Apr 19, 2022

Antitrust Policy and its Different Perspectives

Where do the Antitrust professionals agree and disagree?
Alden Abbott Senior Research Fellow, Joshua Wright, Doug Melamed
Mercatus Policy Download

Discussions of public policy featuring Mercatus scholars.

Contact us
To speak with a scholar or learn more on this topic, visit our contact page.

On this episode, Alden Abbott, a Senior Research Fellow here at the Mercatus Center talks to Douglas Melamed and Joshua Wright about their different perspectives on pressing discussions around antitrust policy. If you would like to connect with a scholar featured on this episode, please email the Mercatus Outreach team at mercatusoutreach@mercatus.gmu.edu

Note: While transcripts are lightly edited, they are not rigorously proofed for accuracy. If you notice an error, please email the Mercatus Outreach team

ALDEN ABBOTT: Welcome. I'm Alden Abbott, senior research fellow at the Mercatus Center at George Mason University, and former general counsel of the Federal Trade Commission. My views on breaking antitrust developments are found my free monthly newsletter online, The Competition Corner. Just enter competition corner and Alden Abbott, that's A-L-D-E-N A-B-B-O-T-T in your search engine to find and subscribe to The Competition Corner.

Today I'm honored to host a conversation on antitrust policy between two of the nation's leading antitrust scholars, Douglas Melamed, professor of the practice of law at Stanford University Law School, and former acting assistant attorney general for antitrust under President Clinton, and Joshua Wright, university professor of law at George Mason University Scalia Law School and former Federal Trade Commissioner during the Obama administration.

As the Moderator, I will proceed by asking a series of questions. Each question will be directed to one of our two scholars, and the second scholar will be invited to respond. My goal is to highlight differences and similarities and perspectives on antitrust policy to a friendly dialogue between two of the most experienced practitioners and best thinkers in the field who are known to hold somewhat contrasting views. Our eminent speakers have agreed to let me address them by their first names. Let's begin. Doug, the first question is for you.

There's been a lot of talk in recent years about neo-Brandeisian antitrust, which features a “big is bad” philosophy and de-emphasizes consumer welfare as a guiding star of antitrust enforcement. You have emphasized that you support the consumer welfare standard which US courts have accepted for 40 years. Nevertheless, you and UC Berkeley economics professor emeritus, Richard Gilbert, recently wrote, "Where we agree with the New Brandeisians is that antitrust enforcement has been too weak, largely because many courts appear to have accepted the idea that the costs of too-aggressive enforcement are greater than the costs of too-weak enforcement."

"Encouraging courts to include innovation concerns in their reviews of unilateral conduct and mergers in high-techn industries could alter this calculus in favor of more aggressive antitrust enforcement. In this respect, it is a movement toward the New Brandeisians, although not a bridge that connects their movement to the views of those who support more traditional, economically-focused antitrust enforcement." That's a mouthful, but in short, could you elaborate on your statement that antitrust enforcement has been too weak, and can you briefly explain how innovation can better be taken into account in judicial antitrust reviews?

DOUGLAS MELAMED: Sure. Let me add a third question if I can do that briefly, Alden. That is the issue about the consumer welfare standard with the New Brandeisians. The fundamental issue, the core issue is whether antitrust law is going to remain singularly focused on economic welfare, which is what I think the end of the day the consumer welfare standard means, or whether they will be incorporated into that in addition to or in lieu of consumer welfare or economic welfare other objectives, inequality and so forth. I strongly believe that antitrust, for a number of reasons we can get to later, should stay in flame and focus on economic welfare.

The notion that antitrust law, in my view, needs to become a little more aggressive, it means that I think it's been too deferential to defendants, too laissez-faire oriented with respect to its objective of maximizing economic welfare. I think that's been true. I think the data suggests that there's been some increases in concentration and price-cost margins and the like, they're not unambiguous, I concede. More than that, I look at it at a micro-level and I look at what might be called micro rules, the rules about predatory pricing, the rules about refusals to deal, some of the recent decisions, the American Express decision, I think they're wrong.

I think many of them, not all of them, but some of them are wrong. I think those that are wrong tend to be wrong because of a notion that's been repeated over and over and taken to excess that we should try harder to avoid false positives, false convictions as it were, than to avoid false negatives. I think it's going too far. The role of innovation is really a little orthogonal to that, although I think what we were trying to say is if antitrust gets incrementally more aggressive, which is all we're proposing, it might take a little of this political steam out of the New Brandeisians because some of the concerns they have would be marginally ameliorated.

The innovation idea is this. We all have known for a long while, that innovation is much more important to economic welfare than avoiding deadweight loss. Innovation is often recited in antitrust documents, but it's almost always a make way – rarely ever specifically analyzed. Our thesis, Rich Gilbert and mine in the article to which you referred, is that if innovation effects were assessed directly they could, in appropriate cases, warrant sometimes both interventions and sometimes justifications and defenses that are not recognized by a more traditional antitrust focus simply on price or quality-adjusted prices.

The reason for that difference is that most antitrust comes down to two issues. Was there any competitive conduct? Did it create market power? Market power by definition is the ability profitably to increase prices. If you increase market power, you have an ability and an incentive to increase price, so price harms can be presumed from a mere increase in market power in a product market.

Innovation is a more complicated story, you have to look at two different markets. You might call it the market for innovation resources and the product market and some more complicated analyses. If innovation affects, whether as justifications or as a basis for liability, are going to count, they're going to have to count by a somewhat different and more complicated analysis than that which we traditionally use for price effects.

ABBOTT: Okay. Innovation, and I think that's probably a lot of people of all philosophies in antitrust perhaps agree innovation is terribly important. Josh, do you have any comments on what Doug has said?

JOSHUA WRIGHT: Always. Sure. It's a pleasure to be here. There's nobody I enjoy discussing these issues more than Doug. I think we're very thoughtful about the areas we disagree with, but we're going to start on something we agree with and we'll find some fun things to disagree about too. I think I will comment in the same order as Doug addressed the question to leave the specific for innovation toward the end. I absolutely agree that when we talk about the neo-Brandeisan influence on antitrust, if there will be any on the ground, is about a movement to a standard that asks antitrust to have multiple goals, multiple objectives rather than a singular focus on economic welfare.

That's, I think, the $64,000 question about whether the neo-Brandeisian’s impact the antitrust world at the end of the day will be are we left with a standard that has a singular economic welfare focus or not? I think where we start on agreement is Doug and I both agree that antitrust is best served by a singular focus upon economic welfare. No disagreement there.

