This episode is the "New Guidelines for the New Economy" panel at the Revising the Antitrust Merger Guidelines event. It was recorded on September 29, 2023.
It features:
Announcer 0:00
Welcome to Engelberg Center Live!, a collection of audio from events held by the engelberg center on innovation Law and Policy at NYU Law. This episode is the new guidelines for the new economy panel at the revising the antitrust merger guidelines event. It was recorded on September 29 2023.
Daniel Francis 0:26
All right, well, welcome to our third and final panel, which I think is called new guidelines for the new economy. That is code for the tech panel. Yeah, so my name is Daniel Francis, I teach anti trust here. And this is an unspeakably distinguished group of humans. You'll notice the how Chilean skis seat is vacant, he is battling traffic by the sound of it on a surfboard coming in from Brooklyn, he will probably get here in some number of minutes from now less than 15. So you know, it's not just a power move, he has told us that he's, he's running a little late, but we're gonna go ahead and start on time and welcome him when he arrives. So I could spend a long time introducing this fabulous group. But I'm going to keep it to one or two sentences each so that we can jump right into the heart of it, my love languages and trust. So I'll just introduce beginning with Harry Harry, of course, is the Charles Denison, professor of law here at NYU. And our featured honoree for this panel. Eleanor Hoffman, is the chief of the antitrust bureau in the New York IGS office. Doug Melamed is a visiting fellow at Stanford and now a scholar in residence at USC. Howard Szalinski, who will be with us shortly is the Joseph and Madeline she chair in antitrust law and trade regulation at Georgetown and a partner at Davis Polk and Spencer, we have a Waller is the Justice John Paul Stevens chair in competition law, and is returning from a tour of duty at my beloved FTC. So we're delighted to have assembled this group. Let's dive in with a question about the state of the world before the 2023 guidelines hit the world. And that's by assessing how we were doing on tech merger enforcement under the 2010 guidelines. So the current administration has brought some technology cases right meta within Microsoft Activision Illumina Grail, we might think of as a tech case, and the NVIDIA ARM processor case where the party's abandoned. And this follows a pretty lengthy streak now of interest and activity at the agencies in tech merges. In the last administration. We had Facebook, Instagram and WhatsApp, Visa plaid save a fair logic, a whole bunch of other tech merges. This has been a frontier for a long time. So how is this administration doing on the technology front? How's the 2010 framework holding up? Has anything in there been holding us back? Harry? What's the state of the world before 2023 draft comes along?
Harry First 3:21
rainy and cold. So thank you, Daniel, of course, thank you for organizing this. And Scott hempco, reminding me I should thank the organizers for not only putting on this wonderful thing, but then forcing me to work by being on the panel. So but I'm glad to do it. Actually, I want to take a little just a step back. I've even ever said in our morning session that oh, you know, not much doing here, you know, nothing much is happening. That much changed from 2010. Which led me to wonder why they didn't just, you know, do a little cut and paste and change a few words and be okay. Eleanor's view, I think was a little different that these are being changed in that's my view is Well, I think Dan's view as well. I think everybody's view as well. These are a change. The only question is what do they mean? But that's another story, which we're trying to figure out. And I'm trying to figure it out as we go along. So I do want to, you know, your question about how are we doing is sort of is also a more general question. I think if we, as listening to the discussion as sort of merger professionals, we might think that merger enforcement has been just fantastic. over all these years that we've done a really great job. But I think and I think that these guidelines are responding to criticism and hasn't been fantastic, bad It's it's had a lot of failures. And that the but Howard did get here. So that's good. And that these guidelines and the current enforcement in the Biden administration is responding to the criticism of where we've come to. So when you say, how are we doing? And you mentioned, Instagram and WhatsApp, I thought that in the past, under the old guidelines, nobody brought suit. So I don't think we were doing very well at all. And what I want to suggest is that these as we think about the guidelines, however they might be, and whatever the final might be, that we think about them against what I think are the three intertwined purposes of this administration's enforcement policy, or what I would like to call progressive merger enforcement policy. So one is a concern for increasing concentration, concentration. And that's why guideline one is guideline, one. By the way, once we start calling the guidelines by numbers, we know that we're not going to have this document read by our parents. This now becomes an inside game. So guideline one talks about concentration, be able to tell you about guideline five should be put to guidance, a guideline six, I have to go and look and see, which is five, what are they what are we talking about. So this is a technical exercise, we've talked a little bit, I think, on this panel about who the audience is, we've talked about it, but so the first the first criticism is about concentration. And this is started in the mid 20. Teens, I guess we should call it Council of Economic Advisers report on concentration and competition, and has been surrounded by a lot of economic studies that a lot of industries and markets have become more and more concentrated, because merger policy has let too many mergers go through. And it has all sorts of bad effects. You know, margins have gone up, entry hasn't actually occurred. We have skewed distribution of income, all the bad effects that are coming through this. So that's the first part. The second part is concerned for firm size, the growth of firms just becoming larger and larger. And the third strand of this policy, I think, is concerned for the rise of the big tech platforms, which is part of our discussion now for this panel, and their power, their control of data and their ubiquity. So these are the things that I think are animating, and then I'll I'll add a fourth enforcement reason. We have a lot of trouble dealing with monopoly. There's a lot of trouble legally, there's a lot of trouble and litigation. The idea of a strong anti merger, I'll quote Peter Carson said any merger policy is so we don't get to have to deal with monopoly. We nip it in the bud. We stop it before. We don't have to deal with tacit collusion if we have less concentrated markets, the idea that we should do it before. So and many people have written about this rich Gilbert has Lena Khan repeated this and seven for addresses. It's important to have a strong merger policy. These guidelines, I think, from the outside are not intended to be incremental change from 2010. They are intended to turn the ship some people might say blow the ship up, but I think turning the ship and they're meant to say we're doing something different. Now what exactly that is another question. Yeah. And I do want to, I want to highlight this by reading to you, paragraph 65 of the Amazon complaint. You know, Amazon, anybody use Amazon? Okay, we're attacking it. Now. You know, whenever I would teach it, you know, students would say do anything you want monopoly, but don't take my Amazon because I need diapers tomorrow. Okay, but here's here. This is in the monopoly complaint. Amazon is one of the largest companies in the world ranked among the five largest publicly traded companies. Amazon's business spans vast portions of the American economy extending from its core of online retail into media cloud computing, brick and mortar, grocery stores array of logistics and operational service expanded in part is expanded in part through an acquisition spread, buying up more than one 100 companies and sectors spanning entertainment grocery health care, its reach ranges from selling socks and making movies to running a pharmacy and operating data centers that house exabytes of data that's in the Amazon complaint. And I think what these guidelines what I think a progressive merger policy is trying to do is to not have Amazon Amazon's in the Amazon ik sets, get it, there's a play on Amazon. So I think that's the lens to look at it. Now one of the criticisms that people have made about the, this this draft is that it doesn't have any examples. And the 2010 guideline and examples of you know how this would work out. There were a few of those. I never trusted the examples because as a teacher, when I do hypotheticals, you will know this, you create the hypothetical to get the result you want. So, you know, I'm not sure what they teach. But we do have the three cases, or I'll mention three cases that Daniel mentioned, matter within these are three recent cases. And your question is, how did they fare? How would they fare under the guidelines matter within this? It's true if any of you do virtual reality, for virtual reality. Okay, first you put on those virtual reality Oculus glasses. And so, meta notice that changed its name, because they want to own the metaverse and meta has the hardware, the App Store, you know, the whole ecosystem, not a concept mentioned in these guidelines, by the way to move into the metaverse. So they acquire this fitness Virtual Reality fitness company called supernatural 800, fully immersive VR workouts. This is the case on which the Federal Trade Commission climbs the hill on potential competition. But it was really about this concern that gets reflected in the Amazon complaints. The effort to expand its empire. This is an Commission's complaint. One step closer to owning the metaverse so a concern for expanding out the system that
