A PYMNTS Company

Antitrust Policy in the 21st Century: Is There a Need for Reform? – Session 3

 |  September 17, 2020

Below we have provided the full text transcript from the third panel of our recent roundtable discussions, Antitrust Policy in the 21st Century: Is There a Need for Reform?, from July 24.

Randall Picker Speaker

Randal PICKER:

Hi, I’m Randy Picker, and I’m a professor at the University of Chicago Law School. As the house is doing an investigation of digital marketplaces that’s been ongoing, they reached out to a number of people and said, “What do you think?” A number of us wrote submissions. When I wrote my submission, I tweeted it out and CPI reached out to me. I said oh, we should do some hearing, not hearings, but have some sessions on these. CPI reached out and that got us here. We’re doing four of these in July. It’s basically four Fridays in July.

We’re very excited. We reached out to two speakers each time. We have Bill Baer and Spencer Waller here today, and then a group of people to engage with them. This is the third such session. We’ve done two already, those are on the web. You can watch those at your leisure.

Speakers, you should identify any possible conflicts you have as we go to you. We’re going to go work our way around the room person by person. I don’t think I have any relevant conflicts, so I will leave it at that. With that, we’re going to start with Bill Baer.

Bill Baer Speaker

Bill BAER:

Thanks, Randy. Other than age, I have no real conflict to report. Thanks to you and to a CPI for organizing this. I was honored to be asked to contribute and to share the front end with Spencer Waller is a great honor. Spencer, I thought your comments to the subcommittee were great, and I urge other people to read them as well.

Before I get into a few remarks, I did want to confirm what I think your screen is probably showing. I have not taken advantage of the reopening in suburban, Maryland to get a haircut. I’m thinking that in the next three to four weeks, I’m going to have a choice to make. It’s either going to be to go back to my college ponytail or perhaps to have the second-worst comb over in contemporary American politics.


Ponytail’s classic, you should go with that.


Okay, good. Good to know. All right. Look, I’m going to walk through how I see the current debate regarding the present and future of antitrust support and antitrust jurisprudence. I conclude with a few thoughts about what Congress might do about it.

I start with, and this is in my submission, now there has been certainly decades since the ’70s, sort of a rough consensus around the consumer welfare standard as an appropriate analytical framework with which to judge antitrust issues.

Now, we’ve seen an increasingly vocal, and I think a very thoughtful view but that approach has left too much bad conduct, inappropriate consolidation on the table. I do share the concern with under-enforcement, but I’m less convinced that moving from an assessment of the impact on behavior on competition and consumers to something that’s not yet defined is a workable idea.

In other words, I’ve not really seen or read a lens that provides greater clarity. I doubt frankly, whether the federal judiciary is currently constituted would accept some kind of the stall switch to some other point of view. But that said, I think the critics have a point. Bob Piotrowski put it up a few years ago that the Chicago School overshot the mark.

That since the ’70s, while we’ve had a consensus about analytically sound and fact-based enforcement, that is more, in my view, into an overly cautious approach by the courts and to some extent by enforcers themselves. In the ’80s and ’90s, the court seemed hostile to just about any government challenge. Few merges were blocked, challenges to lateral conduct by dominant firms were infrequently brought and rarely successful. Consolidations has increased to a worrisome levels across many sectors of the economy from hospital and pharma, retail and telecom, insurance, and travel.

Vertical relationships between upstream suppliers and downstream distributors began to be seen as uniformly efficient and pro-competitive. The standard per section to misconduct seems to have gotten tougher and tougher. We went, as I said, in my submission from an antitrust cultural, where in the words of Justice Stewart, the government always wins to one where enforcers almost always lost or were fear of losing caused the government not to act at all.

Now, Spencer, [Robert 00:06:15] Camp, who was one of your earlier speakers, did a nice job of explaining the subcommittee out more recent decisions in Trinko, Amex have narrowed the ability of enforcers and private players to challenge monopoly maintenance. The current DOJ has decided that a holder of a standard-essential patent of antitrust immunity regardless of its conduct in contrary to the position the FTC has thus far successfully litigated.

The much-discussed and criticized decision on the Sabre Farelogix failed merger shows at least to me how this increasingly cautious, excuse me, approach to antitrust jurisprudence has infected merger enforcer. How do we get there? As I said to a subcommittee in my view the fear of getting it wrong as kind of warped antitrust enforcement.

That’s really my fundamental concern is the courts are too cautious, too worried about adverse effects of over enforcement of so-called type I errors. The attitude that any uncertainty should result to inaction has caused many courts to demand a level of proof that’s often unattainable. It’s sort of a one way ratchet at work in today’s antitrust jurisprudence that results in increasing under enforcement.

Now, that bias against type I errors is really sort of perversely elevated in my view the standard in proof, lack, or the losses of the civil antitrust case. It’s a preponderance of the evidence standard. Something that’s proven more likely than not. And when you’re dealing with the antitrust laws, where is the risk of problematic outcomes. When you ignore the degree with the antitrust statutes want them to be proactive and protective of a competitive economy. The effect of all of this in my view is we are under enforcing significantly.

I understand the reason why a lot of people, Chicago School folks have argued for under enforcement is fear that we’re going to kill the golden goose, that somehow we’re going to kill innovation. I actually find that argument almost silly. The notion that erring on the side of somewhat more enforcement is going to cause firms not to innovate or it’s going to cause venture capitalists not to invest in startups because they cannot be bought by a dominant platform. I don’t think we’re going to kill investment. I don’t think we’re going to kill competitive behavior. I don’t think we’re going to fail to reward success. It’s just not going to happen.

What do we do given the current state of affairs, at least as I see them and I’ve described them? One strategy has antitrust enforcers attempting to articulate better antitrust guidance in hopes that the courts will take a more assertive approach to applying our antitrust laws.

That strategy took a long time, but it worked with regard to the horizontal merger guidelines. When they first came out in the ’80s, they really wouldn’t apply to merger decisions and it’s taken 15, 20, 25 years. But those are sided now with authority. If we can develop more guidance, the ways to both conduct, vertical mergers as new guidance out. There is some hope, I think, that we can move the courts a little bit towards eliminating that bias against type I over enforcement here.

That might work. But there are reasons to doubt whether it actually will. It took so long for the courts to embrace the consensus view about horizontal mergers that whatever is done today may not actually affect decision-making for 5, 10, or 15 years. Can we afford to wait that long? I don’t think so. There’s no guarantee that this current judiciary is going to embrace that new guidance.

My bottom line is that absence in further direction from Congress as these biases has elevated standard of proof is not going to change. I think subcommittee is doing the right thing by looking at changes to current laws. Spencer’s laid out a bunch of interesting ideas. I don’t think they need to be dramatic.

What all in effect I think need to do is to remind the courts that they need to be assertive in applying the antitrust laws to avoid risk of anti-competitive outcomes, whether it be behavioral outcomes or outcomes related to consolidations. With some modest tweaks, I think Congress can redirect the courts and the forces in a constructive way.