I suspect we'll find lots of ways to talk about this disagreement throughout our conversation, but on the question about the quality of evidence that antitrust enforcement has been too lax, I think we land in a very different place. I for one would like to see the agencies bring all of the good cases that they can bring. By good case I mean, and I think Doug and I agree on what a good case is, a case where there's conduct where the defendant acquires market power by agreement, by conduct, by merger, to the detriment of competition and consumers.

I think the quality of evidence that suggests systematic laxity on the part of the antitrust enforcement regime in the US is not very persuasive. Doug's got a list of cases he doesn't like. He doesn't like the decision in Amex, He doesn't like the decision in, and I think I've read or we've talked about Qualcomm before. I've got a list of cases where the plaintiff won where I thought there were bad cases and bad decisions too. We can do our list all day, and maybe we'll do some of that here, but in terms of the quality of evidence of systematic underenforcement, most people point to industry-level sector data suggesting sector level concentration is going up. Anybody who follows antitrust knows that's neither here nor there for the question of whether there has been increasing or decreasing competition.

Further than that, most of the more recent evidence suggests a really interesting economic phenomenon that is sector level concentration up, local concentration down, increase in output and that margin increased, that everybody's second favorite thing to point to for systematic evidence of a failure in antitrust enforcement. Margins can go up for two reasons, the price goes up or marginal cost goes down, or both.

Unsurprisingly, I think the best evidence is a mixed bag on those questions. Nathan Miller has a great paper that looks at 400 product markets as well. When everybody points at those papers with the big increasing markups, about 70% of that comes from a reduction in marginal cost, mostly from the already large firms entering new markets or new geographies. That's an output-expanding thing. It has all sorts of interesting impacts on the economy. We get an increase in firm size, you get distributional effects, maybe impacts on income inequality.

The premise of the argument that the agencies are asleep behind the wheel and should be more aggressive because look at all the market power that's accumulated throughout the economy – I reject the premise of that question and we can talk more about the data. That doesn't mean that I don't fully agree with Doug's proposition that we ought to find good cases to bring, and there are places where the agencies have shied away from good cases. Doug likes the innovation cases, I would like to see the agencies put all of that energy into being more creative in the way that we attack public restraints.

Where I think everybody agrees are pernicious and long-standing sources of monopoly power where there's not so much disagreement about the effects. I would like to see more of that. I would like to see more criminal enforcement than we have now and more creative thinking about sanctions. Long story short, I think plenty that we agree on a broad level, maybe disagree over the current state of affairs and how much of a crisis there is in markets themselves, or in the courts or at the agencies. Plenty to improve on, and just a word about innovation.

I'm all for the idea that to the extent we can do better thinking, more clear thinking, and measurement with respect to directly assessing innovation symmetrically on the plaintiff’s prima facie burden and in terms of defenses, I am all for it. I agree that innovation, and I'd probably add entry to the list, are two places where antitrust analysis is the most hand-wavy on both sides of the analysis, both when plaintiffs invoke barriers to entry or harm to innovation and on the defense side. Those are two places where I think the antitrust toolkit and rules can be improved greatly. Whether it means more enforcement or less, I don't know but I think those are two places where I really wish more of the intellectual energy in our field was going into solving those problems and a little less into debates over whether we should bring back the antitrust regime of the '60s.

ABBOTT: Thanks, Josh. Next question is for you. In his 2021 executive order on competition, President Biden stated, "We're now 40 years into the experiment of letting giant corporations accumulate more and more power, and what have we gotten for it? Less growth, weakened investment, fewer small businesses." He then outlined plans for more aggressive antitrust enforcement, as well as government investigations into competitive problems in a host of industries, for example railroads, pharma, many more. What did the executive order get wrong, and what, if anything, did it get right? Then I will ask Doug, of course, for his reaction.

WRIGHT: The executive order, like most executive orders that do things with competition, says a lot, but without much analysis or thinking. There are things to like and things not to like in the executive order. You'd listed four. I think it named 70-something industries by name and says the agency should look or think hard about changing rules or et cetera. What comes out of the executive order, I think, as an aspirational document, I'm not sure. Antitrust is moved by winning and losing real cases in real courts in front of real Article III Judges. At least on a good day, that's the way antitrust moves.

We'll see what happens in the courts when these agencies bring cases or try to promulgate rules, or when those other industry sector level regulators try to do the same. As an aspirational document, look, there was a lot of political pressure on the Biden administration to walk away from some of the commitments the Obama administration made, and Trump after that and administrations prior on occupational licensing, and some of the state action type that are public restraints. I think the politics on that have changed. For example, I think if you lined up modern progressive antitrust thinkers with the neo-Brandeisian, the neo-Brandeisian reject most of the using antitrust to attack state and local restraints of trade, and actively support antitrust immunity for state boards and the like.

That part of their view has not gotten a lot of attention. Not sure exactly why, but the politics on those issues have changed. I think it was a good thing and kudos to the Biden administration for sticking to their guns on addressing occupational licensing and public restraints as a real problem. I think it's a real problem that affects a lot of Americans. That was a thing in that document I liked. The rest of it, I don't know. It's hard for me to get excited about what an executive order says about a preamble about loss of competition in the failed antitrust, but most of that's wrong.

What it means at the end of the day is more or less nothing. The agencies have a mandate from the administration to go out and bring the case they can find. So far, they've not done a ton. They certainly haven't done more than the administration prior. They haven't had a fifth commissioner, at least at one of those agencies and so the jury's still out on what they actually do. I'm not a big believer that the EO has a big impact other than telling those agencies symbolically that they should be aggressive and I suspect they were going to do that anyway.

ABBOTT: Doug, your thoughts.

MELAMED: I don't think I fundamentally disagree with Josh about that. I think there's a little more on the upside to say from the order just because it was so sweeping and so strongly worded. I don't recall an executive order of the breadth and strength of that one on competition in a broad way like this.

I do like very much what I guess has been called the all of government approach because it's extremely important that we focus on the ability of other governmental regulatory and executive agencies to ameliorate anticompetitive restraints and promote competition in their industries, and to focus also on the extent, as Josh was saying, of the regulations at every level of government that restrict competition, and to take a hard look at whether we can get rid of those in the name of competition. I think the executive order was good in terms of pressing in those directions.

