Acting Assistant Attorney General Renata Hesse of the Antitrust Division Delivers Opening Remarks at 2016 Global Antitrust Enforcement Symposium

And Never the Twain Shall Meet? 

Connecting Popular and Professional Visions for Antitrust Enforcement

Good morning.  I am honored to have the occasion to address this impressive group of people interested in antitrust—practitioners, scholars, and observers alike.  And I am especially excited about this opportunity because it is my first major speaking engagement since I became Acting Assistant Attorney General at the Antitrust Division this past July.  My predecessors – Bill Baer, Joe Wayland, Sharis Pozen, and Christine Varney – did tremendous work on behalf of the American public, and theirs are very tough acts to follow.  All of us, and more importantly all of the exceptionally dedicated professionals who devote their careers to the Division, are working, every day, to keep antitrust enforcement going strong.

Antitrust is making headlines again, and I don’t just mean in antitrust publications.  It is, as it was at its inception, the stuff of popular imagination.  Republicans and Democrats in the House and Senate have been encouraging the federal agencies to be vigilant in their merger-enforcement and competition-advocacy roles.^[1]  President Obama issued an executive order requiring agencies across the federal government to consider “specific actions” to promote competition.^[2]  And numerous op-eds and think-tank reports this year – from folks across the political and policy spectrums – have sounded the alarm on increasing concentration, have called for stepped up enforcement of the antitrust laws, and have implored federal and state agencies to tailor their regulations and policies so as to promote competition by lowering barriers to entry and expansion.^[3]  At bottom, these diverse voices agree on the basic proposition that it is unfair to allow companies to grab unearned monopoly power over markets that they can wield at the expense of consumers, workers, and would-be competitors.  By and large, I think this increased public interest in antitrust and competition is a good thing.  It is good for the public – because antitrust enforcement promotes the interests of the public over the power of the few – and it is also good for antitrust – because it keeps enforcers focused on the ultimate goal of antitrust, economic fairness.

We, as antitrust enforcers, have been engaging in the general public’s dialogue about antitrust by litigating more cases in the public forum of the courthouse rather than settling them in the professional halls of the antitrust agencies.  Accordingly, professionals and the public are moving more toward a consensus vision of antitrust focused on protecting competition and the fairness inherent in it.  When we try cases in court, we are forced to present our proof – even where it includes sophisticated economic analyses – in a straightforward narrative that is compelling to a lay judge or jury.  Successful litigation, therefore, requires focusing on the value of protecting competition – the value that motivates popular interest in antitrust today and motivated the passage of the Sherman and Clayton Acts years ago.

Popular and Professional Views of Antitrust – conflict or conciliation?

The rising popular enthusiasm for antitrust also reveals, however, some tension between the views of the general public and the views of those who practice antitrust professionally.  For the last several decades of the twentieth century, antitrust functioned largely as a practice of experts – economists and economics-savvy lawyers – confined to the halls and conference rooms of the expert agencies.  This happened for a number of reasons.  The conservative leaning “Chicago School” made economic efficiency synonymous with the goals of antitrust in the 1970s, which incorporated theoretical economics into mainstream antitrust scholarship and practice.  Later, more centrist or left-leaning post-Chicago and Harvard School scholars showed that sophisticated empirical and theoretical economics tools can be used to support more aggressive enforcement agendas.  Together, these developments resulted in many technical discussions about what impact a business practice will have on consumer welfare mathematically measured – involving supply and demand curves, triangles representing “dead weight loss,” and so on.   But that sort of conversation is one that resonates very little – if at all – with those engaged in the straightforward, popular dialogue about the dangers of increasing corporate concentration.  The language of economic theory does not sound like the language of economic fairness that is the raw material for most popular discussions about competition and antitrust. 

Despite this divergence in language, I don’t believe the tools of professional antitrust practice – when used correctly – conflict with the popular perspective.  The ultimate concern of antitrust law has always been protecting competition at all levels of the economy.  Animating the beliefs of ordinary Americans who demand vigorous antitrust enforcement are the value of fairness and the belief that properly functioning competitive markets are themselves fair.  To say it another way, competition is fair because it gives a chance to the small business owner to succeed in her business venture, because it delivers lower prices to consumers, and because it drives the innovation that improves products, business processes, and more.  Competition among employers to attract workers is fair because it yields higher wages, better benefits, and safer working conditions.  In general, competition is fair because it distributes these rewards broadly to participants in the economy.  But when companies harm competition – choking off competition or agreeing with rivals not to compete – they infect the economy with unfairness by accumulating power that the few can wield at the expense of the broader American public. 