It’s not going to be easy. A bipartisan consensus these days seems to be an oxymoron on anything. I had some hopes that maybe antitrust would be different, but I think in fact it’s actually going to be harder now given the recent reports the antitrust division is making some decisions that aren’t being based on antitrust merits but on other policy objective.

That’s terribly worrisome, and it risks calling in the question whatever DOJ decides to do with regards to platform they’re investigating. Once you sort of break that egg when you eliminate or fail to adhere to this non-partisan consensus about analytically sound antitrust enforcement and start using the laws to get from place A to place B. And it has avoided the antitrust merits. I think we have some challenges ahead. Thank you.


Good. Thank you. Here’s what I heard, I guess. I think you think the consumer welfare framework is actually the right framework or at least like Churchill and democracy, you don’t see something better. You think the calibration is not right and that calibration has led to under-enforcement. One possible path to changing that would be guidance from DOJ and FTC with the courts possibly relying on that. That’s going to take a while, so maybe a few more targeted interventions. And then nervousness at the very end about where antitrust might be going if it’s not doing antitrust proper. All right, Spencer, you’re up next.

Spencer Weber Waller Speaker


Well, thank you, Randy. Thank you, Bill. I’m a big fan, and I appreciate your submission as well. CPI is doing a great job in having all these different conversations. I’ve had a chance to watch the first two. I think a lot of my remarks are going to be teed up by some of what Eleanor Fox said. I guess in fairness to Randy let me just say that neither me individually or the institute I run at Loyola have received funding from anybody who’s affected by these hearings or any-


Great. Thanks. Good.


… of this. For what it’s worth, my only involvement in the tech sector was as an expert witness in a matter adverse to Microsoft outside the United States a couple of years ago. Eleanor and I think it was round two of these conversations talked about using just a broader lens.

I didn’t coordinate with her on this, but I did something that wasn’t tech-specific, but address a longstanding concern of mine that the United States has been a little bit parochial in its approach to antitrust and has become rather isolated in the world conversation about antitrust law. I wrote an extensive article called the Omega Man or The Isolation of U.S. Antitrust Law, it’s out in Connecticut Law Review. If anybody wants an old fashioned hard copy reprint, I will send it to them electronically or over the mail.


Did you mean paper, Spencer? Is that what you mean.


I got paper. I got paper right here.




My submission was to some extent kind of a summary of the arguments in that article. I think the United States has become isolated, and I wanted to lay out for the committee, and I’m thrilled that the committee has undertaken this task. I really have spent a lot of time trying to think about the connections between antitrust and democracy. I do believe that Congress is the place where debates about the fundamental values of our antitrust system belong. And that is the job of smart, unbiased and nonpartisan enforcers in the way that Bill was talking about to enforce those values. I think it’s a mistake and I get very nervous when they appear to be doing otherwise.

I wrote a little piece for CPI recently about the political misuse of antitrust that address some of those themes. I think Congress is the right place to set the fundamental ground rules. Most people have noticed that Congress hasn’t done much of significance in the antitrust space arguably since Hart-Scott-Rodino and the other related changes in the mid-’70s.

I suggested to the committee that they look at what I think is the dominant system of competition law in the world, which is roughly the EU model, which obviously applies out of Brussels to all member States, was 28, it’s now 27. The British are still using the model, largely derived from EU competition law.

Dozens and dozens of countries that either aspire to EU membership have preferential trade arrangements with the EU or otherwise have studied it and have filled some colonial historical linguistic or civil law reasons to sort of pay close attention. Even countries that don’t have those connections like China in deciding what they wanted to do about their own competition law have looked more to the EU than the U.S.

The distinctions are sweeping, and they’re fundamental. While I don’t particularly get into how any of this should apply in the tech sector, I think the biggest source of divergence, whether you like it or not, most people acknowledge that the biggest source of divergence is in the European approach to abuse of a dominant position, article 102 of the treaties that they enforce. And the relative poverty of section two of the Sherman Act in addressing these issues.

Many of you have heard Bill Kovacic talk about the toolkit or the toolbox. I can’t remember the phrase that he uses of antitrust. I’m afraid the U.S. toolkit is kind of empty and unless Congress does something the agencies no matter what they choose to decide in their current investigations are running up against some very tough doctrinal and procedural barriers. Bill alluded to some of them. Most of the other speakers on the panels have touched on them. Trinko is an err example of that in terms of something the supreme court has done. But it’s not the only one.

I suggested in my submission that the committee think long and hard about the converging standards of the rest of the world around a model that uses language akin to an abuse of a dominant position as the lens to look at the behavior of powerful single firms. That includes a bunch of things. It includes, in one way that cuts against my personal feelings about this. It probably puts a little less emphasis on how you got the dominant position than what you do with it, but it is a much more expansive lens about what violates the law once you’re found to have a dominant position.

It has lower thresholds. By and large, 50% creates either a statutory or a case law presumption in most of the jurisdictions around the world, sometimes a little lower. I’ve seen some research that on the monopsony side, one might want to worry about market shares as low as 25%. But the thresholds for what is a dominant position is simply less than what the US takes for a proof of monopoly power.

Most of the panel and most of the listeners are going to be aware that the abuse of a dominant position can be either exclusionary or abusive. As a result, it covers things like excessive pricing that the US law is clear that just by itself cannot form the basis of a violation.

There are two kinds of cases that they’re not terribly common nor should they be. But one stream of case is akin to price discrimination by a dominant firm where you’re doing some comparison between either closely similar markets or between different time periods to figure out whether prices are excessive. Somewhat more controversial, and I think it just depends on the facts and the evidence, some jurisdictions including the UK have done what they call a pure excessive pricing case where there’s no easy comparison. Most of those have been in the pharma sector. That is one set of tools that we just don’t have by design.

In general, price discrimination is a tool we don’t use by choice. Congress obviously established it, beefed it up and the agencies have chosen not to proceed. I’m fine with most of that, but I do think it is a valuable tool to analyze what dominant firms do in the marketplace as an abuse of their dominance. We use it a little bit to show the dominance sometimes but not as a… it doesn’t by itself violates section two.

I’m not going to go through the whole laundry list of my 13-page contribution, but it talks about the differences between predatory pricing in the United States and in the EU and any country that follows it. Margins squeezes where the supreme court says not a thing and the rest of the world says definitely a thing in virtually identical circumstances.

Tying law is different. The essential facilities doctrine and unilateral refusals to deal are alive and well, and a potent source of investigations and liability when there’s dominance outside the United States. In the non-litigation area, I think market studies are really, really important.

I know that section six of the FTC can be used in certain ways that duplicate what a market study is outside the US. But it’s uncommon for it to be used that way. The DOJ isn’t really in that line of work. I think both agencies could profit from it, particularly in two ways.

One, I would hope that when an agency completes a market study, it comes with recommendations for legislation that in an ideal world would be automatically introduced, or at least considered in a hearing at the congressional level, rather than just put aside on a shelf.

And then, in addition, market studies create the possibility of not having one size fits all rules. At least in the UK, a very thoughtful market study in the area of grocery stores retail power led to some bespoke rules for that sector of the economy. That’s something that might be important in the tech sector.