On antitrust enforcement itself, it did say we have to be more aggressive. All of this rhetoric you sometimes see throughout the economists, throughout all the Democrats from the Obama administration is nonsense. My point is that, at least my concern where I differ from Josh, is that I think there are aspects of antitrust doctrine, burdens of proof and so-called micro rules and aspects of merger enforcement that have been too cautious and have resulted in bad outcomes in those particular markets cases and sectors.

Whether they show up in the aggregate data or not, I agree is a hard argument to make. I welcome all that from the executive order. On the antitrust stuff, however, I will say that some of the rhetoric in the order disturbs me. Repeated reference to unfair conduct, whatever that means, but I have a hunch it doesn't mean anything that I would consider or it's much broader than anything I would consider to be anticompetitive conduct and encouragement of FTC substantive rulemaking in the competition area. I think we'll probably talk about that later. I'm not a big proponent of that at all.

ABBOTT: Doug, following up, you mentioned mergers. A few months ago, the FTC withdrew its acceptance of the 2020 Vertical Merger Guidelines. In January, just three months ago, the Justice Department and the FTC issued a request for information on issues that might inform new merger guidelines including both horizontal and vertical issues potentially. Do you agree that new guidelines are needed, first of all, and if so, what major changes would you suggest?

MELAMED: I don't know if needed is the right word. I don't know that you need guidelines, but I think it's probably appropriate to update the guidelines to make them touch more aggressive in some of the places where that's warranted and reflect some newer learning. Fruitful topics, it seems to me certainly vertical mergers. I think it would be important to make clear how price effects unrelated to exclusion of rivals can violate Section 7. That was the theory in AT&T and Time Warner which interestingly, the court, as I read it, accepted but then they rejected the facts.

Certainly elaborating about how you'd apply the antitrust laws to restraints or mergers adversely affecting labor markets I think would be helpful where market definition might not be necessary in order to resolve a merger case. Nascent competitors is a tough one to sell, but one that deserves a lot of thought. The issue of presumptions, and where we should go in that direction. I will say I think the request for information that the agencies put out is a little worrisome because I think it's very tendentious. At the outset, they say, "We're interested in information that will help us strengthen merger enforcement."

I would have thought the appropriate question would be information that would help us improve merger enforcement. They ask for information about false negatives, they don't ask for information about false positives. They suggest cost reductions as a result of doing away with unneeded superfluous workers, might not be a cognizable efficiency. I just find that, obviously, a departure from an economic welfare focus into a distributional focus on whose ox is gored.

I'm a little concerned that the authors of the request for information didn't keep in mind that if the guidelines are going to be useful as guidance to the business community or as aides to the courts, the courts are going to have to accept them as consistent, with albeit, obviously, nudging and shaping on the basis of agency expertise, existing law, which includes cases decided since the 1960s. It's not clear to me that the authors of the request for information contemplate new guidelines that will be so restrained.

ABBOTT: Josh?

WRIGHT: There's a lot I agree with, almost all of it. The RFI is a propaganda document. It's not a serious request for information to improve the methodological foundations of merger analysis. It's just not what it is. It's not what it's sold as. I think where Doug ended is where I will start, which is the reason that the merger guidelines have been successful in federal court is because it's a pretty good document in terms of laying an economic foundation for analyzing whether mergers result in an increase in market power. That's what the guidelines are meant to do.

They're meant to communicate the analytical framework the agencies use for internal decision-making and tell courts, who are usually generalist federal judges who don't do a lot of antitrust cases, to tell them, "Hey, this is how you can think about these problems." The reason the guidelines have been successful is because federal Article III judges have picked up the guidelines and said, "Goodness, these are in fact helpful. I will use them." To the extent that if the question is, is it a good or bad thing to revisit guidelines? Goodness, I would love for there to be a process where we sat down and figured out how to write a better section on entry.

Whether we could do something better on unilateral effects, whether in the vertical space, it makes sense to write down, as the 2020 guidelines attempted to do, a little bit more of a serious exercise of writing down how to think about foreclosure concerns and vertical mergers. All of those things are useful. Thinking about double marginalization and transaction cost efficiencies in a way to communicate both how the agencies analyze those issues, but also how court can and should, but that's not what this process is.

I'm happy to bet Doug a beer that he doesn't think and I don't think that the guidelines that come out of this process are going to do anything to improve the economic foundations of thinking about vertical mergers or anything helpful for judges. They’re going to be full of presumptions that are divorced from economic evidence and theory. The document in which the agencies announced the repeal of the guidelines got successive monopoly and double marginalization wrong at the level that an undergraduate economic student would be able to point out the errors. It was bad and clearly never went through BE in any meaningful way.

That same document said there can't be as a matter of Section 7 law, as a legal matter, the statute doesn't say efficiencies and so there can be no efficiencies defense in the case of a merger that improves welfare because of marginal cost reductions. Section 7 applies to horizontal mergers too. They obviously think it both with respect to vertical and horizontal mergers. There's very soon to be three votes for that proposition inside the Federal Trade Commission. I find that depressing on the one hand because it's clearly wrong about the law and the economics and it's bad policy.

On the other hand, the nature of this RFI and the process, if what we get is a set of guidelines that say, "Hey, there's a nonsensical, not based on economics presumption for shares and there's no efficiencies defense," and all of that, we'll be sitting here in a couple of years saying the Biden FTC went and lost a ton of cases, nobody adopted the guidelines, and in 2024, whoever is there will need a new set, and maybe then, we'll have a serious discussion about all of the issues that Doug raised, which I think are the right issues for a real updating of the guidelines, which we did in '97 and 2010. I'd be happy for that conversation to happen, but I think that is a couple of years away.

ABBOTT: Thanks for your crystal ball gazing, Josh. Very interesting. Speaking of crystal balls and predictions, legislative proposals before Congress would, among other things, focused primarily on big data platforms and prohibit them in effect from doing a number of things, including entering into a large number of mergers, engage in self-preferencing, however that's defined, and perhaps require them to offer interconnectivity and other forms of cooperation to rivals. Obviously, just touching the surface, what's your take on any of the proposals? I'm not in the predictions market, but any thoughts as to whether they're going to go anywhere?