that Facebook then But then became meta had, how do you fit this within any trust doctrine? Or how do you then accommodate this within new guidelines, first case, Illumina Grail? Illumina has this platform for gene sequencing. It's a pretty much unique the gold standard for gene sequencing. Grail is developing this cancer test, literally the Holy Grail, it's going to be a one shot blood test. Early detection of 50 different cancers. They're developing it they've sold 900 hasn't been approved by the FDA yet. Illumina acquires Grail. How do we handle this sounds like a vertical merger like we're familiar with. And the FTC uses the incentive and ability test incentive to foreclose other makers of the test from access to the Illumina platform. And the third is Microsoft Activision, how many of you play Call of Duty? Great. I impressed my 10 year old grandson by telling him I was working on something writing about video games, he has not forgotten this, you know, so Call of Duty is, you know, it's earn $30 billion in lifetime revenues. This is this is like real money. So Microsoft, which has the Xbox and has video games acquires this upstream producer of this really important franchise, Call of Duty, among other things that Activision does. And you know, what's going to happen to its downstream to Microsoft's competitors downstream, like Sony, or Nintendo? In case you don't know, these are also game consoles. But also, the industry is moving towards streaming streaming games, which is where the future looks like it's going to do. So how to how do we handle that? And I think Spencer is going to say a few words about other jurisdictions. Eleanor mentioned that these are cases that appear in other jurisdictions. as well, not just the US Not, not the metal within but this case done in the EC, the EC said European Commission said it's illegal. And the FTC litigated this administratively. We haven't touched on that difference. All our discussion has been about, I think litigating the district court, rather than before their own administrative court and found it to be a violation of Section seven now on appeal before the Fifth Circuit. So that's in process, how should that be dealt with? Sorry, that that was aluminum, hail, Microsoft, there's so many of these cases Microsoft, Activision. They did, the Federal Trade Commission did go into District Court, lost in district court. But this was also subject of enforcement in the UK, the CMA said it violated their competition law, same, same standard, basically the see sane, and said that they were in violation. But and this is a thing that we mentioned before. Also, there's a deal in both cases to allow downstream use. So we have to think more about how those deals are handled. This is a major problem, which I agree from a morning discussion would be nice to see reflected somehow in these guidelines, without waiting for new remedy guidelines, which were withdrawn in 2022. So I'm gonna just stop there, maybe people will I don't want to take up all the time, people will answer some of these hypotheticals. They these cases are handled to some extent under the guidelines. But we can discuss this we go.
Daniel Francis 17:04
So our format is going to be that for each of these big picture questions, we've got two folks who are going to give sort of framing reactions and then we're going to have free interchange among the panel. So Harry's just given us a tour of the tech merger frontier as it stands. Let's hear from Eleanor.
Elinor Hoffmann 17:22
So first of all, disclaimer, the views that I expressed here today are my own only and don't necessarily represent the views of the Attorney General of the State of New York or any member per office. And on a personal note, but I'm sure the attorney general would agree. I am really honored to have been asked to participate in this conference. It's been fabulous. So far, I haven't checked my phone once I've been too absorbed. So I don't know what's going to happen in my office. And I'm especially honored to be sitting next to Harry first here on this panel, who I might add, was a predecessor of mine, actually, before I joined the chief's office as chief of the antitrust bureau. And in the audience, I see two former ag sorry, bureau chiefs as well, Scott Hampel and Jay Hines, I really am trying to fill very big shoes. So, Daniel, the first question you asked is, How is the administration doing? And I think they're doing well. And I think the reason they're doing well, is because they're bringing cases that push the edge of the envelope. While I think trying to stay consistent with existing law, I think that's the job of an enforcement agency, that enforcement agencies don't have to worry about being profit maximizing entities. Their litigation is not generally I was a large in large firms for 25 years. Litigation is not generally a profit center. Doing business as a profit center, introducing new products is a profit center innovating can be important to getting profits back into the firm, but litigation is generally not. So the agencies are doing their job. We can disagree with their risk assessments in some cases. They have lost a number of cases. I think someone said this morning that the DOJ has brought four merger cases in the last three years or something and lost three out of the four. That's probably true. The FTC has brought some very challenging cases. I'm not going to speak about Amazon because New York is a plaintiff in the Amazon case. But I don't necessarily disagree with what Harry said. You know, Amazon has been called publicly The Everything Store, which I think is a good description. So, the other question you asked Daniel is, has anything in the 20 twin guidelines been holding back merger enforcement? And again, I think the answer is no, I at least I don't think so. I view the current proposed guidelines as building on the 2010 guidelines and its predecessors. Yes, there are differences. And there are significant differences that have that, I think, reflect the agency's learning over the past bunch of years in new markets in particular. And also, you know, the guidelines are there to, as many have said, to tell the antitrust community and the business community, what the agencies are thinking, what questions are they asking, how are they framing their theories? What kind of how do they do their risk assessment? How do they decide whether a particular merger is more likely than not to lead to a lessening of competition? So I think the guidelines are important from that perspective. So let me just touch on some of the ways in which they differ. And these have all been mentioned at some point today. They're structured differently. There are some guidelines that are new, like the entrenchment guideline, number seven, if I remember correctly. They don't, for the most part, draw bright lines between vertical and horizontal mergers. And they fight case law. So just let me touch on the last point, when I saw they cited case law, my first reaction is, yeah, why didn't the other guidelines like case law? You know, we are after all, lawyers were bound by the law, were informed to a great extent in any trust by economics and economic learning. But essentially, we're lawyers. And, yes, the case law they cite, include many cases from decades ago. Perhaps there should be inclusion of more recent cases. I'm sure the agencies are thinking about it, because that has been kind of a constant criticism of these guidelines. But I'm really happy to say case law actually. Let me just talk a little bit about structure. So when I first saw the guidelines, I did see, okay, this is how you litigate a case, prima facia case, you know, rebuttal. And that made sense to me, listening to some of the criticisms, reading some of the criticisms, it perhaps might be useful to bring some of the rebuttal stuff upfront. And I say this for one particular reason. So I think the guidelines are useful in to the antitrust community when there's going to be an investigation. And I think they're useful when there's going to be litigation, but there's a difference. When firms come to us and they come to us just like they do to the federal agencies to talk about a merger. They come in with everything, you know, and very often we hear about the efficiencies before we hear about why there's no anti competitive effect. So there's no real distinction. There's not like the same kind of burden shifting, at least, as far as you know, what facts what theories are presented to us during the investigation. A court on the other hand, has a framework of precedent, including burden shifting and when they write their opinions, they're going to talk about whether the government made out as prima facia case, and whether the firm's rebutted that prima facia case. So I think, you know, there there could be a little bit more about what considerations The agencies have worked into the prima facia case. There has been some talk about there's no mention of elimination of EDM, elimination of double marginalization. Yeah, well, you know, maybe that should be moved up. My personal view on that aspect of it is unless you condemn Straight pass through. Shouldn't be really nice, simple. But so I think there's room for improvement. But I think the agencies have done a good, credible job. And then finally, I guess I'm puzzled by some of the comments that I've read, that criticize the guidelines for introducing new or dredging up old theories. And there seems to be a resistance to change. So to me, antitrust law is all about evolution and change. So in my career, I've been stressed, I've been aggravated, I've been challenged, but I've never ever been bored. And that's because there's a constant theme to anti trust. And it is that there are no bright line answers. And nothing, you know, over a period of decades, nothing. Nothing stays the same. So some of some people earlier today talked about the introduction, the 1968 guidelines, those came before my time. But I was a young associate when the 1982 guidelines were introduced. And I literally remember being at a New York state bar association meeting, when people were yelling at each other. Because it was such a radical change such new novel ideas. Some of the partners at my firm, who and it was largely a defense firm, were enraged by the new guidelines. So change happens, it will take time for courts to adopt this new framework if they ever do. But I wouldn't say they never will. It may just take time. So I think I'll stop there.