And then finally, we are the outlier in terms of our agency remedies, particularly in terms of not having any civil or administrative fines. I had recommendations around all those lines. And I would say if I had one takeaway, I would hope down the road, we could add to section seven, section five, section two, the words karma or the abuse of a dominant position, period.


Good. I hear a certain amount of European envy there. Exactly.


I’ve been called an American.


Yeah, exactly. Really focusing on our section two, their section 102, I guess, and I heard what you said at the very end, which I obviously saw in the piece as well, this idea that we just ought to add this idea of an abuse of a dominant position to the statute. That would give us closer to parody. You also say at some point you almost want to let those decisions from overseas be relevant binding, I don’t know, in the United States. That was interesting, I thought.

The other thing I heard, an overlap with Bill is a greater role for agencies, I guess, right? You want the agencies to do these market studies. Bill wanted them to do guidance and market studies leading to legislation. And then I heard the piece about fines at the end. Thank you so much. Our speakers have teed this up. We’ve got a sequence we’re going to go through here. We’re going to Josh and then Michael. Josh, you’re up.

Joshua Soven Speaker

Joshua H. SOVEN:

Hey, Randy, thanks very much. And thanks for CPI for inviting me to speak. That’s a lot of firepower to follow, so I will do my best. I’m really impressed with presentations. Briefly for background, I do do work for Google, I do some work for some large hospital systems, I do work for a lot of people. But before that, I worked in the antitrust agencies, both of them a few years, shorter forever for both Republicans and Democrats. I’m either proud to say or ashamed to say I have never held a job for more than five years. With that, I guess have a multi-faceted perspective.

Let me start with what I agree with what was just said because most of it, I don’t. I agree that the consumer welfare standard is the right standard. I think it’s worked incredibly effectively across a bipartisan group of leadership of the antitrust agencies and it has produced an economy that is the envy of the world. Notwithstanding all the problems out there, and there’s a lot, everybody wants to be us.

Our economy continues to boom, and it continues to boom with respect to the metrics that antitrust is focused on across the board. Lower prices for consumers, more choice, more innovation, just a really dynamic space that Europe isn’t even close to. I disagree, so to perhaps anticipating a little bit with what Michael said and Bill will say, and Bill picked up on this as well with these market studies about increases in concentration out there. They might be right, they might be wrong.

I don’t think they’re a very good basis for making antitrust enforcement decisions. What’s proven to work again, regardless of which party is running the show is a fact-based evidence-based analysis on what’s going on with a particular deal or a particular set of conduct by a particular company. Antitrust in the US has never done this sort of let’s stare at the cost and what’s going on in between and make some generalizations on what to do.

I really disagree with the idea that the antitrust agencies are not vigorously enforcing the antitrust laws. Again, I look at this from a perspective of having worked for both Democrats and Republicans. I keep waiting for the list of cases that people have purportedly missed, all of the under-enforcement that’s going on. It’s a really short list.

Going back to Bush 43, well, there was some a little bit of concern about water heaters and XM Sirius, and maybe some front loading washing machines. I don’t know, maybe there’s a few things that people think… deals think were missed right now. But it’s not a long list. It’s not a list of 10, 20, 30. You can usually fit it on one hand. With respect to big tech, every competition agency on the planet including the US has been investigating large technology companies basically on an ongoing basis forever. So, I don’t see that.

I also disagree with the idea that the courts aren’t getting it. I hear Bill a bit, maybe they need a refresher that the antitrust laws sort of mean what they say, and the government gets to hit on the doubles alleys, as well as the singles part of the court. But if you look at the box score, the government almost always wins. Bill won almost everything. He won the challenges to the large health insurance mergers. They stopped GE Electrolux. The FTC has won every hospital case they’ve brought, I don’t know, since for the last 15 years.

As Randy, you’ve heard me say before the Ls on the box score all have a pretty predictable pattern, which don’t show a lack of sort of adherence or fealty to the antitrust laws. The government lost Amex. My view is a little bit out there, it’s low tech because the merchants didn’t testify that Amex’s conduct was producing consumer harm. That’s missing from the trial record.

DOJ brought the incredibly aggressive case of AT&T Time Warner vertical case really aggressive, Makan stepped up and brought that one against enormous criticism that he was missing in. I credit them for trying to do it. But what was missing from the trial record was clear evidence of consumer harm.

When full disclosure, I represented Dwight to telecom and the T-Mobile/Sprint litigation, a bunch of States, very well resourced with very experienced trial counsel on their side tried to bring it. The problem with the case wasn’t that I think admittedly somewhat left to center judge was missing the antitrust laws. The problem was that in the trial record, there wasn’t any evidence that prices were going to go up to consumers.

It’s really hard to prove that a deal, which is going to produce a massive merger specific increase in wireless capacity is going to result in higher prices. At the end of the day, I think if you were to ask the state what they have is they had some market shares and squared them, and a couple of old documents from McKinsey and Morgan Stanley and that was it. That is why they lost.

On Saber, I hear Bill that one can look at that record. I guess what I point out on that one is it was never fully litigated, had it gone up to the Third Circuit. The Third Circuit has certainly reversed defense rulings in the past, and who knows what would’ve happened. But in any event, the Saber decision is not a licensed or a good basis in my mind to sort of redo the antitrust laws as many of the challengers are proposing to do.

Just a brief sort of thought about the legislation and some of the ideas that Spencer put out there and ideas for reform. I don’t think they’re antitrust litigation reform or antitrust enforcement reform, what they are is regulation. You can say, okay, I want to regulate certain sectors or the economy as a whole, and I feel better if that’s done under the guise of the antitrust agencies as opposed to something else, but it really is regulation.

There’s calls for strengthening presumptions and making certain rules really bright line and removing defenses may be well and good. I strongly believe it’s not well and good. It’s going to hurt the economy. But it’s not antitrust enforcement. You will have to retrain the women and men in the antitrust agencies if you want to do that because that’s not what they’ve been doing for the last 30 years.

What they’ve been doing, and as the people who train me is they’ve been going out and blocking and tackling and finding evidence on the ground of how markets work. When there’s evidence of higher prices, evidence of reduced innovation and evidence of fewer choices, they win the case and they win the cases overwhelmingly. When the evidence is missing, they lose. Where they’ve lost some of those cases.

Two quick points and I’ll stop. The idea of sort of, for example, micro-acquisitions or so-called killer acquisitions, and this is a problem of large companies buying small companies. The reason the government really hasn’t brought those cases for the most part, despite intense investigation is there hasn’t been any evidence of consumer harm in the foreseeable future. It’s not as if the government was sort of asleep at the switch and wasn’t paying attention to these deals. They looked at them really hard. I’ve been involved in a number of those investigations and it wasn’t there. Both Rs and Ds have passed on those deals.

If you want to ban them, you can ban them, again it’s a terrible idea, but it’s not antitrust enforcement. What it is basically coming up with a bright-line regulatory rule and saying I think on balance I don’t want these deals to go forward. And so we’re going to remove the antitrust analysis from the antitrust investigation and litigation and skip right to the answer.