WRIGHT: I'm not in the prediction market either, but it's free so I'll offer predictions anyway. Buyer beware, you get what you pay for. I don't think any of the pieces of legislation are going anywhere. Pure politics for a second, I think the moment for that has passed. I think one of the reasons the moment for that has passed is you had a lot of bipartisan fervor about perhaps there could be some gains from trade on competition legislation between the Republicans and Democrats because of the conservative anger over big tech. I think there was a window, and you've seen certain senators in the Republican caucus stand up.

Goodness, I was jealous of how well the Republican commissioners treated Lina. I don't think I got that when I went through, but I think that moment has passed. I think the moment has passed in large part because now that the agencies have run under this administration for a while, people are seeing what's happening. Lina, to her credit, has said, "Hey, I'm not here for big tech. I'm here for all of you." That's how the agencies have proceeded. It has been a general world view that all transactions probably violate Section 7, or whether or not they do, they should be stopped.

A preference for obliterating or eviscerating the HSR process that covers merger review at the agency, removal of, particularly near and dear to my heart, the 2015 policy statement under the unfair methods of competition to pave the way for rulemaking, and to get rid of a constraint that said in the rulemaking context, we will think only about economic welfare, doing away with that on their first day.

It is not just the big tech industries that are suffering under the leadership of the new antitrust agencies, it's everybody and I think there's been enough time to pass so that the people in those other industries, be it retail or private equity or manufacturing or whatever else, are appealing to Congress and saying, "These are bad ideas and they are going to hurt us. Please do not give superpowers to the United States Federal Trade Commission or the DOJ." I think that's a much more believable plausible story than it was 6 months ago or 12 months ago because there's evidence of some of those policies. The FTC advocating for prior approval of all deals when there are acquisitive firms. I had to look up what that means. It turns out it doesn't mean firms that do a lot of acquisitions.

What they mean, and that you see in their policy statement about prior approval, is they contemplate prior approval not just in consent, maybe a rule that says if you are a firm that has done a lot of acquisitions or thinks about acquisitions often, you have presumptive illegality and they would like you to, in exchange for your merger, give up judicial review rights on all future transactions. This stuff is, I think, really far out there, and pretty obvious that it's not narrowly tailored or targeted to big tech. I think the more that lesson has been understood by the antitrust bar as a whole, by the economy as a whole, the more that that political window has closed.

My view in $5 gets you a cup of coffee, but my sense is that window is all the way closed. If I were betting on whether the AICOA, the self-preferencing bill, that also replaces the consumer welfare standard, but for what it's worth, with essentially harm to rivals. My view is that the probability of any of that goes through is next to zero. Some of that becomes from, I think, the smart money is always bet against Congress so some of it is just general skepticism. I also do think that the political moment for some of that has passed.

On the substance, I think nearly all of those proposals do a couple of things that I think really don't have a place in a healthy antitrust enforcement system. One is they create different rules for different industries and different technologies who they deviate or undermine the consumer welfare standard. Those are the two. I think those have negative knock-on effects for what antitrust becomes, but that's my take on the legislation.

MELAMED: I noted at $5 a cup, Josh drinks more expensive coffee than I do. I don't know. I'm going to just add a couple of thoughts. In some ways, I think the most interesting thing that Josh said was near the end of his comment when he said that this litany of things which he complained of the FTC were aimed not just at the tech platforms, but at antitrust in general. I think it's a really important point. I actually interpret a lot of what the commission has done. I have a list of about eight or nine of them. I won't bore you with trying to recall what they all were.

Have the property and therefore, give rise to the inference that they were intended to implement an enforcement strategy by an agency that thought it couldn't win in court, and it wanted to deter transactions without ever subjecting itself to judicial review. That's obviously a very worrisome development, I think, for those of us who believe in the rule of law. Against that background, one could imagine, "Gee, if we can actually change the law and make it a little bit more maybe agreeable to some of the folks in the agencies, maybe we will at least get back to the rule of law," that would imply some general antitrust reform.

I suspect you could put Josh and me and three or four other people in a room, and have maybe not just coffee, maybe something stiffer to drink, and after a while, we could probably come up with legislation that might make some sense, but I don't believe Congress is going to do that. I don’t think Congress is capable of doing it. The interesting question that you raised initially, Alden, is what about sectoral-specific legislation.

I can imagine an argument I wrote about a couple of years ago, you should have, arguably, specially antitrust laws for the tech platforms on the ground that they seem to have inexorable growth potential with network effects, scale economies, and so forth, but that's not going to happen. To me, the central problem with the tech platforms is that the problems they raise are not exclusively antitrust. There are the economic problems of economic power and inequality, but there's also sociological problems of addiction and disintermediation, and privacy. Then there's the political problems of disinformation and censorship and all that.

The remedies for some are inconsistent with the remedies from the other. If you want to open up entry barriers by going in the direction of interoperability and data portability, we create cybersecurity and privacy risks. I can imagine an argument that what you need is a sectoral regulator reconciling all these competing concerns. It's a huge task, not going to happen. Instead, we have these primal screams on the Hill about fair treatment of others using the platform. I don't even know what the problem is. An antitrustor would look at the platforms and think of issues like self-preferencing and non-discrimination, they would say, "Is that a way of aggrandizing existing monopolies?"

It doesn't appear to be, but that's what’s on anybody's mind. Is it a way of extending market power into adjacent markets? Could be, but these bills aren't focused on that at all. They seem instead to articulate a general notion of fairness. It's not entirely clear to me what's unfair about taking advantage of scope economies. I think until they have a better definition of what the problem is, I guess maybe for a slightly different reason, I come out where Josh does, I don't see how they're going to get a consensus to come up with a solution.

WRIGHT: If I can just add one small thing there, Alden, before we go to the next one. I'm going to triple down on this judicial review point, Doug. I agree wholeheartedly, I think if you look at the strategy and tactics of the agency as a whole, there are a couple of pieces that fit together; the removal of the policy statement and the intention to do competition rulemaking instead of litigation, the evisceration of HSR to put pressure on people in the review process to walk away from deals or let them expire or just make it generally burdensome in order to take deals off the table, sort of in terrorem effects of procedural tactics inside the agency along with with the rulemaking and the prior approval.

I think there's very much a concerted effort to move away from the courts. I think a really interesting point with that, however, is that what's happening on the ground, and I'm talking now about the merger side more than the conduct side, on the merger side, it used to be the case that you had a merger in front of the agencies. Think about a retail deal or something, you go in on day one and you say, "Listen, we agree with you that in these 10 markets, there's a problem. In these 50, we all agree they're fine. Let's focus our advocacy on the other 10 and we'll find a divestiture that solves the competition problems." Everybody's happy and we walk away.