In my view, the tools of economics simply provide enforcers with a better means of detecting situations where companies and individuals have subverted – or threaten to subvert – the competitive process.  It is our job as public servants to explain to the public why we do what we do; for example, when we use economics tools with obscure names like “Herfindahl-Hirschman Index” or “Gross Upward Pricing Pressure Index,” we are simply measuring intuitive phenomena like the concentration of economic power or the tendency of mergers to reduce competitive pressures that keep prices down.  At times, we have left these concepts largely unexplained and allowed expert practice to remain isolated from popular relevance.  But I believe strongly that in the last decade we have been reducing the gap between expert and popular antitrust as we have been litigating more and more cases, forcing us to explain our claims of harmed competition to lay judges and juries who must determine the rightness of our causes.  Antitrust is too important to be left solely in the hands of antitrust experts.  I’ll return to that trend at the end of this speech.

That is not to say that there is no contradiction between popular and professional views of antitrust.  Each camp contains a diversity of views, and some variants are more compatible than others.  For example, there are some versions of the popular view that reflexively conclude that “big is bad.”    This view sees the rise of large corporations – and the effects they have on communities, culture, and politics – as a proper focus of antitrust enforcement.

The big-is-bad view takes aim, as I see it, at the wrong target for antitrust enforcers.  First of all, many of the measures of concentration have from an antitrust perspective the wrong measure of bigness in mind.  Antitrust is concerned with a situation where a firm or firms are large enough in proportion to the rest of the market – and thus face too little competition – that they can raise prices alone or take actions that prevent new competition from undercutting high prices.  Such firms, we say, have “market power.”  Many concentration studies simply do not measure market power in this way.  For example, an Economist magazine study divided the entire U.S. economy into 900 some sectors and measured the change in concentration from 1997 to 2012, finding “[t]wo-thirds of them became more concentrated between 1997 and 2012” during this period.^[4]  Other recent studies have employed similar approaches.  Some people have suggested that greater antitrust enforcement should be used to combat these trends.

A simple example, however, illuminates why these measures are prone to error.  In the complaint the DOJ filed in 2011 to block the proposed merger of AT&T and T-Mobile, we alleged that the most important competition for consumer cellular telephone services took place in local markets among providers that had assembled nationwide networks.^[5]  Under this framework, the proposed merger between AT&T and T-Mobile threatened competition between two of the most important national competitors.  By contrast, a merger among multiple non-overlapping regional networks might enhance competition by enabling a regional network to become more relevant to cellular customers by becoming a national network capable of competing with AT&T, Verizon, Sprint, and T-Mobile (the only competitors with national networks).  But the measure employed in the Economist magazine study I mentioned would treat the merger among regional networks – just as it would an AT&T/T-Mobile merger – as an increase in concentration.

Of course, I don’t mean to suggest that mergers between non-overlapping regional telecommunications networks can never present competitive concerns.  When Comcast and Time Warner Cable proposed merging their non-overlapping cable and internet networks, we determined it was likely that their combination would dramatically increase Comcast’s power in nationwide markets in which content providers seek to distribute their shows.  The moral of this story, I suppose, is that there is no substitute for the hard-earned understanding that comes with the detailed work of market definition that antitrust enforcers do on a case-by-case basis involving a wide range of qualitative and quantitative evidence.  Simple, quantitative measures used in industry concentration studies can be a useful way to bring attention to a potential problem, but they often shade over important details that make them difficult to use for antitrust enforcement or competition policy.