Daniel Francis 27:00
That's wonderful. And let me just invite others in on this sort of big picture question. And I'll throw in my sense of it, which is, there are plenty of things holding merger enforcement, and tech merger enforcement in particular back. But those things, almost without exception, cannot be solved by changing guidelines. There are things like lack of agency staffing, funding, largely hostile courts, problems with evidence and witnesses difficulty showing effects in vertical or certain kinds of tech cases. Those are real problems, the inaction of Congress. Those are real problems, not obvious to me that the guidelines are a place to find win those battles. But let's hear from others.
Doug Melamed 27:43
Let me just a couple of thoughts. First of all, there's a normative manner, I start with the notion of the antitrust laws about economic welfare, not about some of the other non economic objectives that that people who disagree with me think it shouldn't be about. And from that perspective, I don't see the 2010 guidelines as being an obstacle except in one respect, I'll get to in a minute, because I think they set forth a coherent analytical framework for analyzing mergers. And I think what we see in all the birds occations that we talked about, and one that we didn't talk about, except for when we didn't talk about, these aren't new theories. The agencies have actually been litigating very conventional theories. The one exception was AT and T TimeWarner. On the prior watch, which which articulated a theory of hiring based on Nash Bargaining model rather than foreclosure. Even in that case, and in all the cases that the current administration has brought the court except of the legal theory, from the wrong with the legal theories, articulated in the guidelines, the agencies have lost on the facts. I think the problem played us face and mergeable nonwords a law and anti trust law, is that the courts had been much too demanding for meticulous proof of matters that can't be proved, with that kind of precision. In the merger context that manifests itself, among other things, in the general rule of thumb that many courts have embraced it, others I think implicitly embraced, which is in order to prove a violation, section seven, you have to prove that the bad things are more likely than not. Now, a case where that's exquisitely inappropriate. This is so called nascent acquisition, or even a potential acquisition case. Facebook buying Instagram, there's no way you could prove that it's more likely than not that but for that Instagram would have evolved into the next big thing or being a huge, a huge threat to at least not based on the information available at the time of that merger. But if you move to an expected value approach, I think, or something like it, where you can say wait a minute, what are the upsides from a letting off Facebook tried to improve Instagram on the one hand versus the risk that it will use Instagram as a compliment rather than a substitute. On the other hand, we may conclude that looks like a bad deal. So my, what I'm trying to say is, if I were drafting the 23 guidelines, I'd say the real problem is, courts had been much too demanding on proof. The 2010 guidelines may contribute a little bit to that, because the emphasis on modeling and analytics and numbers may have led courts to believe that's a necessary element of proof. And I'd be writing guidelines that wouldn't be trying to reinvent the analytical framework, but rather would be emphasizing a richer portfolio of ways one can assess risk and prove risk, and so forth. And in that connection, by the way, if there's two or two thoughts about the guidelines, if their purpose is to is to keep Amazon from being Amazon that I think that Miss focus, I just want to agree with you here. On the site and case law, if the agencies really want to move this tanker when he changed the needle it you're saying wrong in principle we cite in case law, the problem is is unpersuasive as well as being analytically unsound. Just to cite a case, like Philadelphia National Bank. Yeah, it said 30%. But look at the facts in that case number two, and number three, and a heavily concentrated industry, and liquids happened to emerge the law post PNB, you can't just cite it say 30%. we're entitled to it to this. This presumption. You got to build a legal argument. If they if they do something like that. I'm all for it, because anything wrong with citing cases, but citing snippets is not the same the second case,
Daniel Francis 31:52
Spencer, I think you want to jump in?
Spencer Weber Waller 31:54
Yes. Just two brief comments. And then I want to talk a couple of things about the framework that we're talking about first is, I just want to take a second and thank Eleanor and Harry and Dan, for being among the kindest and most wonderful mentors that I and many people in this room have ever had, as a young lawyer and a practicing lawyer, and academic, all three of you were kind to me and helpful to me when there was literally nothing in it for you. And I appreciate it and learn so much from all of you and continue to this day, and hope that all of us who have benefited from this in this room and beyond, will will be able to continue that tradition with those who are more junior to us. So just thank you, from the bottom of my heart for all of you.
And Harry, thank you for putting up with all my nonsense as a co author. So thank you in particular for that. And then just one brief disclosure, I was at the FTC in 2022. As a senior adviser to the chair, I did not work on anything relating to the mergers and had no no involvement with this. So I'm giving you personal opinions that are just based on public knowledge. At one point, I was vaguely aware, because I was at a public pre op staff briefing on an earlier version that isn't the little I can recall very similar to what we have. But but in general. A couple of things. I think these guidelines are important. And I think they make some changes. They do certainly kind of Blender eyes some things that were in the 2010 guidelines and just move them around in different places. I have no objection to putting some stuff in an appendix that economic in nature and refer to throughout. I'm just agnostic to whether that's a good thing or bad thing that's drafting style. I think it's fine. I don't have a problem with citing cases. Most of the other guidelines and other areas, cite cases, other FTC things like policy statements and other stuff that the agency publishes cite cases. And and I guess I differ with Doug, on a couple things. But one is the Supreme Court has been clear that lower courts are under a duty to follow their law and not read the tea leaves until the Supreme Court decides to change the law. And I understand that that's, you know, you're you're kind of daring the Supreme Court to change the law when you do that approach. But I do think something I always respected judge posed before was that he often disagreed with the law and would simply state that and then follow the law and urged the Supreme Court did take serve on a case where he was bound by something and he would explain why that was suboptimal. State oil versus con is one of those things. But in any event, so I also think the 68 guidelines for which I was, you know, just a child were born from a place of humility and came at a time when the the Justice Department FTC would probably win any case that they could bring, you know, an argue with a straight face that certain and structural guidelines. And that administration, as I understand, it kind of just said, that being the case, here's what we care about the most. And this is one will act, or at least be more likely to act. And I think the 82 guidelines were different in two ways. One, I don't think they were born from humility, I think they were born from a type of arrogance about, we're just going to change the way we do things. And you know, you're going to go along. Now, as at that time, I was a junior lawyer at the Justice Department. And at least I appreciate the fact that here's a document I could read that tells us how we would do things. What I didn't appreciate is my experience in those some people who were there at DOJ with me may agree, is you, you do your best to investigate a case with the help of the the Econ people, and you at some point you bring up, you write a memo to the front office arguing you should or shouldn't bring a case. And the problem in my view was that the guidelines didn't represent what was actually going on, because the front office simply didn't bring any number of cases, that that solidly met what was said in the 82 guidelines. So I do think this set of guidelines that in draft format will be revised, in response to the conversations we've been having in the comments that have been filed. I think they represent what the agency cares the most about. Now, I, one of the great things that I think got overlooked in the 2010 guidelines, as they said, rather, clearly, this isn't a book, you don't just do A, B and C and D. And the biggest thing is theory of harm. But then the courts continue, and the parties often just did a, b, c and d, including to find a market and then do this and then do that. And the courts just kind of went along with that. And I do think this set of guidelines expresses some serious concerns that the staff and hires up have in the agencies that the courts have been getting it wrong. And one of that's about concentration. And I think the guidelines is mostly about power and not size. But I could see why some people might think differently. I think the staff and the results are also represent a concern about verticals. For example, in Microsoft, Activision, that I think is on appeal. Still, there was a amicus brief that I signed that Jennifer stereology wrote very short, basically just says that the court got it wrong as to the difference between total and partial foreclosure. And the court basically said, well, we don't find any evidence of of total foreclosure in the sense that the government was arguing, therefore, it's fine. And she persuasively wrote this nice tight eight to 10 page Amicus, that partial foreclosure is a different problem. The incentives are different. It's actually in the dominant firms interest to do this, and it can, you know, mess with your competitors as well.