You go when you do the analysis and you agree on the questions and staff recommends the solution and the commissioners accept it because everybody's reading from the same page and thinking about things the same way. You can't be a practitioner in front of the agencies and do that anymore. You don't agree to a timing agreement with this agency. You come in and you say, "You got 30 days, sue us or don't," because whatever the staff does, they recommend a settlement, a divestiture package, whatever, the commission rejects it and says they're going to sue you.

The folks who are around the antitrust bar are pretty smart. They've been playing this game for longer. I think people are learning that the optimal strategy is to hold the agency's feet to the fire and say, "I don't know. If you're going to sue us, sue us." Eventually, the bullied says, "You got to fight me to take my lunch money," and I think you're seeing a little bit more of that. I think the interesting point that that raises is I think the FTC has opened up a lot of battlefronts and it's not clear they can cover them all. The idea is avoid litigation and spend our time on rulemaking.

I think they simultaneously have adopted a strategy that almost guarantees that we're going to get a lot more litigation. We're going to have to wait and get their third vote, but I think they're going to either end up in a lot of litigation that they don't want to be in with bad cases on the merger front, or the world's going to learn that their threats aren't credible. I don't know which one of those is the equilibrium.

I think that the next move with that strategy, and I think we're starting to see it in the merger world already, is people saying to the FTC when the HSR clock goes, "Yes, I received your letter that says you can sue me anytime you want. Sue me." I think we're going to get a lot more of that. Maybe that's good, maybe it's bad. It's good for my casebook, there'll be more cases. I'm full of predictions today, but I guess that's going to be what I think the next year looks like.

MELAMED: I completely agree. It certainly has seemed to me since I saw some of these things roll out from the commission that if I were still in the business of practicing law, I'd be telling my clients, "Look, you want to bring a deal, get ready to litigate. It's a different game." Here's my worry, Josh. It's not just that everybody's going to say, "If you don't like it, sue me," and eventually, maybe the house of cards will crumble. What I worry about is that in the meantime, how many deals?

The lawyer goes to the CEO or the board and says, "Look, this deal looks like it should be lawful, but you're going to have to litigate it because it's not so obvious that they're not going to sue, but you have to litigate it. You have a 70% chance of winning and it's going to take three, four years, or two years," whatever it's going to be. How many deals does the board say, "It's not worth it, I don't want to do that."? That's the price, I think, there, one of the prices that you pay for this approach to merger enforcement

WRIGHT: Absolutely agree. It's an area, I think, because of it, look, I think at the end of the day, fast forward two years when we look and see how many cases did the FTC and DOJ bring in when, do they have any rules? Did we get any major legislation? My guess is the answer will be no more wins than the Trump administration or the Obama administration, probably fewer as a percentage. I don't think that there'll be any competition rules on the book because I don't think the FTC has the authority to promulgate them, and I suspect that the courts will agree.

I think when people look at the scoreboard, that will suggest no harm, no foul with this administration's antitrust enforcement choices, but that misses the story that you just told, which is, I think, these in terrorem effects are general deterrents of competitive conduct and mergers and transactions. I think that's the real story and the real loss that arises out of this strategy.

ABBOTT: Very quickly. Josh, you preempted my question on FTC competition rulemaking under Section 6(g) of the FTC Act, which refers very generally to the authority to issue rules, and part of a larger section dealing with largely procedural matters. There are a number of people who have argued, myself included, that the FTC would lose before the courts. I think probably doesn't have that authority, but it clearly is going to try. It said so in a December 2021 regulatory agenda release. Doug, what are your thoughts on the future of this FTC competition rulemaking?

MELAMED: Well, I don't profess to be a real expert in the law in this area. I'm sure Josh has studied that more closely than I, but from what I do know, I think it's unlikely that a court will find that they have the authority to make, do substantive rulemaking under the competition prong of Section 5, but I want to just add a little something on the desirability of rules. I could imagine a conclusion, this goes to legislation as well that says, look, the courts in these, what I called micro rules earlier have gone too far. We want to recalibrate them ordinarily because these are judge-made rules. You expect the courts in the common law process to do their own recalibration.

These courts are very conservative. The Supreme Court is extremely conservative on antitrust so it'll be a long, long time before the common law corrects these, what some people like me might regard as errors. Let's try to pass rules, legislation or FTC rules. I'm not talking now about codifying decision rules. I'm talking about clarifying principles in a way that you would hope wouldn't become obsolete with changes in market circumstances. I could imagine a theoretical case for some kind of of rulemaking by some organization, but I wouldn't want it to be the FTC.

That's because of the specific FTC concerns I have, it's hard to imagine consensus on the commission. I don't know how long the rules would last if the commission- I don't think you should have different antitrust rules for the FTC than the ones you have for the DOJ because of industry inconsistencies that that would imply. Also, this is something near and dear to my heart, FTC rules would be under Section 5 of the Federal Trade Commission Act. They would be enforced in the administrative proceedings rather than federal courts, and for reasons I've written about, Josh has written about, he and I have lamented about for a long time I think administrative adjudication of antitrust principles is a really bad thing.

Not because of the ALJs, but because basically, you have the prosecutor sitting as a judge, where the FTC makes a final decision. I think it's not just that it's unfair to defend and to always lose in contested cases involving Sherman Act issues have lost everyone since 1984 at the commission level. I think it leads to an impoverishment of the adjudication process when it does get to the Court of Appeals, because you have the commission defending and locked in under the administrative law rules to its opinion, which is often not- but your compromise job rather than being advocates in a court, defending a result by a lower court.

I would think you have to rulemaking is probably not legally authorized, is probably undesirable in general, because rulemaking may not be the right way to solve antitrust problems, but it's certainly undesirable if it's confined to the FTC in Section 5.

ABBOTT: That raised an even bigger question for both of you. Senator Lee has argued that there should be only one federal antitrust agency, the justice department, for the sake of policy coherence and efficiency. What are your views, Josh?

WRIGHT: I'm happy to start. I think as -- A couple of things, I'm sympathetic to the idea that the cost of multiple agencies exceeds the benefits. We ought to reconsider the institutional structure and whether we would do one, whether put it in the DOJ or what have you, I'm sympathetic to that argument. I'm sympathetic to that argument in large part because the agencies over time have proven, for one reason or another, that they just can't be happy operating under the same rules. The FTC has chosen to interpret, its Section 5 UMC authority to give it different rules, or at least to argue to the world it has different rules, say, in areas to do with in IP and antitrust is one example, but there are others.