It’s the same thing for some of the pricing practices that Spencer was talking about and so-called sort of bands or changing of how you weigh self-preferencing and the like, all which I think would do some harm. Let me stop on that again. I really appreciate the opportunity to speak and look forward to listening to everyone else.


Okay, good. Here’s what I heard there, I think. You started really with sort of the economy, and you say, look, we’ve been in the tech sector maybe in particular, extraordinarily productive. There’s a reason Europe isn’t regulating their own big tech firms, they don’t have any. So, we start there. Then we go to the idea that yes, there are some prominent losses, Amex and AT&T Time Warner. But if you actually look at some of the less visible cases, the government’s done well, and the government’s done well when they’ve had the evidence of consumer harm. So, follow the question of the evidence. You want to say something, it looks like.


Yeah. Not less visible cases, high profile cases.




eBooks, Microsoft-


That’s fine.


… the Michigan MFN case, these are big-ticket wins-


Fair enough.


… which Bill and others get a lot of credit for.


Okay. And then third and last point was is that there’s a series of other ideas out there, regulation, micro-acquisitions, killer acquisitions. That’s a different conversation. Your view is not an antitrust conversation. We’ve got Michael next. Perfect. Michael, you’re up. Thank you.

Michael Kades Speaker

Michael KADES:

Well, first of all, thank you Randy and CPI, it’s quite an honor to be on this panel and with such impressive folks. Bill, since your children are at home, you can do what I do which is my daughter cuts my hair. Since we’re basically bald, it’s not that hard to do while it makes a very excellent electric shear and it’s made in America even.

I’m going to start where a little bit off of Joshua’s remarks. One, I think it’s a little bit of a hubris to suggest that the state of the economy is due largely to antitrust enforcement, one. And two, I’m not so sure we are the envy of the world. I work at the Washington Center for Equitable Growth and I do not believe I have any complex. But most of our work is about inequality and how inequality affects growth and vice versa.

For example, I think most labor economists would disagree that labor markets work well. They would say there’s tremendous amount of employer market power that has all sorts of negative effects. Macroeconomists also are continually finding that problems with the way the economy is functioning, business dynamism is way down, innovation is slowed. How much that can be attributed to antitrust enforcement, I’m not saying it’s all due to that, but I would not be so copacetic about the state of the US economy.

And so then I wanted to sort of… I don’t know if Josh said it today, but he said it the last couple of weeks. Usually, when people talk about the consumer welfare standard, I have this little riff that I hate talking about the consumer welfare standard because I don’t think it concludes consumer welfare standards really what’s been the problem in antitrust. I don’t want to die on that hill, and I don’t want to die going up that hill.

But then Joshua suggests that what the consumer welfare standard means is that you have to have harm to the ultimate consumer. That seems to me, if that’s true, then it is a disaster, an act of pure judicial activism. The antitrust laws are about the improper creation abuse of market power. Particularly buyer power is a problem if you can… Workers, if they suffer are the victims of anti-competitive conduct, that’s all cognizable.

If that’s true, then, in fact, Josh’s point that it’s all we care about is the ultimate consumer sort of proves the point that there’s something wrong with antitrust enforcement. I wanted to sort of pull back out and talk about why this debate is occurring. I think it’s interesting what you have, there was a consensus and there seems to be among some folks almost forlorn nostalgia for the consensus.

But that consensus is really broken down, and it’s been attacked from a number perspectives. So first you have sort of what I would call outside critics. So people not necessarily IO economist, not necessarily antitrust, bringing sort of very but important concerns as Bill suggested. But it doesn’t stop there. I think Spencer and last week Professor Spencer and Fox both make the point that Europe has definitely diverged and we seem to be losing that debate in the world. That should also give us pause.

We have people on the outside looking at the antitrust world and saying you’re not doing a good job. Europeans and much of the rest of the world is not adopting our approach. And you have sort of Bill Baer, and his statement to the committee is really this very… Bill’s obviously spend his entire life practicing antitrust law, been on both sides, and an insider’s view about why it is that antitrust isn’t working. There’s the deterioration of the consensus from the practitioner point of view.

I think honestly, Bill Baer’s testimony, I believe it was 2015 or ’16 where he was saying that they were reviewing deals at the Justice Department that shouldn’t have got out of the boardroom because they were so anti-competitive. I honestly think on the hill that added a tremendous effect. That totally raised concerns across on both sides of the aisle that something was wrong, and why we’re now having this debate.

The final piece of this is you have the submission by John Baker, and his group, which I was a part of. One of the most striking thing about that is those are largely very sort of standard IO economists, and they are taking the position that our legal rules do not reflect where the economics is. Not to say that there’s a consensus with that position, but that consensus has broken down. There is the outsiders compared to international, the insider’s view and now even the economists view that the standards and approach of antitrust in the United States is failing. And then I want to add to that resources.

It’s hard to think that we’re adequately enforcing the antitrust laws. Even if you disagree with all that, when resources today are roughly almost 20% lower in real terms than in 2010. Or another way to look at that is between 2010 and 2018, the economy grew twice as fast as the increase in resources. Now, that doesn’t necessarily need to be one to one, but look, if you have more people to police and you don’t ever hire more police officers, you’re going to end up not effectively enforcing the law.

The failure to recognize that I think also sort of suggests that there’s something wrong within the antitrust community. I think we’re beginning to see that between 2000… sorry. My apologies. That phone almost never… That’s the first time that phone’s rang in three weeks, of course.

The merger filings are up and between 2010 and 2018, the merger cases were flat, non-mergers enforcement between the two agents is all-time low. What’s really striking is if you go talk to the enforcers themselves, they’re all suffering from litigation fatigue. They’re just constantly in trial. They’re bringing fewer actions and those actions are taking ever more resources. I think all that it’s hard to come to the conclusion that everything’s fine and well in the antitrust world. I’m sorry Randy, I’m thinking probably done on time.


You’re fine. If you’ve said all want to say, stop, and if you’ve got more, just say a little more.


Where I think this affects… Bill talked about this, a little bit about the error cost analysis is wrong. That we have been way more concerned with over enforcement than under enforcement. I think that it is actually almost ingrained in the antitrust enforcement communities DNA at this part. Government is… As Bill Baer said, the standard is more likely than not.

I think that we’ve really underestimated what the costs of under enforcement are. At this point, we now have some examples. There’s a period of time for a decade where the government could not challenge hospital mergers based on where the courts were. There are a lot of anti-competitive mergers that got through, and they’re not getting fixed. The market isn’t going to fix that.

I spent over a decade of my life trying to being involved in showing that patent settlements in which a branded pharma company pays its potential generic competitor to accept the settlement date can be anti-competitive. It took a decade and by one vote to convince the courts that actually that kind of agreement is not per se legal. The rough calculation, the more lenient rule that the courts have applied probably $60 billion of consumer harm over that decade. All the concerns about over enforcement, there’s no evidence that any of that occurred after the Supreme court adopted more reasonable view. I guess it’s-


Go ahead.