Second, even when we have the right measure of bigness in mind – that is, market power – market power by itself is not the focus of antitrust.  This does not mean it is a good thing when one or a few firms dominate a market.  A monopoly may charge higher prices.  And a market with few competitors may be more conducive to coordination on higher prices or other non-competitive outcomes.  But the antitrust laws do not target concentration as such; nor is it clear they should.  Markets can become concentrated for benign reasons.  For example, where fixed costs are high, new entry may be sufficiently unattractive to all but a handful of firms.  Alternatively, network effects could lead one firm to gain all or the lion’s share of a market for a time.  Moreover, so long as large firms do not wield their power to exclude new rivals, charging higher prices could accelerate the process by which new firms challenge the dominant firms of today.  In other words, antitrust enforcers don’t go after firms that become large just because they are good at competing.  So long as competitive processes are not subverted, new firms can rise to displace today’s winners.  That is how competition works.  We are concerned with situations where market power is achieved or protected by anticompetitive means.

Just as some popular views of antitrust miss the mark somewhat, so do some of the expert views.  With the rise of precise tools of economics toward the end of the twentieth century, some practitioners and theorists began to mistake sufficient proof for necessary proof.  To put a finer point on it, the tools of economics have allowed us in some circumstances (where the right data and enough time exist) to estimate a reasonably precise impact on consumer welfare of certain business practices.  Some commentators have accordingly suggested that the antitrust laws should judge all practices by their impact on the welfare of downstream consumers, as measured by price and output effects in downstream markets.  But, although we believe competition maximizes consumer welfare, the ultimate standard by which we judge practices is their effect on competition, not on consumer welfare.  It is certainly relevant when a merger will lead to higher prices and reduced output because these results are hallmarks of reduced competition.  But the law instructs us to examine whether a merger may substantially lessen competition and that means we must sometimes look to other evidence of harm to competition.  For example, let’s say we have documents showing that two technology firms have been investing in research and development in a race against each other to develop new products, but the two plan to merge and cut R&D expenditures substantially due to reduced competitive pressure.  That is a loss of important competition even if it is difficult to measure its exact consumer impact. 

Similarly, this focus on competition makes it clear that we are just as concerned with lost competition among upstream input suppliers as we are with lost competition among sellers of finished goods downstream.  As the D.C. Circuit reminded us in its seminal “baby-foods” merger decision, “no court has ever held that a reduction in competition for wholesale purchasers is not relevant unless the plaintiff can prove impact at the consumer level.”^[6]  What we are concerned with is mergers and other practices that create market power—giving individuals or companies power over prices and output rather than having competition drive the setting of prices and output.  Under this view, it is the risk of creating power over price that makes a merger illegal; we need not know exactly where the price will be set.  In any case, competition is impaired.  And that is true at any level of the economy—upstream, downstream, midstream.  The antitrust laws protect competition throughout.

Competition as Fair Process – the shared language of popular and expert antitrust

This view of protecting competition throughout the economy finds its ultimate support in the roots of our antitrust laws.  When Senator Sherman promoted passage of the country’s first antitrust law (which still bears his name), he and other supporters of his bill focused on targeting business practices – especially the “trusts” that had risen to dominate many industries at the time – that are likely to harm the processes of competition by which economic rewards are normally distributed.  Harm to competition, and the injury that it tended to wreak not just on consumers, but also on workers, farmers, and would-be rivals, was their concern.  Senator Sherman approvingly quoted the following depiction of trusts: “They operate with a double-edged sword. They increase beyond reason the cost of the necessaries of life and business, and they decrease the cost of the raw material, the farm products of the country.”^[7]

With this concern for the protection of competition in mind, the antitrust laws are an effective weapon to protect the economy against the threats posed by increased concentration, which motivates many of the popular calls for vigorous antitrust enforcement.  An easy way for firms to sidestep the process of competition and enhance their power artificially is to acquire a significant competitor and thereby eliminate the competition between them.  But the antitrust laws give us the means to stop those mergers: Section 7 of the Clayton Act declares them illegal if their effect may be to lessen competition substantially in any market.^[8]  The antitrust laws protect against the harms of increased concentration in all varieties of markets.