Daniel Francis 37:58
So I think, Sally, I'm sorry, I mean, to cut across you go ahead. Yeah. And
Spencer Weber Waller 38:00
just two other things. I think also, there's important language about the situations in this area of the law and others where the agencies believe there's direct proof of harm, and they're still stuck in a market definition framework. And this, this set of guidelines provides alternate paths to making the necessary showing. And then finally, I think there is a deliberate choice. I have no nothing to cite for this or proof of it. But I think there's a deliberate choice to dissenter, the hypothetical monopolist test, and provide a broader palette of types of broader sources of evidence. And I think that's healthy. And I think you'll even like in something that's not a merger case, same trend in the Google case where you're seeing different kinds of experts, in addition to the type of gifted Iowa economist, you know, that we're familiar with, in these cases. And so you know, you saw behavioral psychologists who talked about why defaults matter. You see some business strategists, you'll see some marketing people. And I think that's all all to the good. And so I'll just stop there in terms of the the framing.
Daniel Francis 39:14
So you touch on something that I think is a perfect pivot point. To get a little bit more specific. You mentioned Microsoft, Activision, like so many of the other headline tech merges, that's a platform case. And for the first time in the 2023 draft, we have guideline 10, which puts a bunch of stuff about platforms in some of its descriptive, theoretical, some of it outlines possible theories of harm, and buried in there not really emphasized but pretty clear is this concept of a conflict of interest as a reason why a platform merger might be unlawful. So Spencer, let's go back to you briefly. Is this the tech update that we need in the context of the kind of thing you're talking about is this introduction of platforms the right path?
Spencer Weber Waller 40:02
Oh, thank you. I, I think it's gonna be I think it's gonna be helpful. I think it's really interesting. I, I read this, I don't know if Howard or others agree, but I kind of read this as a turning, turning two sided platforms as misstated in AmEx, from a defense into a weapon. And I thought that was really interesting. And we can talk about that more. But they again have given several paths to showing what the harm is, in terms of to the extent there's provable evidence about how platforms compete with each other, and a merger harms that or how the platform can take advantage of its suppliers or customers. And then all the sort of the technical stuff in there, what I kind of like about it is, and maybe this about the guidelines as a whole, I think you have to read them in, in conjunction with all the changes that the agencies are doing and saying about Section Two law, and Section seven, and section five of the FTC act. And we'll get to this when we talk about serial acquisitions. But if you read the complaint, in the anesthesia case, that it's not, you know, it's not a high tech platform, but it's just a roll up of anesthesia practices. You see a blending of all these theories, there's our count set, count one, count two, standalone, Section Five and sub section seven all together. And I think you got to sort of look at that in context, even though that's not the literal topic of the guidelines. Finally, what I like about the platform section, is it nudges the United States into closer harmony or relationship, it doesn't get you all the way there with the rest of the world? It you without saying it? It sounds to me like it has echoes of abuse of dominance. When you talk about entrenchment. That's another way of you can say in conventional us terms, monopoly maintenance. But also you can put it in conjunction with the rest of the world that is by and large, rooted in their unilateral conduct around an abuse of dominance concept. So I think it's several steps in the right direction,
Daniel Francis 42:21
how it platforms, is this the way?
Howard Shelanski 42:24
It's part of the way? So, first two disclosures, I have provided legal advice for and against certain large platforms. So I've taken money, exact Well, it depends on the platform, Richard Yeah. So so so so so I have received remuneration for making arguments related to platforms. The second disclosure, I was one of the co authors of the 2010 guidelines. And so that will have some, some influence on how I think about these 2023 draft guidelines. Look, I think the platforms needed to be tackled, and I commend the agencies for doing that. I think one of the most terrible cases in in antitrust jurisprudence in recent years was the Supreme Court's Ohio against Amex decision, I think it's wrongheaded on multiple fronts, I think it eliminates the concept that as a plaintiff, you should only have to show harm in one market, it confuses what should be a defense burden on rebuttal with a plaintiff's burden on the primary case, I think the economics are terrible. I think the language and the definition of what is a platform with partial cross network effects is poorly thought out. So I'm glad that they're kind of trying to pare back their interpretation of how broadly that case sweeps and I'm glad they're taking aim at platforms. I do think that the conflict of interest notion that is introduced, one has to be very careful here. I do think it's an appropriate consideration when it comes to mergers. I think it's a much harder consideration when it comes to section two. There's even a citation to justice Marshalls concurrence in the Falstaff case, in these guidelines, you know, which says, Well, there's a preference for organic entry and development. Well, what do we do when you're a platform? And you're organically entering into competition against those who are interoperating with or using your platform? You know, should that be a problem? Is that a monopolization issue? I think that's a very separate question from what they're addressing here. If you're making acquisition, that introduce conflicts of interests that could lead to changes in how you deal with those already operating on your platform and from whom you're making a lot of money. I think that's a reasonable consideration in a merger. The question I have is how that differs from a vertical vertical foreclose Was your analysis. And I don't see that it really does. So I would have preferred sort of this being an example of a kind of foreclosure or way to think about foreclosure in the in in the, you know, in the vertical, or you can call it conglomerate, or you can call it non horizontal, sort of sets of considerations. And the reason that I think that that matters, is when you start to introduce new new categories of harm, you know, they will get talked about and challenged as well, there's no identified conflict of interest, you know, antitrust harm. And I think that if you put it into a more conventionally understood harm, that would have been somewhat more effective. But look, that's drafting, it's not really substance, I don't begrudge the agency for pointing out that it's an issue that's going to arise more likely with, with the platforms with big platforms and with others, I have some concerns with some of the other presumptions that are sort of embedded in the agency's thinking about platforms as it comes out in the platform guideline, but also some of the others. But you know, but I think it's a start, and also just want to emphasize something that Daniel said, you know, the agencies are acting in a in a bit of a vacuum here, you know, Congress has not acted on platforms, we are ceding the intellectual lead on how to address platforms to other jurisdictions. If you look at what is happening in Australia, and the United Kingdom, in Europe, one of the things I actually liked about the platform provision was kind of the US agencies saying, you know, we're not totally asleep at the switch here, they might be, you know, over on, you know, up the hill, in terms of legislation. But we're actually thinking in some very similar ways to the way other jurisdictions are thinking about platforms. And you as a platform should not think that just because you're being, you know, reviewed here, and not by the DMU, in the UK, or under the DMA, in the European Commission, or under, you know, the legislation in Australia, that, that you're going to be confronting a different or more sympathetic, you know, audience, we share these concerns, and we will use what tools we have to vindicate them. So I thought that was a useful part of that provision.