I’ll just finish up. If we’re having problems with those easy cases, it’s very hard to imagine that the antitrust laws is currently understood by the courts are going to be capable of addressing any anti-competitive activity if it is occurring in high tech markets.


Okay, good. I heard a different view on the economy, and one situated in inequality. A sense of, I love what you said, you don’t want to die on the hill of consumer welfare. And then the question is how do we understand the consumer welfare? We had Josh right on last week, he absolutely wants to bring workers within that framework. It is a question of how you frame that.

Then this question about how the consensus has changed and the breakdown and the fracturing of that. That’s why, as you say, we’re having these hearings right now. And then I heard what you said about reverse payments, how long it took the supreme court to take that situation. Understood. Okay, good. We’re going to go to Koren next, and to Greg. Koren, you’re up.

Koren Wong-Ervin Speaker


Great. Thank you, Randy. I’ll start by saying that I represent Google. Thank you to Bill and Spencer for your statements. I read them over the weekend and they made me really think. I guess I’ll start with sort of what Bill mentioned about the concentration problem. I talked about this in I think the first session. I think there’s some problems with the studies, but mostly I think that aggregate statistics are ultimately tangential to the question we really care about, which is whether specific conduct is anti-competitive.

I want to spend most of the time talking about this notion that we have become too sensitive to type I errors. And that gave me pause because I thought about what it’s based on and the supreme court precedent like Trinko that says type I errors are especially costly because they deter the very conduct the antitrust laws were designed to promote. And the idea that markets are much more likely to self-correct over time at least then having enforcement that have a chilling effect in the market.

I asked myself is there any empirical basis to support that, and I came up empty. What I did come to is that if it’s true that we have a lot of type I errors or false positives then we should expect to see a lot of type II errors. It seems like what I’ve heard today and from other critics is that we are seeing a lot of type II errors in monopolization cases in particular. I asked myself, if they’re rampant or at least the systemic problem, why aren’t they showing up in all the retrospective analysis?

There’s a body, there’s a number of leading meta-studies including those done by DOJ and FTC economists that survey dozens of empirical papers from 1984 to 2018. That literature suggests or indicates that vertical restraints are generally pro-competitive or benign. Now, at the very least that literature does not support overturning decades of supreme court precedent and returning to the long ago abandoned presumptions of the legality.

For sure, Michael alluded to this or mentioned it, there’s been criticisms of this empirical research on three main grounds really. Some academics have said first, look, the empirical research, the meta-studies include, not all of the studies surveyed include firms with market power. That’s right but what they missed is that some of the studies do involve firms with market power.

The criticism would have greater force if the meta studies had found that when verticals are used by firms with market power, they’re more likely to cause harm than verticals with firms without market power. But that’s not what the studies find. Again, if it were really true that verticals and monopolization is causing all this harm, I think we would expect to see that show up in the dozens of product market survey.

The second of the three criticisms levied against the meta studies are that they cover limited industries. But if you look at the long list of industries, there’s dozens of them from grocery stores to electronics, to pharmaceuticals, and these are just not niche products or industries.

The third thing that the critics like John Baker point to is there’s this unpublished study. This study compares states that retained per se illegality for resale price maintenance following the supreme court’s decision in Legion with those states that follow Legion and adopted a rule of reason.

The study purports the find that following Legion their price increases were higher for household consumer goods and output grows smaller in states that follow the rule of reason. But Lambert and Sykuta have a nice paper where they point out the flaws in this study.

I mean, the first thing is that the study doesn’t show that any actual RPM agreements caused harm because it doesn’t even attempt to show that there were any RPM policies implemented in any of the products that they surveyed. The second thing is that the study excludes from its analysis or its review most of their products for which we would expect to find one of the pro-competitive justifications for resale price maintenance namely the elimination of free riding. So, the study excludes large appliances, complicated electronics, and other relatively high priced products for which are normally sold along with free rideable amenity.

The third problem is that it ignores that when you look at anti-competitive theories of harm for resale price maintenance, they predict not only higher prices but also reduced output. And only 1.6% of the product surveyed by the authors had both a higher price and reduced output.

The second thing I’ll just briefly talk about is Spencer’s proposal that we adopt a presumption of market power or dominance as he says based on market share thresholds of 50% or higher. I think there’s a lot of problems with this. For one, market structure as presently or is currently defined primarily by reference to share or ease of entry is at best a very crude indicator or signal of the likely effects that a merger or conduct will have in the marketplace.

I think the dangers of relying on share are particularly acute for multi-sided platforms. There’s also problems with zero price products, right? If you shares and you base them on value of sales, you’re not going to take into account the constraints posed by zero price goods. If you rely on unit sales, then you’re not going to take into account quality differences for which price is a common proxy.

With respect to Spencer’s pointing to foreign jurisdictions including China in support for this proposal. Even China’s Supreme People’s Court, and China’s agency’s been around a lot shorter than ours has. But even in their 10 plus years of experience, China’s Supreme People’s Court has said that there are dangers in relying on market shares especially in dynamic markets.

In the court’s 2014 decision involving Chinese internet company, Tencent, the court held that Tencent’s long-held market share of greater than 85% could not be too heavily relied upon. The court said instead, we’re going to rely on more traditional and economic indicators such as did the platform have the ability to raise price or reduce the output? And they also said, they’re going to look at actual competitive effects.

I know one of the things Bill said in his statement was moving more towards whether things are there’s a risk or likely to cause harm. I find that problematic. I think we should actually have evidence of substantial foreclosure that deprive rivals of minimum efficient scale, evidence that they’re actual, especially when you’re looking backward at conduct where you can actually compare to see what actually happened in the marketplace. Did prices go up? Did output go down?

I think with a lot of the, or some of the so-called killer acquisitions we point to, some of those markets are thriving, output is exponentially growing. It seems odd to me to point to that and say, “Aha, that,” because we normally think of exponential output growth as pretty good. In the interest of time, given there’s chance, I’ll stop there for now.


Good. That’s good. Let’s see what I heard there. I guess what I heard there was sort of a defense of the current framework, especially with regard to where we’re drawing lines on the type I errors and the type II errors. A sense that if we look backwards, we ought to see a bunch of missing cases and problems that should have arisen, and we don’t see those is what I heard. A really deep dive on verticals, good for you. That was really interesting.

Then at the end, a discussion of how we should think about market share figures, especially as applied to two-sided markets and whether those figures don’t really capture the nature of market power in those situations. Good. We’re going to Greg next and then to Edith. Greg, you’re up.

Gregory Werden Speaker

Gregory J. WERDEN:

Thank you. I worked in the antitrust division of the Justice Department for 42 years before I retired last year, but at present, I have no affiliation. I wouldn’t say that I have any conflicts, although I have done some consulting for a tech firm.

Quick note, I agree with Michael Kades, and this is in Bill’s testimony, but he didn’t mention it in his remarks that we should devote more funds to antitrust enforcement. My submission to the subcommittee suggested a 15% increase in the antitrust division’s budget each year for the next five years. That would result in a doubling over five years.