One bedrock tool for protecting against anticompetitive mergers has been recognized by the Supreme Court for over fifty years.  In Philadelphia National Bank, decided in 1963, the Supreme Court announced that “a merger which produces a firm controlling an undue percentage share of the relevant market, and results in a significant increase in the concentration of firms in that market is so inherently likely to lessen competition substantially” that the law will presume it unlawful.^[9]  But in the wake of the Chicago School’s influence, antitrust commentators started to call into question the validity of this common-sense presumption, believing that economic theory showed that mergers tended to be beneficial or, if they resulted in harm, that harm was fleeting.  Those skeptics demanded more detailed proof of consumer harm in place of the presumption.  More recent economics studies, however, have given new life to the old presumption—in several ways.  First, we are learning more and more that mergers among substantial competitors tend to lead to higher prices.^[10]  Second, economists have been finding that mergers often fail to deliver on the gains their proponents sought to achieve.^[11]  Taking these insights together, we should be skeptical of the claim that mergers among substantial competitors are beneficial.  The law – which builds this skepticism into it – provides an excellent tool for protecting competition from large, horizontal mergers.

For an illustration of why the Philadelphia National Bank presumption makes sense in practice, I want to return to the Division’s investigation of the proposed Comcast/Time Warner Cable merger.  The combination of those two large cable providers would have made one firm the provider of internet service to nearly 60% of the households that purchase high-speed broadband internet.  That would have substantially increased Comcast’s bargaining leverage, and allowed it to charge significantly more to large providers of internet content that wanted to exchange traffic directly with Comcast’s network.  Moreover, because Comcast also sells cable television to many of the same homes to which it provides internet, its control over internet service would have given it reason to harm companies, like Netflix or Amazon Prime, that use those high-speed connections to challenge Comcast’s cable business.  We could not predict precisely how Comcast might have used its new-found power; it could have increased prices broadly to providers of internet content or more specifically tried to harm its online video rivals.  But the combination of such substantial competitors gave us (and the Federal Communications Commission) sufficient reason for concern—so much power could not be entrusted to one company.  As an early common law case brought against the Standard Oil Company noted, “[e]xperience shows that it is not wise to trust human cupidity where it has the opportunity to aggrandize itself at the expense of others.”^[12] 

In other words, measuring the effect on price and output is not required in every merger case.  This is not because we are unconcerned with price increases and output reductions that harm consumers—far from it.  But we best protect consumers (and others in the economy) by stopping anticompetitive practices, including mergers among substantial competitors, that experience and evidence – including company documents and customer testimony – suggest are likely to harm competition.  If we required particularized and quantified proof of consumer harm in every case, we would simply make it more difficult to stop harmful conduct.  That’s the lesson of Comcast/Time Warner Cable.

That said, in the more routine horizontal merger case we almost always do introduce proof of consumer harm where it exists – along with demonstrating likely reductions in quality and slowing of innovation – because we want to present a full account of the anticompetitive effects of challenged practices to the finder of fact.  Moreover, we rarely rely solely on the presumption: even where price and output effects cannot be quantified, we usually have some concrete evidence such as documents or lay testimony showing that competition between the merging firms is important and will be lost with a merger.  And even where the presumption is satisfied, we would not bring a case where a fuller picture of the evidence has shown us that competition is unlikely to be affected significantly by a merger.

Some commentators have suggested, however, that certain classes of cases are special and do require proof of demonstrable consumer harm.  For example, some argue that where the claim is that a merger creates “monopsony” power – that is, power over the price of some input or over the wages of workers – the plaintiff bears the added burden of demonstrating harm not only to the input market but also to consumers downstream.^[13]  But there is nothing in the Sherman or Clayton Acts that would justify this limitation.  As I said before, the legislative history of the Sherman Act makes it clear that the antitrust laws were intended to benefit participants in the American economy broadly—not just in their capacity as consumers of goods and services.  Senator Sherman said in promoting the Act that one of the problems with monopoly is that: “[i]t commands the price of labor without fear of strikes, for in its field it allows no competitors.”^[14]  So a merger that gives a company the power to depress wages or salaries or to reduce the prices it pays for inputs is illegal whether or not it also gives that company the power to increase prices downstream.  The Supreme Court’s Weyerhaueser decision illustrates this point.  That case involved an alleged attempt by a sawmill at “monopsonizing” a local market for the purchase of logs through predatorily overbidding.  The Court pointed out that the case involved no allegation – and it presented no likelihood – of the defendant’s gaining power over downstream finished lumber sales (a national market) even had it successfully monopsonized the local market for logs, without suggesting that presented any problem for the plaintiff.^[15]