Daniel Francis 47:23
So let's, did you want to jump in on platforms?
Harry First 47:26
Do I have a platform to do that? Yes. So yeah, a couple reactions to that, first of all, in terms of the American Express case, it in my view, it is the Arnold Schwinn case of the was decided in 2018. So Schwinn had a 10 year run until Sylvania. So come back here in 2028, to for the overruling, nobody likes this case, except for Richard. And Richard is not a nobody. But, you know, most people, most people, nobody likes this case. Anyway. So that's sort of the first point. But the second point is the point you're making. However, the this the the analytical structure of that guideline, in terms of thinking about platforms as markets, it really brings us within the way, I think almost every I can't say every but almost every other jurisdiction is thinking about it, and seems quite right. So you know, you look at the markets on either side of this two sided platform, but don't call it a two sided market. And no, I don't think any other jurisdiction does that. And Sabre fair logic is sort of the, the test case for that because the UK rejected that merger and didn't buy the two sided platform. And that's why the merger fell apart, even though the district court judge accepted the that argument in the United States labor fair logics. So the on the I think we should be aware, and one of the things that's not that the agencies don't seem to pay all that much attention to is what's going on in other jurisdictions, and specifically, in particular cases, and I'll come back to Microsoft, Activision, were in the UK and in Europe, they rejected a major part of foreclosure argument on factual grounds that the FTC tried to rely on and pushed in a district court, instead of focusing on the part of the case that may turn out to be the most important which is, where's it going to? How's it gonna be distributed in the cloud, and that developing nascent market so those two markets those two jurisdictions found there were anti competitive effects a lessening of competition and instead, the FTC seemed almost to ignore that and litigation and focus, I think on sort of the loser market. So someone said, it's not the law, it's a province of facts. Well, okay, the facts are there. And it's, it's more than just the facts. It's trying to think ahead and dynamically about what's happening in tech markets, which have not developed yet, which are developing. So you don't have the same data. You don't, it's harder to make predictions and even harder, that dynamic analysis that Aviv said, we're going to definitely have, in the next iteration of the guidelines, I am waiting for it, because that's what's lacking in this iteration. And that's where it's important for the tech mergers, the problem going forward. I mean, we have a relative consensus on horizontal mergers, we can discuss, you know, different nuances, but not on these mergers, a tech mergers that involve future markets, where you have these firms that are dominant in one and moving into another like meta within or like the Microsoft Activision, we don't have as much of a consensus. And for that, I think these guidelines are trying to move it forward. But maybe not paying enough attention to what's happening in the rest of the world. You know, actually, the CMA revised their guidelines in 2021. I hope somebody looked at them, they're pretty good. And they're very current, much more current than the European merger guidelines. So there's a lot to be learned from outside, including how they're approaching the very cases that we're talking about in the tech world.
Daniel Francis 51:54
So I want to pick up Harry on something that you mentioned, which is the problem of dominance and dominance, not just prospectively, but also firms that have achieved that dominance already. Obviously, when people are talking about tech antitrust, they often have some tech monopolists in mind. And this set of guidelines clearly has them in mind. Also, there are a number of places where the 2023 draft innovates in ways that seem to be looking towards firms that are in a dominant position. So one of those is the serial acquisitions, which we hear a lot about roll off acquisitions, stereo acquisitions by big tech. Another is the famous guidelines seven dealing with entrenchment and extension of a dominant position, which kicks in presumptively at a 30% market share. But also rhymes with the 30% share in guideline one. So I'm going to come to dog first dog, we've been talking for a long time about how to handle acquisitions by tech monopolists does 2023 Give us the tools we need?
Doug Melamed 53:04
Well, it probably does. But there's some ambiguities that I think are worth noting. So there. Let me talk about three components of the tools 2023 talks about what are our serial acquisitions. And that means a series of acquisitions by the defendant or the merging party acquiring party usually, in the case before the agency, it says we'll consider we're going to consider serial acquisitions. Well, let's take a case like the roll up case. And that was a Texas anesthesiologist at the FTC broad. That was a kind of the scheme to corner the market. As I understand the allegations in the complaint, I think it's a continuing course of conduct, of course, you want to aggregate all of that. I think it probably fits more naturally, frankly, under Section two, but even if they share some gets into section two, I think you can bring it under Section seven ever kind of. But I think it was you have to be something that you can characterize as continuing course of conduct. And I think that's what the guidelines have in mind when they say the agencies will consider individual acquisitions. In light of the cumulative effect of related patterns or business strategies, that's guideline, nine. But it's not entirely clear that they're tethered to a continuing course of conduct. Because they talk about, about a pattern or strategy of growth through acquisition so they could encompass, for example, Amazon, you know, buying whole foods and and buying somebody else in a different field and so on and so forth. And eventually saying, Well, gee, they've done something bad. They've gotten too big. And the problem here, I think, is something like this. We're dealing with law. We got to identify what's the illegal act that the defendant engaged in. So if we have a bunch of independent X independent mergers, not part of the continuing course of conduct is an opportunity came here and opportunity came there. I call them employee hire, talent hire acquisition there. And then you come to the fifth one, you say, Oh, they're hiring too. They bought too many. They had too big a state to sort of be illegal. That gets awfully close to the discredited. Sherman Act theory of say, well, we can accumulate, aggregate a lot of benign conduct in and then say, Well, overall, it's really hurting competition, that's for sure. Neck violation. I don't think that's how you make conduct illegal ISIL. I think it's truly independent. You can look at the prior acquisitions because they changed the facts on the ground, they may make the merger more concentrated, they may create a vertical threat because of changes in the Related market, they might shed light and give you evidence about the likelihood that efficiencies are really going to be achieved or what's the strategy of the company in the dynamic sense, but I don't think you can just aggregate the effect on the market and turn that in to a violation. Okay. Another aspect of the guidelines is on so called entrenchment and extension of a monopoly position. entrenchment is clearly an important cognizable harm. That's what the Microsoft case, of course, was all about raising entry barriers to protect an incumbent monopoly. And no problem considering entrenchment, or cognizable. In some ways, maybe maybe one of the most important harms that any trust law should be concerned with. But any trust law has two elements, it has the harm. entrenchment, a monopoly is surely one, and it has the bad conduct. So you got to ask where is the bad conduct. And here I'm very concerned about the language and in guideline seven, because it says that they will go after mergers, that entrenched through any mechanism. And they listed their examples creating network effects, your scale economies requiring scale their customers less deprive your rivals, just a couple of days ago, the EC block the merger between an online hotel booking service, and an online airline booking service. And the theory was, well, they're going to have network effects. And it's going to make it harder for their rivals in those two segments to compete. I'm an old fashioned guy, I think that's a way of saying oh, they're going to have efficiencies and be able to bring more value to their trading partners. And I don't see that as an antitrust violation. So I would hope that, that the language is clarified on entrenchment in terms of looking for something other than a so called efficiencies, offense, and then the third one in the in the same guideline, extending into an adjacent area, I think this is somebody that guidelines seem to care a lot about the conflict of interest language, in guideline eight, when it was talked about earlier, I think is part of this story. And here, I have a real problem. If you have, let's take Amazon. And let's imagine losing to an adjacent product Mark, if it's using its market power, in online sales ready to call it stores to leverage its dis leveraging its power by tying by predatory pricing or something like that, or predatory refusal to deal by which I mean a profit sacrificing disadvantaged rivals, I get that that ought to be anti competitive. What if it's simply taking advantage of scope and scale economies, when it goes into an adjacent market using illegally acquired properly used in terms of the obligations to the supplier data to provide a better product in the adjacent market? Again, that's welfare enhancing conduct. And I don't think there will be that all to be competitive. So I would hope that the be illegal. So I would hope that the guidelines would be clear that they're not trying to go after efficient welfare creating conduct, just because it's going to make it harder for rivals to compete against the big guy.