On the other hand, a few years ago, when I looked at the historic data, I was shocked at how constant the funding had been since 1940. If you break it down by decades since 1940, antitrust division budget has been 1,000,000th of the GDP in every decade.




1,000,000th must be exactly the right amount or half as much as it should be. I agree with Bill and many others who say that courts often demand too much from plaintiffs. Many judges are in fact overly skeptical of claims that business conduct other than hardcore cartel activity is anti-competitive. I made this point in my submission to the subcommittee and I made it in a book I published last month. I saw this in some of the cases I worked on at the antitrust division. I saw it in a lot more of the cases I read as part of my job when I was in the antitrust division. Historical research has taught me that it’s not a recent phenomenon.

The Chicago School, especially Frank Easterbrook rationalized and institutionalized his skepticism, but they didn’t invent it. It goes back to the 19th century. Where I part company from Bill and others is on how to deal with the skepticism problem. I don’t think tweaking substantive antitrust laws is going to work. Because it can’t prevent an excessively skeptical judge from finding facts that are fatal to a plaintiff’s case. Those findings will be exceptionally difficult to attack on appeal.

On the other hand, tweaking substantive antitrust law will create enduring uncertainty. There’s no assurance of a satisfactory resolution of that uncertainty over time. Would help a lot to get useful direction from the supreme court, and we’re not getting it. The American Express decision is part of the problem, not part of the solution.

My letter to the subcommittee cited the Foreign Trade Antitrust Improvements Act of 1982, an example of what not to do. That was intended to clarify which claims involving foreign trade can be brought by which people under the Sherman Act. Hasn’t really accomplished its purpose yet. It’s considerable uncertainty remaining after 38 years, and I think it’s going to take three more trips to the Supreme Court to settle things. I’m not sure they’re going to settle anywhere near where they are right now.

My letter to the subcommittee offered a procedural solution, a radical one. I suggested that government merger cases be submitted to mandatory arbitration. I proposed that the facts be found by a three-person panel of which two would be economists, taking judges out of the loop is the solution.

I also proposed that section seven be amended to require written direct and rebuttal testimony by economic experts in all cases, whether tried before judges or submitted to arbitration. It is not possible for a federal district court judge to understand what the experts are saying. I know, I’ve been in the courtroom and I know a lot about the subject and I can’t understand what they’re saying.

I don’t agree with very much of what Professor Waller has proposed. I do not agree with the core idea that we should try to make US substantive antitrust law more like that of the rest of the world. I regret that when we could have had an effect on competition laws in other countries, we stood aside and let Europe proselytize their version of competition law. That’s why it was adopted, can’t go back.

The problem with European approach is that the European courts haven’t articulated the necessary limiting principles. I don’t think they ever will. They don’t like being in the business of articulating limiting principles. They have done it for a few particular abuses, but they haven’t done it broadly and generally. They haven’t annunciated what it means to protect competition rather than competitors. That’s a problem.

Of course, if we did what Spencer proposes and add use of dominance offense to US law, our courts would get plenty of chances to draw limits because we would get a lot of cases at least if we retained the private treble damage remedy. But I think what’s really going to happen as we make billionaires out of plaintiffs.

Courts outside of the United States have done almost nothing to articulate limiting principles for exploitative abuses. I think it’s just a bad idea. Congress got it right in 1890 when it decided to prohibit conduct that creates or maintains monopoly, but not to prohibit lawful monopolists from charging high prices. That’s what monopolists do. The dream of monopoly pricing is a critical driving force of capitalism, at least American style capitalism, which I agree with Josh is working. Maybe not for everybody, but that’s not an antitrust problem.

A few other responses to other panelists. Well, I don’t agree with Michael Kades on a great deal. I agreed with his comments on the consumer welfare standard. I think it’s much misunderstood and much abused. The antitrust is about competition, not consumer welfare. If our eye is on the consumer welfare ball rather than competition, we get it wrong a lot. I don’t agree with his one sentence on dynamism. I’ve looked into those data and believe it or not, the data that are being relied on omit 90% of the new firms in the US economy, 90%. So, don’t trust those data.

One vignette, I love stories, related to Koren’s comments on RPM. When I worked on the Legion case, I read everything that had ever been written on resale price maintenance, I think. Including the entire hearing record from the Consumer Goods Pricing Act of 1975. I roared with laughter when I read the testimony, somebody whose name I can’t remember right now who followed a witness who presented an empirical study comparing prices in drug stores between Arlington and DC. A kind of study, by the way, the antitrust division had itself submitted to Congress on two prior occasions.

This gentleman said that the problem with this study is not a single one of the products in the study is subject to RPM. That can’t be the reason the prices are higher in the RPM state than they are in the not RPM state. Don’t know what the reason is, but it’s not RPM. That is illustrative of the quality much of the research people are relying on now. They make assumptions. They’re not really usually testable. The assumption was that lots of the products, most of the products sold in drug stores were subject RPM, and they weren’t.

There’s just so much we don’t know and too much now thanks to the internet. A little bit of data is going a long way to convince people of things that probably aren’t true. I do think we have some problems in antitrust today, not the drastic problems that people suggest. There isn’t an easy fix, making judges think about antitrust cases differently is hard. I’m not sure how to do it.


I heard a number of things there. I guess I heard the iron law of DOJ financing. It’s 1,000,000th of GDP, and it’s been that way forever. The skepticism that judges have towards the antitrust pre-Chicago, you say, so don’t blame that or credit that to Chicago, but a deep skepticism that goes back. An idea that actually it’s really hard to understand things. We need more economists and we need better data because the data we have right now is just not doing the trick. Okay, good. We’re going to Edith and then we’re going to Chris, and then we’re going to circle back to Bill and Spencer. Edith, please.

Edith Ramirez Speaker


Thanks, Randy. Let me just start by saying that I’m also really delighted to be here with this esteemed group. I’ll start by saying that I am a private practitioner. I do represent some tech companies, but I did also spend seven years at the Federal Trade Commission. I think I bring a broad perspective to these issues, and I am speaking in my personal capacity.

I’d like to start off, I think, by echoing I think some of the comments that Michael made in terms of hubris. I think this broader debate about antitrust is certainly a very positive development. But at the same time, I do think that too much is laid at the feet of antitrust. When I think about the economy and the issues and concerns about dynamism, I really am frankly much more concerned about how we’re educating our children. I’m very concerned about inequality, access to capital. I’m really also very concerned about the impact of technology on labor markets and in particular AI. That’s going to be tremendous issues.

I do think that a lot of these issues do need to be confronted elsewhere in the policy domain. But let me turn back to the topic at hand and focus on and give a few comments on antitrust, react to what’s been said. I’ll make just a few observations given the limited time.

Number one, I do think that courts make a difference. They may not necessarily make a dramatic difference in areas where there is tremendous consensus, and that’s generally the case in the merger arena, but there are certain areas where I think courts do get it wrong.

Future competition is one example of an area that, I think, where certainly in my experience at the FTC with the Steris-Synergy case, I felt that the court placed a burden that was very hard to meet on the FTC. I think there are areas that would be one example in the merger arena where I do think that the courts can very easily get it wrong. Section two, there’s no question, in my view, that enforcers are constrained by existing case law in section two. That absolutely makes a difference as you’re considering what actions an agency might bring.