Much of this talk has focused on mergers, but firms can also sidestep the competitive process by simply agreeing not to compete.  For much the same reason that presumptions in our merger laws exist, the antitrust laws impose no requirement of proving harm to ultimate consumers in cases of agreements among competitors.  Our successful case against Apple in the e-books matter provides a great illustration of this framework.  The Division prevailed at trial and proved that Apple conspired with publishers to raise the prices of e-books.  In the appeal, the Second Circuit rightly rejected Apple’s argument that the conspiracy was justified to facilitate its entry into the market as an e-book distributor.^[16]  According to Apple’s view, allowing competitors to band together to increase prices could help provide an alternative to Amazon’s distribution platform.  But the invalidity of this sort of defense has been long-established.  The Supreme Court in Philadelphia National Bank explained that a practice that harms competition “is not saved because, on some ultimate reckoning of social or economic debits and credits, it may be deemed beneficial.”^[17]  The antitrust laws favor competition; so it is no defense that a practice replaces competition with something else that could enhance consumer welfare.

The same principles apply to monopolists.  When companies subvert competition to gain or maintain monopoly positions, that conduct is illegal whether or not consumer harm in the form of price effects or output reductions can be demonstrated.  Section 2 of the Sherman Act makes it unlawful to “monopolize, or attempt to monopolize, or combine or conspire with any other person or persons, to monopolize any part of the trade or commerce among the several States.”^[18]  Under more than a century of case law, proving illegal monopolization means showing both that a company commands a monopoly position in some market and that it reached that position through anticompetitive acts.  Monopolists wield great power in the economy.  That they might use that power for some social good is no defense to the creation of monopoly power through exclusionary acts.  This principle was well understood at the time the Sherman Act was enacted.  In a common law monopoly case against the Standard Oil Company, the district judge wrote:  “It may be true that [Standard Oil] has improved the quality and cheapened the costs of petroleum and its products to the consumer. But such is not one of the usual or general results of a monopoly; and it is the policy of the law to regard, not what may, but what usually, happens.”^[19]  That was wise in 1892, and it remains so today. 

Sometimes firms get big through normal competitive means but then turn around and try to protect those positions by harming competition to fend off new threats.  The antitrust laws reach that conduct too.  The Department of Justice’s famous monopoly maintenance case against Microsoft confirms that there is no requirement of demonstrable consumer harm in such a case.  There Microsoft targeted a threat that hadn’t yet, and may never have, arisen—a rival operating system that could have grown up around applications that ran off middleware products such as Java and Netscape Navigator.^[20]  Microsoft’s actions simply made it more likely that it would retain its monopoly position.  Because that subverted the possibility that a rival could develop to compete against Microsoft, Microsoft’s actions stifled the competitive process, and that was enough for the government to prove a violation.

The Federal Trade Commission’s case against Intel a decade later similarly shows how dominant firms can cut off the normal mechanisms of competition to maintain dominance.  In that case, the FTC alleged that Intel violated Section 5 of the FTC Act by maintaining its monopoly in central processing units (or CPUs) through a variety of payments and penalties (including loyalty or market-share discounts) to computer manufacturers to induce them not to purchase products from Intel’s rivals such as AMD and Via Technologies.^[21]  When a monopolist pays customers to disfavor its rivals and punishes those customers who nevertheless do business with a rival, that does not look like the monopolist is competing with its rivals on the merits of their products.  Because these actions served only to foreclose competition from rival producers of CPUs, these actions distorted the competitive process.

Antitrust Litigation – the process through which experts and lay enthusiasts converse

As a matter of process, the way that antitrust as expert practice has become more engaged with the general public’s enthusiasm for antitrust, and to make itself more relevant to the public, is to change the way disputes are resolved.  We have been moving from a quasi-regulatory system – in which disputes are typically resolved by settlement crafted largely within the halls of the agencies – to a litigation mode – where disputes are more often resolved by lay third-party arbiters.  In the waning decades of the twentieth century, there was a tendency for antitrust enforcers to want to compromise with parties in antitrust disputes.  With the unfounded Chicago-School presumption that mergers often benefit competition, many have wanted to save mergers by trimming off their anticompetitive effects with minimal divestitures or with behavioral limitations.  Antitrust enforcers at the Antitrust Division and the FTC have become justifiably more skeptical about the promise of procompetitive benefits of mergers and of the likelihood that remedies solve the competitive concerns.  As a result, we are more and more litigating to challenge mergers we see as fundamentally problematic and difficult, if not impossible, to fix.