Daniel Francis 59:03
So I've also sort of struggled to figure out what guidelines seven on entrenchment and extension is trying to capture that isn't a horizontal harm, a vertical harm or an efficiency. And I sometimes have heard, you know, gosh, it's like the Microsoft example where we're going to knock out middleware. To my mind that story is a story of complement foreclosure that fits very traditionally within vertical analysis on the guideline five, and if we mean something more narrowed by vertical than we shouldn't, we should be real clear. That includes complements. Let's hear from Spencer and then we'll come back to Harry.
Spencer Weber Waller 59:40
Sure, so I'm guessing but I certainly don't know that. The stuff about serial acquisitions is designed to clarify, you know, the reach of the NFL As long as to a series of, of small, multiple acquisitions over time, that could either be a roll up or sort of a market extension and, and I get Doug's concerns, but I think it's not as important if if the bad conduct is the acquisition of, or denying somebody, scale by by acquisition as opposed to predation. And so, you know, in Microsoft, they were, they did what they did to prevent various other nascent competitors from achieving scale. And the theory of the FTC case against meta for WhatsApp and Instagram is that they did it by acquisition, in one case, paying billions of dollars for a company that had no consistent revenue stream at the time. To me, they're there, they're roughly the same as long as you you show what the likely harm isn't you, you have the benefit of the Clayton X incipient see standard, which we haven't talked a lot about. So I'm okay with that. I suspect in the real world, it'd be used more for roll ups. It's every venture capitalist dream to sort of fund an endless series of acquisitions, so that, you know, veterinary clinics, and anesthesiology. And at one time, they were trying to roll up direct mail advertising firms that didn't go so well. So I mean, you know, it never got challenged. So they're, they're just successful roll ups that harm competition, they're probably successful roll ups that don't, because the nature of the market, and then there's some more they try and just fail. So but, you know, the FTC brought a different Vons case than the ones that that is famous from the Supreme Court. And this is in the late 80s, and early 90s, where Fonz was doing a series of rolling acquisitions of grocery stores and small chains throughout Southern California, none of which reached anywhere near Hart, Scott level stuff, and they sell them off here and there. And the idea is, over time, they extended their dominance to a bunch of areas where they weren't dominant before and made it harder for any other grocery store chains. I have a serious scale, and that's a problem was resolved by consent decree. I don't recall it being cited in this. I do believe it's cited in the section five policy statement. And so I think at the end of the day, you'll see, you know, 70 80% of the cases being more like that, using this part of the guideline. But again, I can't remember the the technical term for this acquisitions that are mostly employment and talent acquisition. It's a very technical term, we meet up but again, if you can make the case on the ground, that a tech platform is just doing a rolling series of relatively small acquisitions, to prevent anybody from reaching scale and to knock out the most likely ones who would grow over time, or at least have the potential to, I'm fine with that. I think it's a good idea. And again, I think the laws there and probably the facts are going to be hard. If that's the sole theory you go on. And again, does this sound more like a guideline five and six? Probably, I appreciate Aviv explanation on the second panel. I think I for the first time understand a little bit of the difference, as written between five and six, but I'm still pretty sure they overlap heavily.
Daniel Francis 1:03:29
So I do want to make sure we get to the articles. I saw a brief interjections I think from Harry and then Howard and Eleanor is are you wanting to come in on this? Oh, wait. Tober?
Speaker 4 1:03:39
I I wanted to make one very short comment about Cyril acquisitions, I think, you know, section seven does say substantial lessening of competition or tend to create a monopoly and from an enforcement perspective, does tend to create a monopoly is enticing. But it's very hard to know when, you know, when when do we see an acquisition or a series of acquisitions that tends to create a monopoly? So I think in a way the serial acquisitions guideline is attempting to get to that problem, like what Spencer said about scale, you know, depriving rivals of scale. That's another aspect of it that I think is particularly important in tech mergers. So yeah, so I'm not answering any questions. I'm just saying that I think this is one of the one of the questions that Cyril acquisitions tries to address
Daniel Francis 1:04:37
critical from TF Ushaw. Harry, did you want to jump in briefly,
Harry First 1:04:41
just briefly, I just want to say Eleanor's notion of tend to create a monopoly I've always wondered with that part of Section seven min. I've never seen a case that's litigated. That part of Section seven, and I've never gone back. I actually tried to do a little looking at the legislative history. I'm not sure What it means exactly, but it's there. And there are hints of it in the guidelines as well. But I do want to answer your question, what's guideline seven? What's a case for guideline seven? To me, I read that I said, That's meta within. So it's a trencher. Extend. So meta is making this presence in the metaverse through all kinds of acquisitions. This is the way in sort of an answer to Doug's. You know, are we worried about this? I think this is what the FTC is worried about the way platforms grow. And there are lots of studies including the FTC zone study about lots of small acquisitions in sort of related markets that are not adequate hires. And, and, and that's how they grow. And we don't want them to grow, and entrenchment. This is for Peter, the Procter and Gamble case 1967. Procter and Gamble acquires Clorox, and part of the court's opinion is about entrenching Clorox in this broader market by acquiring the leading firm, the leading producer of household bleach. And that's the theory, I would have liked to have seen the commission pursue and meta within, for which there is one paragraph in the complaint, and barely a mention. But if you're going to talk about case law, that's my kind of case law. And I would like to see them use it. And I think entrenchment in that guideline seven, fits nicely. 30 seconds
Daniel Francis 1:06:31
from dog and then we'll go to verticals.
Doug Melamed 1:06:33
Yeah, it occurs to me, I think there's a really interesting idea that if a company A buys Company B, and as a result, no one else in the market could get to efficient scale. And therefore the merging party is going to have market power. That strikes me as a cognizable. harm to competition. I think that's independent of whether it's a series of acquisitions or one acquisition, whether you can call it entrenchment, or whether you can call it expansion, it's simply a kind of harm that could be realized. I don't, I think they should bring that case, I don't think you have to get hung up on series of acquisition and stuff like that.
Daniel Francis 1:07:08
So, verticals, you know, came up was a critical frontier in the last discussion, it's kind of been lurking in the background of a lot of what we've been talking about so far. Let's take it head on. A lot of the tech acquisitions are vertical, whether it's a platform and a compliment, or games console and games console maker and a game studio. It's pretty clear that guidelines six in particular is an effort to hit the reset button or to turn on some additional machine least on vertical enforcement. Kind of has shades of brown shoe looks like an argument the FTC tried in Microsoft Activision, is guideline six gonna do the trick? Is it gonna perform some function that we need? Let's get an enforcement perspective. Eleanor.