I also want to echo the comments that have been made in support of studies. There are a number of different areas where there’s a significant amount of internal debate that takes place at the agencies that can also hamper enforcement. It really, in my view, is important for agencies to engage in retrospectives for the FTC to be using it XP study.

Those endeavors, however, take a lot of resources and a lot of time. The agencies need much more resources in order to be able to truly engage in the type of work that I believe is necessary to address areas where additional work is absolutely needed and certainly the digital arena when you’re talking about platforms. There’s a lot that we still don’t understand well, and I think it’s imperative and incumbent on the agencies to undertake that work.

Josh cited the example of hospital mergers, that the string of victories came after, of course, an extensive efforts to take a close look and engage in retrospective analysis of the work that the FTC had done in that area.

Let me also talk a little bit about the need for consensus. In my view, when we’re thinking about reform, I would like to see a lot more consensus both between the two agencies. Also, frankly, it’s a very divided era at the FTC. It’s concerning to see such wide chasms amongst commissioners. In my view, engaging in studies and trying to really make sure that any proposals for reform are really based and sound.

Facts and understandings of markets, I think that could help to bridge some of the really divided views that we have. But in my view, it’s important to have that because only then by getting some level of consensus at the agencies in my view will efforts to move antitrust forward. Only then will it succeed in my view. Only [crosstalk 01:08:30]


Good. Here’s what I think I heard there. I heard the idea that maybe we shouldn’t ask too much of antitrust, but there’s some other issues out there, some really important things. We shouldn’t necessarily think that antitrust is our first tool of response there. Some situations where you think the courts have gotten it wrong.

What I heard that I thought was really interesting was this idea that if we do the sort of hard work, and the FTC you say did this in the hospital mergers area of doing the studies, figuring out what’s going on. That once they understand the facts that can drive both the litigation strategy. But also maybe you help to achieve more of a consensus internally in the agency.

I thought that was a really interesting point on strategy. That’s why I like that a lot. Chris, you’re up next. And then we’re going to go back to Bill and Spencer who have had a chance to listen, and we’ll have things to say, undoubtedly. Chris, you’re up. You’re muted, yep, but you know that.

Christopher Yoo Speaker

Christopher S. YOO:

Thank you very much. Just one point of clarification; I actually go by Christopher because I married a woman named Chris.


Thank you so much, Christopher.


Just to avoid confusion.




Delighted to be here. Thanks for having me. In terms of conflicts, I actually do not do any direct consulting on antitrust matters. I have done some work in privacy and some regulatory matters. The work I’m doing in antitrust right now has focused largely on due process, which is a pretty vanilla argument as it goes. Although we all are probably on the verge of Ken getting a grant to study the economics of big data from a foundation. And so that’s not… I can’t go to the details because it’s not done. And until it’s done, you never know. We actually, the seller one takes support from a broad range of actors who are on different sides of these issues. I’m happy to talk. If all of that’s disclosing on the website, I’m happy to go to that.




What’s interesting to me is I’m really resonating with what Greg is saying about the need for limiting principles because what I see is this weird stampede to judgment. I really see it at the international competition network when I sit there. You see these old things about instead of ready, aim, fire, it’ll start with fire, aim, ready, aim.

It really strikes me that there is a lot of undisciplined analysis where we just think that collapse of the idea that digital markets are bad that I think are not helpful. To give you a good example, I spent a fair amount of time reading the various reports that are salient now, the Furman Report, the ACCC Report, the DG Comp report, the Stigler Report.

There’s a thread in there where they say, oh, network economics affects leads to winner takes all markets and therefore is a sort of market concentration of market failure. Well, just eyeballing it, that doesn’t seem right to me in the sense that for one of the things that they’ve been looking at, for example, is travel sites. Travel sites do not, multiple ones seem to be existing on sort of a arbitrary parasitism.

But when you get right down to it, basically there’s a long-standing critique that if you basically posit inexhaustible returns to scale, we know on the supply side, it leads to market failure and you get natural monopolies. Guess what? That happens on the demand side too.

If you look at the models, many of them assume basically that there’s this constant returns to scale or over increasing returns to scale even with the decay. And in that sense, sometimes the market failure is seen as a product of model or even just as much so many of them assumed constant zero marginal cost, which is essentially are saying assume market failure.

And what strikes me is there’s an interesting question is, is there no diminishing marginal returns to scale? Are there no diseconomies of scale on the demand side? And there’s a nice literature going down to the computer science network side that says, look, if you have access to 100 online eCommerce sites, is there really returns to scale to another one? Or if you lost phone contact with a large part of the world right now, given our actual calling patterns, the fact that we prioritize certain things more than others, that doesn’t happen.

There’s a nice literature that says anytime you have multiple places you can go, you can organize them at high value to low value, which creates a natural decay of scale based on which ones you value more. Many of the assumptions are that increases in the number of connections you can make, each of them contributes the same and therefore adds to value. That’s a nice theoretical critique.

I’ve actually been looking at the empirical literature to try to back up this stuff, and it’s generally not there. What you generally see is people who just sort of assume that network effects leads to market concentration, and inevitably leads to winner take all markets instead of actually analyzing real markets. And it’s based on pure theory.

This is what I’m going about, sometimes what Greg is doing it and what Koren was saying, theory is important and it’s a lot of what we use initially and it shows what can happen. But unless you’re disciplined by your empirics eventually, anyone can make an argument about what’s theoretically possible.

So maybe of the modelers are just so gifted now because they’re just about anything is theoretically possible under certain conditions. And then the question becomes what are the validity of those conditions and how robust they are. That just hasn’t happened.

What really strikes me is we’ve forgotten the long standing critique we’ve had of the structure-conduct-performance paradigm, which is we wanted to measure effects, but we couldn’t measure them directly. So we said, we drew an inference, certain structures lead to certain kinds of conduct, which leads to certain kind of performance.

And now what we’re seeing in the model of empirical antitrust, we’re actually able to measure performance directly and you see this in the horizontal merger guidelines. If you’ve got direct evidence of market performance, you don’t even need to define a market. In some sense, we need to take that critique back to the digital markets because in a lot of times they’re saying we have network effects therefore we have bad performance. It seems to me to be skipping the same steps or shortcutting the analysis in exactly the same way that we criticize the structure-conduct-performance analysis to go.

Two sort of larger thoughts, one responding to something that Spencer said. Looking at sort of the European style and not so much looking at bad conduct, it really brings back the conversations we’ve had about no-fault monopolization. I think about excessive pricing, I also think about the EU doctrine of collective dominance, which brings us to conscious parallelism and tacit collusion.

There’s a bunch of interesting theoretical problems, but the big one, one we always had is remedial, which is if you’re penalizing someone for engaging in profit-maximizing behavior, you’re going to have to write a remedy, which basically tells them how irrational you would like them to act. We have to understand what’s a fair assessment here requires us to assess ex-ante risk in ways that have always made us extremely uncomfortable because price setting is something that antitrust cohorts going back to the Trenton Potteries and everything are very, very bad at doing. So, finding a lot of the hesitation there has been trying to figure out how to solve that.