In 2011, the Department of Justice filed a complaint to challenge the proposed merger between tax software providers H&R Block and TaxAct.  This challenge was significant in that it would lead to the Division’s first trial (and trial victory) in over half a decade.  It also illustrates how a successful merger case should involve a range of quantitative and qualitative evidence that reinforces the same message of likely anticompetitive effects.  The Division won that challenge,^[22] and it later prevailed in a litigated trial against Bazaarvoice.  The Bazaarvoice case, in which the Division successfully sued to unwind a consummated merger between the top two providers of online-ratings-and-reviews platforms, provides a great example of a trial that packaged sophisticated economics evidence into a compelling narrative.  As in most merger cases, we put on expert testimony to define the relevant market using the technical “hypothetical monopolist test,” and our expert found that the merging parties faced little competition from remaining rivals in that market.  But the centerpiece of our proof was the companies’ own documents, which showed that the merger was planned to “[e]liminate [Bazaarvoice’s] primary competitor” and “reduc[e] comparative pricing pressure.”^[23]

In total, the Division has been in civil trials for 75 days during the Obama Administration—as compared to only 27 days during the prior administration.  Showing to the public that we are prepared to litigate has paid off: in this administration, a total of 40 mergers have been blocked by court order or wholly abandoned by the merging companies in the face of our investigation, a stark increase from 16 in the prior administration.  These efforts have protected competition in many important sectors of the economy—cellular phone service, computer technology, cable and internet service, cooking appliances, canned foods, oil and natural gas, tax software, and online-business-review platforms, to name just a few.  The FTC has been litigating more too, especially to protect our economy against increased consolidation and anticompetitive practices among hospitals and pharmaceutical companies.  Our most recent efforts in merger litigation resulted in the abandonment of the GE/Electrolux appliances merger and of the Halliburton/Baker Hughes oilfield services merger.  But our work continues: we are currently in active litigation in a handful of civil cases, including challenging the two largest health insurance mergers ever. 

Litigating more means changing the way we talk about our disputes.  When cases are settled, the conversation hardly leaves the corridors of agencies and law firms, and so it can stay at the rarified level of economics tools such as HHIs, cross-price elasticities, and GUPPIs.  When we litigate, we put our dispute before a neutral lay arbiter—a judge or a jury.  That means we have to tell a compelling and coherent story about why certain business practices are harming competition and thereby participants in the economy.  The proof will be varied, and it will almost always include sophisticated expert presentation of economics evidence – theoretical or empirical or both – as part of the evidence.  But that must be packaged with qualitative evidence that confirms in a palpable and intuitive way the story told through the numbers of the expert.  And ultimately the plaintiff’s story should highlight the moral underpinnings of the antitrust laws—fighting against the unfairness of concentrated economic power profiting at the expense of consumers, suppliers, or competitors who could challenge the defendant’s dominance.

So I think the way that professionals practice antitrust is already changing.  As popular enthusiasm for antitrust has surged, so has expert practice moved from the quasi-regulatory work of negotiating consent decrees to the mode of challenging anticompetitive mergers and restraints of trade in litigation.  As that move has occurred, our language has shifted as well.  Rather than focusing on measuring consumer welfare in an academic fashion, we are looking more broadly at the effects of business practices on competition, and that is getting us back to what the drafters of the Sherman and Clayton Acts intended.  When we explain our cases to generalist judges in public proceedings, we necessarily engage in the popular conversation about the purposes of antitrust.  By focusing our enforcement efforts at harm to the competitive process wherever it occurs, we target the basic problem animating the popular enthusiasm for antitrust—attempts to obtain or keep economic power unfairly by sidestepping the normal competitive process to gain profits at the expense of the public.  These efforts benefit consumers, because competition leads to lower prices and greater quality goods and services, but they also benefit workers, whose wages won’t be driven down by dominant employers with the power to dictate terms of employment.  And they also benefit small business owners, who are given a fair shot in competing against larger, more established companies.  Our efforts protect competition, and that helps keep the economy fair.

(A)AAG Hesse 2016 Global Antitrust Enforcement Symposium Remarks with Footnotes