, based on the historical relationship between the antitrust laws and the labor markets.  We do not have all of the answers, but we think these are the right questions: whether the agreeing entities can act concertedly, whether they are horizontal competitors, whether their agreement is ancillary to a larger arrangement, and whether they merit a quick look.

Through this discussion, I have endeavored to both argue and demonstrate that we are motivated in both instances by our devotion to the rule of law.  Our only job is to see that justice is done.

Thank you.

Updated March 8, 2019