Speaker 4 1:07:55
So, as I think Jennifer Malachi said earlier, I'm not a scholar, you're not an academic, but I do hold the title of an enforcement official. And I want to just talk about vertical, vertical mergers and enforcement here. So the standard that is applicable, generally, that we've been following, and is that in my view, is incorporated in guideline five is ability and incentive. From an enforcement perspective, that is a very, very big hill to climb. So this is what happens on the enforcement side, we have economists, highly respected, very smart people who can bring to bear their theories about whether the, the the emerging firm will have the ability and incentive to foreclose arrival at some level of the distribution chain. And if their theories are grounded in sound economic thinking and reliable methodologies, they'll get past a Daubert motion. Right? That's the first build test. And sometimes we have third parties, but increasingly, it's very hard to get third parties to testify in a case where there is a an elephant, right? They're worried about continuing to do business with their business partner. So yes, there are third parties, sometimes their testimony can be a little bit equivocal. It can be difficult. So you get to trial. And we have our economists on one side and the merging firms bring in the people in the trenches, the executives who run the company, the people who will stand up there and say very often credibly, no, my firm never would never could Never will engage in this kind of conduct. It's our, it's our against art, it's against our interest, it's a, it will damage our reputation in the marketplace, it will deprive us of the profits, if we, for example, don't allow the company we're acquiring to provide inputs to our rival. They'll lose money. But well, maybe, you know, the real thing that the executives thinking about is the overall profitability of the firm. So if you make money on this end, but you lose money on this engine, you make more money on this end, it's going to be worth it. So a judge sitting there and I have I have sympathy for the judges here. They're sitting there with factual testimony on the one hand and theory on the other side, the economists offer theory and opinions. The guys in the field offer expertise, daily experience with the firm in the marketplace. Who are they going to believe? I think the judge almost has to choose the factual testimony over the theory. As long as the witnesses are credible, and I'm thinking of the UHD, change trial, I was sitting there, okay, maybe not a tech case, in my view, it's very close to Ted pace. Uhg is the one of the big, biggest health insurers in the country, the largest owner of providers, health providers in the country. And it's acquiring change, which is basically the pipes through which claims are made and information flows. I was sitting there in the courtroom the day that our economist DOJ in New York was a plaintiff and Minnesota was plaintiff got up and testified. And he did a good job. And then Andrew Woody, the CEO of Uhg, came in a very, very credible gentleman with a lot of expertise and a long standing reputation with this system. And I wasn't surprised at all, when we read the decision and found out which way the court went we lost. So I think we needed a reset, as Daniel put it, we need something along the lines of guidelines sips, which attempts to I think, do kind of a shortcut. Through the presumption. I don't know whether 50% is the right number. I don't know where the 30% plus plus factors is the right number. But I do think that we need a standard beyond incentive and ability. We know got reaction, and I think Aviv, Neville said pretty much this, that not all vertical mergers are okay. There's no rule of per se, legality for vertical mergers. There are some that we know are not okay. And we need a way to prove that in in an enforcement context.
Daniel Francis 1:13:19
That's a wonderful enforcement perspective, Doug, what do you make of it?
Doug Melamed 1:13:24
Well, I mean, overall, I like what the guidelines are trying to do in the vertical area. But I want to talk a little bit about about what Elon was talking about. I think in principle, I assume we would probably all agree that the ability and incentive test is the right test, right? It doesn't, it's not going to happen, and we don't care about it. So what would cause it to happen? Well, you got to have the ability to make it happen. You got to have an incentive to make it happen. The problems with it is like not smelly garbage from others as well, I guess are twofold. One, it's hard to prove incentive because you have your economist with his model up there. And then you have the CEO coming in there. And after also a lot of CEOs how do they get to their job because they're, they're good, impressive salespeople, and they're kind of charismatic and then used to talking to the, to the street and of course, you get good at bullshitting when you're doing that. So they're gonna they're gonna win win the bat. I think that's a legitimate problem. I think it's a subset of the broader problem I referred to earlier about the difficulty anti trust plaintiffs have in general, because courts have been so so demanding on their proof that it's not just on motive and incentive, it's also on costs and in predatory pricing. You can go on and on. And I think that's what the the Reformers ought to be focusing on. Not removing from the analytical framework, an element that so plainly is necessary to make a coherent theory, namely the element of incentive. Now the other argument against I'm having a prove incentive and ability and I guess it would it would come down to just ability being sufficient is the one that a be referred to earlier that day, which is what you really want somebody to have a gun to they had to have the trading partner, you know, maybe hit two weeks from the Senate to shoot it. But it's kind of changes. Well, that was kind of the theory that the government used in, in ATT, Time Warner. Remember, they didn't say they were going to cut off competitors. They said they're going to use it in the bargaining to disadvantage rivals because of the risk of the of credible threats or whatever it may be. And I guess what I'm trying to say is develop analytical tools to analyze directly if a gun to the head risk, that doesn't require proving more likely than not that they're going to going to foreclose. In Activision, for example, as I understand it, and I've been told this, I haven't read the record. The FTC did not put in evidence on partial foreclosure, their expert, addressed only at the issue of foreclosure. Well, I think partial foreclosures let's theory the agencies ought to work on and try to develop, I guess what I'm trying to say, oh, it'd be an accessible be a long winded way is instead of inability should both be required, but the agencies have to think about as more creative ways to address the problem persuading a court that they've done enough. And then finally, the statement in the guidelines, it says efficiencies are not going to be cognizable. And it's not clear to me whether they mean vertical emerges or trend to vertical integration, if there's little ambiguous because of the placement of the word trend in that phrase, but either way, I just don't think you're gonna throw efficiencies out of out of consideration, that has to be a part of the analysis, at least by the agency, if not by the court.
Daniel Francis 1:16:51
So I have a favorite question for events like this, which is some version of if I gave you a magic wand, what would you do with it? So if you had a totally free hand on the 2023 draft, but maybe only the opportunity to do one or two things with that freehand? What would you focus on changing and why, Howard?
Howard Shelanski 1:17:16
Boy, so I think that the where to start, I only get one wish?
Daniel Francis 1:17:25
A couple if you need a
Howard Shelanski 1:17:27
look, I think that there, I am very concerned that these guidelines mark, possibly the beginning of the end of the guideline guidelines experience, I think there are too many fronts on which these guidelines step significantly ahead of or to the side of consensus in a way that could lead to a ping pong game where another administration comes in and says, your reading of those cases you cited is those those cases aren't examples. You're reading the doctrine in a particular way, we have a different reading, your approach to vertical mergers is outside of any kind of established consensus, these are new things, we're going to pull these back, you can imagine sort of getting into that kind of ping pong, at which point, the guidelines really become mission statements of the administration that's in power, not the folks that are there. So I am actually quite concerned about these guidelines that I think risk the flaw of being a sort of transparent articulation of the many things agencies are thinking about, which may or may not be good things to think about. I think people will disagree. But they're not going to form enduring guidance. And I don't think they're going to inform an enduring path, the policy change on a number of issues. And I think, you know, I think the vertical mergers is sort of, you know, guidelines, the two of them, I don't really know why there are two. And unlike Spencer, I haven't really figured out what the difference between them is or why they're two separate ones. But But I will, I will note that there, I think is a very good example of something that highlights this concern. Vertical mergers, I will, I will agree fully have not been, I think well thought about for decades. I think there's terrible ignoring of a lot of important economics, the whole notion that a vertical merger might be a good way to better govern a transaction has been sort of not thought about, and the way that could be used as a way to screen out bad vertical mergers. The very good economic research that shows you can create a lot of elimination of double marginalization contractually and don't need a merger to do it also was ignored for decades. So I think vertical mergers have not been well thought about. But when you have something in the guideline that says efficiencies don't count Well, governance efficiency. So the main reason vertical mergers happen. There been at least two Nobel Prizes given for that insight by Ronald Coase and and Oliver Williamson and literally hundreds of empirical papers. I know because I wrote one of the three Because of that empirical literature, and when I stopped, there were like 250 of those articles, and that was in 1995. So that's so efficiencies are a key part. So when you start to have something that calls that into question, I think that's a big, that's a big problem. And just Second of all, a structural presumption where there hasn't really been one before. And you point to a trend towards vertical integration is counting against the further vertical integration, that trend could be highlighting the fact that this is an industry in which there is integration, because that's the beneficial way that the technology or the industry is going. So I think that's sort of emblematic of we're a sort of not well thought out set or consensus set of principles is being thrown into a guideline that's easy to react to and push back. And I don't think it's going to advance the ball. But my one magic wand wish, I think, I think the structural presumption has been overstated in these guidelines to weigh that significantly, outside of where the law has gone, and as well outside of economic consensus. So that might be the one thing I can change.