I think the FTC insisting that it could use administrative adjudication if it loses a merger case in federal court, which is something it's done from time to time. I think on the other direction you've got not too long ago the DOJ intervening in FTC cases and Amicus briefs, even in places where I agree in most of the substance of the DOJs brief. It's just an unhealthy situation where you've got different rules, different agencies, and you've got an industry sort of coin flip to see which one governs. Right now, I think you're also going to see some process-based differences and merger review as well between the agencies.

I'm an FTC guy, I've worked at the FTC four times, from being an intern to a commissioner. It's very difficult for me to feel sympathy for the argument that we have to take the competition mission away from the FTC and hold it all in in one agency, but goodness if they're not making the case for it. I think the behavior of the agency with abuse of administrative adjudication in particular, but now expanding into the interum effects we're talking about in merger review, some of the differences between the agencies, I think -- I am sympathetic.

I don't know if I'm in a place yet where I can give full-throated support to the idea but the continued divergence and policy differences and enforcement differences between the agencies that have been persistent over time and across administrations I think is really inexcusable. I think it makes the case for structural change. What I would like to see first before any of that is, I would, if I could wave a magic wand, I'd get rid of administrative adjudication at the FTC tomorrow. I would at a minimum restore the 2015 policy statement if not go a little bit further, both as a constraint on the Section 5 authority generally, because I think that would be desirable policy but it also minimizes -- It's a really attractive tool to an agency that wants to make its own rule and standards, and doesn't want to think hard about whether that curates divergence with its sister agency. At a minimum, I would get rid of administrative adjudication, I would impose constraints on UMC interpretation, and I'd probably do away with the FTC ability to do competition rules for all the reasons that Doug said.

ABBOTT: Doug?

MELAMED: Well I think it makes no sense to have two different agencies. If you were starting from scratch it's hard to imagine that's what you'd come up with. I think we ought have one agency, for consistency and efficiency. I think it's a better model when we're out proselytizing and with good, sound antitrust policies to foreign countries. I think that agency should be the justice department. I've heard it argued that having the consumer protection, a mandate enriches competition analysis at the FTC. I haven't been inside the FTC enough to know, I'm skeptical that that's really important or at least shall we say institution specific benefit.

Certainly absent of that, the DOJ is a better agency. First of all, it has a criminal authority. Secondly, it's a law enforcement agency with a law enforcement culture, which I think is a healthier one for sound and antitrust enforcement which is about competition, not regulation. It's a single-headed agency so I think it's more likely to have -- to act in a coherent and consistent, internally coherent way. The theory is that of the cases it brings are going to be coherent and intelligible, more so I think than in a multi-headed organization.

WRIGHT: The synergies of consumer protection and competition missions inside the agency are -- the story of synergies between those are -- the stories I wish were true. I really do. I think one can imagine making that case in theory, you wouldn't have to search far in my own writing to find me making that case previously, but nothing disabuses one of that notion more than working inside the agency.

ABBOTT: Good interesting point. Let's turn quickly to monopolization. Now given current case law what do you think is a prospect for such big cases as FTC v. Facebook, which until a few months ago was on life support until the second crack at the motion to dismiss by Facebook failed. Of course, DOJ v. Google -- that leads to the separate question about a potential amendment to Sherman Act Section 2 language to overturn such precedence as Trinko perhaps linkLine.

I don't think that's going to happen, but in general, I'd like the two of you- the thought two of you on monopolization, and as a footnote, I might mention that there's an ongoing effort in New York state to amend the New York state competition law to adopt an abuse of dominance standard, European standard, as opposed to the Sherman Act Section 2 standard.

MELAMED: Well, let me just try a brief response to this. I'm not going to prognosticate the outcomes of cases. I probably might feel only slightly more confident doing that than the outcome of legislation. I think the FTC's case is a really interesting case, but I think they're going to have enormous difficulty, even if they get past all the analytical problems. Explaining to any judge and certainly, the judge is sitting in that court – “Why did you wait so long on the mergers?” I think the Google case, as I understand, it's a very complicated case to litigate but I don't think it's making particularly pathbreaking in terms of a new law or anything like that. State cases is more so than the federal cases, but not so much I think.

Look, I think the area of monopolization law is an area that I certainly had in mind when I said earlier that we need some recalibration. The law, I wish the law would be nudged here and there. I don't think you need to change the statutory language. Frankly, I think if courts were willing to listen and look at Trinko and say, "Let's imagine that Trinko" -- I think Trinko was rightly decided. The problem is it was written by Scalia and not Breyer. There was a lot of broad language there, which in lower courts, in many cases have run with it, taken that whole idea, in my view, way, way too far on what duties a firm might have when it's dealing with an outsider using the platform and so forth or wants to use the platform.

I think we could get, we mean the law, the legal system could get to where we are now to where I think they ought to be with cases in principle, like the cases that are pending. The judiciary is very conservative. We talked about that. I don't know that either of these cases is likely to nudge the law. I think that's unlikely either because they'll lose or because they'll win on conventional legal theories.

ABBOTT: Josh, your thoughts.

WRIGHT: I'm not much in the business of prognosticating the outcomes in those cases as well, but I certainly agree with the proposition that neither case is one that's going to move Section 2, generally. I do agree, I think in the Facebook case in particular, because we've got more of a track record to look at what's going on there, I think they've got some real problems. I think they've got summary judgment type problems and the judge has highlighted them already.

I also think when push comes to shove in a case – this is agreeing I think with where Doug was going – when you wait 10 years to bring a case, and you've got a consummated deal, the burden of proof facing the agency on the books isn't different, but goodness if the judge isn't looking for real evidence that compared to the counterfactual “this deal's anti-competitive”. If you're going to wait a decade to bring a case, you better have the goods.

There are styles of consummated deals and killer acquisitions where the theory is, well look, they acquired the thing and they buried the product. As long as in the but-for world, the product is out there and doing something and making some consumers happy, maybe you've got a case, but this isn't that case. My children telling me that WhatsApp and Instagram are still things that people use. This isn't one where the theory is that they bought it and buried it. They'll have documents to show and the like, and they waved them around in the congressional hearings, but short of that, I'm not -- I worry about the agency meeting its burden in that case.