The last overarching theme I’d like to mention is a discussion broadly about the role of agencies. Spencer talked about administrative enforcement, I’m thinking about I’ve got some writing. Actually, administrative enforcement in EU is under some fire because a right to fair trial requires a court to look at things and give it to novel review.

In fact, what happens is in the agency, there’s a classic problem where you have the same people both developing a prosecutor in the case and judging it. If you’d have differential judicial review, then you don’t actually have the separation between prosecutor and jury that is generally required by most principals. No one can be a judge in their own cause.

The other thing I’m thinking is about administrative enforcement, a bunch of things. There’s an interesting question about faith in industry studies and having agencies play certain roles. I see a lot of different things about this, which is what are the issues? Spencer mentioned price squeezes that’s often where you have a regulated wholesale price that’s too high that doesn’t allow them to make profit on the retail level.

I always understood that to say, well, what you’re really saying, either A, the regulator got it wrong and set the wholesale price too high, or B, I’m not efficient enough to operate and survive there. The remedy would naturally be to go back to the first wholesale price center and do it. It seems as it’s very complex interaction between the two, which price squeeze cases don’t really overlook that we don’t really consider.

Two last thoughts on the administrative action. One is, having looked at industry studies, I do think they can get politicized and they don’t necessarily do great analysis. One of the things that was in the ACCC Report that I was thinking about is one of the things they kind of confront is whether online and traditional offline advertising are substitutes. It seems to be in terms of market definition, understanding competitive effects core.

They looked at two studies. The empirical literature is very thin, but there were two studies that basically concluded they did act as substitutes under those circumstances. But they said, look, we’re not saying… they did a caveat saying at the level of antitrust market definition, we’re not going to say that because we don’t really have a SSNIP test. We’re not doing a 2% test, we’re not doing the usual things.

ACCC said nothing in those studies disprove that they could be. And therefore, we’re going to say that they are. Now, for an agency that bears the burden of proof, it’s taking what was a caveat and saying, well, we’re not claiming this is an antitrust market, for sure, because we don’t actually have a test. It was against the spirit of what they’re concluding, but frankly, it was a pretty cynical manipulate sidestepping of what I thought, what should have been a real empirical analysis.

The last comment goes to Bill Baer’s comments about guidance. That’s another role for a string of agencies to come in. I can’t help but be struck with how the merger guidelines have really killed judicial development. Bill had a hope that it would actually spear the courts in reconsidering the doctrine.

What strikes me is the guidelines have really made the agencies driving them and to many of us think has killed the judicial cases because when it runs across public enforcement, they just pull them. In some ways, this turn of guidance wouldn’t necessarily help courts, but actually would make the agencies that drafted guidance the primary locus for determining law.

There’s just a whole discourse you can make about whether that’s good or bad in the nature of agencies versus courts. But that to me has to be part and consider the role of the agencies are and possible expansion of them. It has to be something we have to think about.


I heard three quick things there. I heard the idea that maybe we’re making assumptions about the structure of winner take all markets. We actually need to look at those markets in reality more carefully to see whether those assumptions are borne out. As to Europe, the question is what remedies would you actually implement? This court system really well situated to do that. And then third, some pushback on agencies. Concerns about politicization, concerns about combination of, as it were judge and jury and the like. So maybe that’s not the right approach. All right. We’re running a little long. We’ve got let’s say five minutes. We’re going to go to Bill for two and a half, we’ll go to Spencer for two and a half. I’ll say, bye and we’ll be done. Bill, you’re up.


I thought this discussion was terrific. I really appreciate the comments and thoughts. Let me make a couple of points. Edith makes a key point, Randy, that you summarize, which is that there are huge public policy questions relating to income and inequality, privacy, all sorts of things that aren’t necessarily best solved by antitrust enforcement and probably in many cases are not well suited to using competition enforcement.

There are things that one can do using competition principles and regulating. You look back to that was part of the theory of the open internet rule was to try and basically deal with bottleneck issues. You look back to 2003, I think it was when the FTC basically used competition principles with some push from Congress to say guess what your phone number can move with you, which hugely facilitated competition both for landlines and in the growing mobile market. There are things that can be done.

A lot of this debate was talking about how much scholarly research do we have that proves there’s under-enforcement, over enforcement. I go to Greg’s point, which is we may need to invest a hell lot more in studying the economy out of both agencies and maybe elsewhere. We obviously do need to know more.

But the question comes back to type I, type II. What do you do in the frigging meantime? What do you do when you got to make decisions? There is I think those who think we need much more conclusive proof who would say in the meantime you just under enforce, you let it go.

Josh talked about the agency’s successes of which there have been some in recent years more than in the ’80s, in the ’90s. But some of that comes from only picking up the ones where the level of proof is so high that a court has a hard time saying no. I don’t think it answers the question of the courts imposing too high of a burden and that causing the agencies themselves to err on the side of not bringing the case that potentially creates a bad law. I’ll stop there.


Good. Thank you so much. Spencer, you’re going to get the last substantive word. Go ahead.


I appreciate it. Thank you all. I appreciate the comments. Koren, just to be clear when I was talking about presumptions, I can’t speak as to how every country around the world does it. In most cases, those are rebuttable presumptions, and I think that’s completely appropriate if the market shares don’t reflect traditional notions of power for whatever reason, fine, the court should hear that. That’s more of a burden shift than a irrebuttable presumption in terms of what I was talking about.

I want to push back on a couple things. I think the United States has done a very poor job of learning from abroad, comparative law in general, and antitrust. I hope we can do better. I think it’s not fair to call the approach outside the country as no-fault or just regulatory. They require an abuse of a dominant position. We can disagree whether the theory is good or the application is right.

But they’re doing a traditional antitrust analysis. I’ve worked with countries as diverse as Egypt and Chile and New Zealand. They don’t view the caseload part of their mission as a regulatory, they view it as antitrust. I would imagine it’s no different than all of us who’ve been involved in private litigation on these types of theories that are just favored. I think they should be brought back. Other people obviously are pushing back on that. But they’re antitrust and they require litigation and they require facts and they require legal theories. I’m arguing for a broader palette of legal theories.

My last point, Randy, is just to respond to Greg a little bit. I think the problem isn’t that we did not proselytize our view. I think we have over proselytized our view. We have spent several decades, more or less pushing the view that we got it right and other countries who want to do real antitrust should do it more or less our way. We continue to do it to some extent in the ICN, and we’re not getting a lot of takers.

I think we have a natural experiment of what’s going on in the rest of the world with some really smart well-trained people who understand what we’re doing and are choosing something differently. I think we can learn from that, not necessarily all of that.


Great. Thank you so much. We’re spending Fridays in July talking about how submissions and their investigation of digital marketplaces. What a better way to spend Fridays. It’s not possible. Thanks for everyone being here today. We’ll be back one more time next week. Thank you. Bye-bye.