Daniel Francis 1:21:07
Harry, you got the magic wand, the benefit of being the honoree, what would you do with it? Well,
Harry First 1:21:13
I would I would waive it. I would waive it. Okay, so I have three suggestions. The first suggestion is based on our first panel, where at the end of the panel, I said, So what do what do presumptions mean? And this because I think, I think nobody on the first panel could quite agree. So I think where these guidelines can really help, and people don't like to think that these are cookbooks for judges, but what do judges like? I think everybody likes cookbooks, a recipe and their best when they have three parts, you know, like the like, entry, there are three things for engine. Okay, I can I can do those. So, I would say lay out the analytical structure for what presumptions mean, the problem isn't the PNV presumption. It's what do you do with it? Once it's raised? How do you meet it? How does that affect the analysis? Okay, so that's one second. EDM sounds like a disease. Elimination of double marginalization should be clearly stated in the officience as an efficiency defense. And it should, and it should indicate how it should be raised. And it should be viewed as skeptically as others. I never understood and AT and T, Time Warner why that point was given away by the government's economist whose name we all know, but won't otter because we liked him too much. But that was a key concession it was this is, and this is a classic reason for shifting the burden. They know it. Okay, so that's the second, the third involves settlements. This is the major challenge of the three cases that I mentioned to were basically done in or ended by settlement. So Microsoft, Activision is about to settle out in the UK, it's settled out. In the EU, I'm not sure what's going to happen in the US. And Illumina GRAYL. Same thing, when it went to court, the judge bought the settlement on the same way that judges buy the efficiencies, Hey, that sounds darn good. It's a great settlement. So the downstream settlement, particularly in these cases, is really important. And again, it's really hard to know all of you know, is very complicated area, but a structure for how they should be taken, taken account of, in the context of litigation, so as to avoid this, you know, constant changing, there's a good part about the changing in Microsoft, Activision, the deal got better and better. But that's another aspect. So I would say the third thing is dealing with that not waiting for a whole remedies approach. But dealing with that the same way you deal with an efficiency burdens on you. It's got to, you know, however, you this is hard, by setting out a way to for a judge to think about it. So the judge isn't just in that position that you point out, where, you know, the party is coming in and say, Oh, we're so wonderful judge. We're promised to have this completely open. We'll offer it to everyone Don't worry about us. Yeah. Right. So.
Daniel Francis 1:24:35
All right. Well, we have lovingly preserved two minutes for question from the audience. Tim.
Speaker 8 1:24:47
I feel like this is becoming a traditional Oh, this one works. I actually have a question for Howard. Since you have the benefit of experience at 2010. I was wondering if the process by which these guidelines is being adopted as any different than the 2010? And whether procedurally or process wise have anything to say about the differences between 2010 and 20? And whether this is a broader process? It's
Howard Shelanski 1:25:08
a great question. And I think both in their production and in their public facing nature, the difference here is really great. There were six people involved with the drafting, two of whom are in this room, I'll call out Phil Weiser. There were six people who are involved with the drafting of those guidelines. You know, it's easy to name them all. And we sat in the room together time after time again, and we came up with a set of guidelines. And then there was a lot of internal socialization. But the internal socialization within within the agencies was very different. There were enormous numbers of stakeholders internally, in each agency through which these guidelines, the current ones, so 2023, were worked on, were vetted, were contributed to, that did not happen. We controlled the drafting process, we took them around to leadership at the DOJ, to Christine and to share us on to a couple of other people and to the commissioners. And once we'd sold them internally, we pumped them up on a website and said, let us know if you have any thoughts about them over the next 30 days, we tweaked a few things here and there, after getting you know, insults hurled at us, you know, by a few people saying that it was too radical, to have upward pricing pressure, so that we could have a more robust approach to unilateral effects. And you didn't have to do a lot of market definition and unilateral effects case, which by the way, was one of our biggest goals to try to make unilateral effects more easily proven. I like to think we accomplish that when you look at the h&r block case and some others and we we push back on Oracle PeopleSoft, which we which we were deliberately targeting, but it was a much more focused, I think, sort of smaller process. I think that has pros and cons, it makes the guidelines look a bit more technocratic, and sort of, you know, provide some more of the kind of guidance for how you will do this, what will but they're narrower. This is more radical change that we're seeing by a, you know, not an increment minded agency, by the way, I'm an incrementalist, a moderate, I'm all those things that get you defenestrated These days, but by people with big vision, Tim Lena, Jonathan, who are like, let's, let's knock some walls down, let's not repair, you know, the bricks. So you're gonna have a very different process when you try to do that to get a much broader stakeholder guide. And we did not hold guidelines workshops around the country, when you're trying to build up a political constituency. That's what you do.
Daniel Francis 1:27:43
terbang on 345? I'm gonna take one more question, sir.
Unknown Speaker 1:27:47
Yeah. Okay, so. So, so, so the guidelines have been 21 years. So I'll repeat reaping the benefits of that in the situation. So can you talk more about the guidelines? The guidelines? that guides my question.
Daniel Francis 1:28:20
So, you know, we've had a merger enforcement system since 1914, it's 109 years old, how would we characterize the place of the guidelines in that framework? Harry, do you want to take that? Yeah, we've had us, right, we've had guidelines in 68. How would you characterize the guidelines central peripheral instrumental.
Harry First 1:28:39
So I used to think of the guidelines as the crown jewel document of us enforcement agencies that were looked at around the world. But things have changed. Maybe not necessarily because of this, but because of the rise of other strong enforcement agencies. And I don't think we do have or should have that sort of hegemony. I think, in fact, one change that I would like to see more of is, you know, more looking at how other jurisdictions are solving, basically the same major problems. And that's not something that we had to think about in 1968 1982. And even not so much in 2010. So I think that's been a shift that we haven't quite acknowledged.
Doug Melamed 1:29:34
One thought, sure, the guidelines as I understand them, starting in 68, and continuing to through 2010. purported to be and we're descriptions of how the agencies think about mergers. They were not explicitly a document intended to persuade the courts how they should think about mergers, although they had that effect, because they had a cold Parents have the credibility within the agencies and the trust community. And I think that's a desirable state to be in, if there's enough of a consensus in the trust community that somebody's document might have that kind of effect. But when you get into the area of telling the courts how they should decide, it's a much nicer undertaking.