What I was saying about Section 2, and we were talking about Trinko and reform efforts through legislation or elsewhere, be it the United States Congress or New York statute, which I actually think is quite a bit worse than Europe's monopolization standard. I think that's hard to do. I think in the US- Doug's got a handful of cases, be they Qualcomm or Amex or other Section 2 type theories where I think the question is, how is the law operating in those areas.  

I used McWane as an example that I think cuts the other way where the FTC came to court and said, "It is not our burden to show that this exclusive dealing contract raised the market price or reduced market output, or harmed consumers. All we need to do is show that the defendant has a high share, and because it has a high share, obviously you get a foreclosure rate that's relatively high as well. One of my favorites for a discussion like this is McWane. Now, I'm the dissenting commissioner. I wake up thinking about the one that got away, but couldn't even persuade one colleague to vote with me. I'm by myself in dissent. If you look at the record, the FTC proudly argues we have not a lick of evidence.

This was a monopolization case involving exclusive dealing – so, sort of past conduct.  Not a lick of evidence offered about what happened to price or output. This wasn't a fancy innovation theory, It wasn't lost innovation in the iron ductile pipe-fitting market. It was they monopolized and increased the price and reduced the output through foreclosure. They said, "Our burden's not to show you any price or output effects." Not an ounce of it on the record. I don't know, but they won. The 11th Circuit said, "We defer to your fact-finding," and a foreclosure rate in a big market share is enough. They just didn't calculate the foreclosure rate right – I think a really weak case that they won easily and they won through the Court of Appeals without even really trying to put on meaningful economic evidence. Both cases are out there but connected to winning in those cases, agencies and private plaintiffs win cases. They win good cases. Sometimes they win bad cases. In my view, McWane [was a bad case].

ABBOTT:  Do you have anything to add about monopolization cases, Doug?

MELAMED: Well, I'd fire one small point just to make sure maybe the record as where it's not incomplete. Josh said, and I don't disagree with this, that the Facebook acquisitions of WhatsApp and Instagram were not killer acquisitions in the sense that they didn't buy the nascent competitor and put it out of business. I would hate to have a rule that would say, "If you don't put it out of business, you're home free." I think the real question in a case like Instagram, is whether Instagram could have been a real game-changer to the whole social networking paradigm and shaking things up and diminish Facebook's role and so forth if they had been in independent hands, rather than is operated as a complement to Facebook.

Well, I think the facts are going to be very tough to prove in a pending case. That, it seems to me is the interesting issue in nascent competition cases, much more than the so-called killer acquisition story.

ABBOTT: Certainly, on the IP antitrust and I know you have argued that the FTC removes- lawsuits monopolization case against Qualcomm, which basically involved claims that Qualcomm had violated the Sherman Act to its restrictive patent licensing policies primarily on standard-essential patents [SEP], patents that read on standards and that it had a no license, no chips policy and it won, the FTC won in district court. In fact, the district court judge as a remedy wanted Qualcomm to renegotiate its contracts, but then it won before the court, but then a Court of Appeals reversed and Qualcomm came out the victor.

In related to that, more recently, the Biden administration appears to be reversing Trump policies regarding licensing of again standard-essential patents, patents that cover standards. Now, under the last administration was argued that holders of these standard-essential patents should be treated the same as holders of other patents. The Biden administration seems to be taking a tougher line and is contemplating a new policy statement on his views on licensing of standard essential patents. Your thoughts, Doug, on that – more broadly the set of IP antitrust issues?

MELAMED: This is an area where I'm pretty sure that Josh and I really don't agree much. The Qualcomm case is a really interesting case. It's a one-off. I'm glad to talk about it, but let me leave that aside because it would take a while to unravel the problems there, which I actually think you have almost nothing to do with FRAND commitments and some of the core issues involving SEPs I think you have in mind. On the general policy issues, I'm completely supportive of the Biden administration. What the Trump administration did in the Justice Department, FTC wasn't with them.

I think it was eccentric, it was a departure from prior almost consensus understandings. I actually think the whole issue of public policy in this area, it's just a wonderful case study of public choice. I think what you have is like a few companies whose huge revenues depend largely on licensing patents and they- not surprisingly have been more committed to it, more effective at lobbying than have the technology implementers, those who make devices and machines using the patented technologies and who care a lot about the patent royalties they pay, but don't look at the royalties as the core of their business.

You have, I think, a very skewed debate that loses sight of some important starting points. I'm just going to mention that and maybe I'll stop. First of all, patents do not confer rights broader than those of owners of tangible property. They were never intended to and they don't –  they're limited in duration. I own my house. I don't have like a next year term, I own the house. There should be no special privilege from the antitrust laws for intellectual property law. That's point one. Point two, most of the biggest disputes in the areas to which you refer to Alden, involve the IT sector, information technology, where for good welfare-enhancing reasons, licensing takes place after firms have begun implementing the technologies.

I just want to pause on this. My cell phone has technologies that are claimed by an estimated several hundred thousand patents. It's inconceivable that any phone manufacturer could get pre-clearance of all the patent issues before we're rolling out a phone if we wanted to see a phone in my lifetime. We should welcome this implement first and then worry about the patent issues later, but the significance of that is that these patent issues, therefore, are resolved when you have a dispute between a patent owner who claims he's owed money by an implementer who already has an accrued potential liability.

The idea that we can look to some kind of market bargain and licensing practices in different contexts as useful information is very misleading because what's really going on, no matter what you call it, is settlement of a threatened or actual litigation. It's not a market transaction in the ordinary sense in which the parties can walk away if they don't reach a deal. Neither party can walk away. If the patent holder doesn't get a license, the other guy's using his technology, he claims, and if the implementer doesn't get a license, isn't happy with the license terms, he winds up in court.

All we're talking about here is the terms on which litigation can be settled, and the law has to take that reality into account, and when it does, I think a lot of the rhetoric about markets and high pollutant theories, property rights, and all that drive a lot of the- what I call this public choice success on the other side of these issues, I think evaporates and we wind up pretty close to where the Biden administration is.

ABBOTT: Josh, your thoughts.

WRIGHT: I think right now, the instinct to point to what one perceives as errors in the courts feed policy efforts to do bad rulemaking or bad legislation, or bad policies that we've talked about otherwise. I don't doubt that Doug and I could get in a room and think about general principles that we'd like to see articulated and some guidelines, or as a high level to guide Section 2 enforcement. We disagree on the outcome of Qualcomm as a particular case, but I'm listening to Doug give his general principles for how he thinks about IP and antitrust and the symmetry between how we should treat IP and real property. These are all things that I think are good ideas that I do agree with, but yet disagree about the outcome in a particular case, given the evidence put in front of the court.

I think we could do that, but I think the policy debate in the country right now and what's happening at the agency isn't that. I think the actual proposals on the ground for the cases the agencies right now want to bring and the pieces of legislation in front of congress right now, aren't to tinker with Trinko or make some general statements about economic evidence. The house report proposes overturning every single Supreme Court antitrust case since 1977.

In a footnote, they proposed overturning Frank Easterbrook's 1984 Law Review article. I don't know how one does that. I'm mostly jealous no one's overturned one of my articles yet – they overturned my policy statement, but not articles. That's okay. Look, that's where the debate is right now. I don't think we're going to get those sorts of principles. I think you could get improvement in Section 2 doctrine. I would like to see more focus on actual proof in Section 2 cases. I would be happy to trade more focus on actual economic proof on both sides of the V in exchange for less broad and sweeping safe harbors.

I'd be happy to do that, as long as we did it symmetrically. That's not the policy discussion that's happening at the agencies. What's happening is can we declare bright-line rules of per se illegality for a broad swath of conduct that we agree are sometimes anti-competitive and sometimes pro-competitive? Where I'm left in the absence of real evidence that there's a market power crisis in the economy or real evidence that to me, and this is a place where I think Doug and I do differ, I don't think there's a real crisis for antitrust plaintiffs in the courts. I think they win good cases. I think sometimes they win cases they shouldn't.

In the absence of either of those, I think in looking at the actual policy alternatives being proposed in places like New York and out of the agency and in the executive order, Doug's ideas for how to fix Section 2 just aren't popular enough right now- but they're more popular than mine, Doug, that's not a shot at you.

MELAMED: That's why I don't do politics, Josh. I'm in a different world.

WRIGHT: I try to be.

ABBOTT: This has been a fascinating conversation. I wish we could continue, but I think we'll have to wrap up now. Maybe, do you want the last word, Doug? Do you have anything to add to Josh's broad comments about the bigger picture?

MELAMED: Well, let me try this. It's a question. I think one of the problems in antitrust is that it's become very complicated, very technocratic, and I think a lot of judges don't understand it. It doesn't explain every case that was rightly decided or wrongly decided but I think it's a problem. That's because after all, antitrust is decentrally enforced by 51 government agencies and every injured party and in hundreds of district courts around the country.

I wonder whether we ought to be thinking in the ivory tower sense, not because it's popular, we can get it done in Congress, about simplifying antitrust or expert tribunal, another issue, another conversation, put that aside. Here's my question. What if we check all of antitrust law, and we change it to the following. A defendant violates the antitrust laws, if it, one, engages in conduct that increases its market power compared to the but-for world and that conduct cannot be justified by efficiency benefits in the nature of improved product quality, lower prices, or lower costs, sufficient to overcome the market power costs.

Get rid of Trinko and Brooke Group and Sylvania and all of it, and just put those questions before the court and then have the parties decide what facts and what economic analysis is needed to address those two questions in the case before them. Would that make it easier for judges to reach the right decision?

ABBOTT: Well, that's certainly a provocative question, Doug. Broader philosophical question, and I don't have an answer to it, but I don't know if you have an answer to it, Josh.

WRIGHT: You frame that as, little last closing comment to think about, and then turn the antitrust world upside down but there's a lot to like, there. Frankly, I think for broad swaths of antitrust, you described what the law is, but for these areas where we've developed safe harbors or presumptions of illegality in another way. This is, do you do economic intensive case by case analysis on those questions, which I think are inherently economic questions, in every case, or do you try to carve out sensible bright-line rules, whether- presumptions of illegality or legality on both sides.

I think that question return us to where we started with the Neo-Brandeisian pull on on antitrust is the far antitrust left is now the owner of the desire for bright-line rules and antitrust and bright-line presumptions of per se illegality for everything, whereas go back to 1984, that was Frank Easterbrook's domain, was simple rules for antitrust and safe harbors. What I think that approach misses, and I do like it for most of antitrust, is there are areas where I think we know enough about the conduct- take per se rules for naked price-fixing.

Take, to do one that's more provocative, I think the literature on vertical mergers is overwhelming, that vertical mergers are generally pro-competitive or competitively neutral. There are exceptional cases where vertical mergers harm competition, for sure, but there are areas where we know a lot about the distribution of the competitive effects of a practice, where it's 95% likely that conditional on seeing this type of business arrangement, it either hurts or harms consumer welfare, it either increases market power in a way that fits your definition or does not.

I think the question for antitrust is, in the area where we have that information available to us, do we want to use it to shape rules or not? It’s a transaction cost question, in a way, about the efficiency of rules versus standards. To me, I think the answer to that question is, yes, we do want that from our antitrust system. We want it on places where we can use it to make enforcement more efficient, our law per se rules. We want it on the safe harbor side, too because first conduct where the answer to your question is obviously no.

The conduct is not likely to result in the effects the antitrust laws ought to be concerned with. I think an optimal antitrust regime has rules that get to that answer faster and with a greater percentage of the time than a case by case – this is rules versus standards 101, but I think the answer to the question should an antitrust regime respond to the economics of the rules versus standards debate? Absolutely. Are we doing that well? Sometimes, sometimes not.

ABBOTT: Thank you.

MELAMED: If I can, one brief thought. A lot of wisdom and what Josh said, but the problem, I think, and I’m not convinced of my ideas, more than half baked at all, but the problem is that when Josh articulates the benefits of rules and presumptions and in effect codified learning, he assumes that we know the category this conducted issue falls into. What I think happens a lot is we have a rule, a unilateral refusal to deal, predatory pricing and then we spend a great deal of time arguing. Wait a minute, is this predatory pricing? Is it tie? Is it exclusive dealing? Is it a refusal to deal a conditional refusal deal? That's where the complexity comes in, but for another month or so, to talk through.

ABBOTT: To be continued. Really provocative thoughts by two great scholars of law and economics. I hope that everyone here has learned something, and look forward to future conversations and future podcasts on antitrust. Thank you.

Photo by Steve Heap via Getty Images.

Support Mercatus

Your support allows us to continue bridging the gap between academic ideas and real-world policy solutions.Donate