A Dawning Era for Vertical Mergers? The New Vertical Merger Guidelines, the FTC’s Challenge to Illumina/Grail, and the Future of Enforcement
The antitrust agencies’ approach to vertical mergers has been the subject of significant debate — with potential changes still on the horizon. Last summer, the Department of Justice and the Federal Trade Commission issued long-awaited Vertical Merger Guidelines, the first update since 1984. The FTC’s challenge to the Illumina/Grail merger — which sits at the intersection of healthcare and developing technology issues — is currently scheduled to begin an administrative trial next month. This would mark the first vertical merger litigation under the new Vertical Merger Guidelines, and one of the first since the DOJ’s loss in AT&T/Time Warner. Our panel of experts discusses the recent developments in the vertical merger space, the theories at issue in the Illumina/Grail case, and implications for enforcement activity over the coming months and years.
Although this transcript is largely accurate, in some cases it could be incomplete or inaccurate due to inaudible passages or transcription errors.
[Music and Narration]
Nate Kaczmarek: Good afternoon. Welcome to this Regulatory Transparency Project webinar. This afternoon’s topic is vertical mergers and whether or not we are experiencing a dawning of a new era. My name is Nate Kaczmarek. I’m Vice President and Director of the Regulatory Transparency Project. As always, please note that all expressions of opinion are those of our guests.
For our moderator today, we are once again very happy to welcome Elyse Dorsey. Elyse is an adjunct professor at the Antonin Scalia Law School at George Mason University. She previously served as counsel to the Assistant Attorney General in the antitrust division of the Department of Justice and as an attorney advisor to the FTC Commissioner Noah Phillips. Before joining the FTC, Elyse practiced at an international law firm where her work focused upon cutting edge competition, privacy, and consumer protection issues in courts and jurisdictions across the globe, including the United States Supreme Court.
If you’d like to learn more about all of our guests today, you can visit our website, regproject.org—that’s R-E-G project.org—where we have all of their complete bios. In a moment, I’ll turn it over to Elyse. Once our panel has discussed our topic, we’ll go to audience Q&A, so think of the questions you’d like to ask them. Audience questions can be submitted via Zoom using the raise hand function. And we will call on you directly. With that, Elyse, Taylor, Bruce, Tom, Michael, Steve, what a panel. Thank you all very much for being with us today. Elyse, the floor is yours.
Elyse Dorsey: Okay. Well, thank you so much, Nate, for that introduction and thank you and welcome to everyone joining us today — all of you out in the audience and, of course, our excellent panel whom I’m very excited to have here today and very thankful that we could get everyone here to join us for what I’m sure is going to be a really interesting discussion. Right now is a really interesting time, I think, for a lot of antitrust issues. I think we’ve all been hearing quite a bit about antitrust enforcement and some of the biggest tech companies.
But one area that’s been at least comparatively flying under the radar lately is vertical merger enforcement. Questions of how antitrust law should approach vertical mergers and other vertical arrangements for that matter have been the subject of some significant debate and some key developments over the last couple of years. And further potential changes remain on the horizon but maybe not getting as much of the popular discussion lately as some of the other issues. So I’m really happy today that we can kind of dive into some of these.
Last year, for instance, we saw the antitrust agencies — the antitrust division at the Department of Justice and the Federal Trade Commission release updated guidance on vertical mergers for the first time in over 30 years. And as our panelists I’m sure will discuss, this was quite an interesting process and not one without some controversy. Additionally, the FTC is currently poised to begin litigation to block the Illumina/Grail merger, which would be the first test of a vertical merger in court or, in this case, in the FTC’s administrative process under these new guidelines.
I am only going to offer the briefest of introductions to our panelists, and it won’t come close to doing justice to all of their tremendous experience and achievements. But it will leave us some more time for substantive discussion. So I hope you all will forgive me. Steve Cernak is a partner at Bona Law and previously served as in-house antitrust counsel at General Motors.
Michael Kades is the Director for Markets and Competition Policy at the Washington Center for Equitable Growth. He previously served as antitrust counsel to Senator Klobuchar and spent several years as an enforcer at the FTC, including as an attorney advisor to Chairman Leibowitz. Bruce Kobayashi is a professor at Antonin Scalia Law School at George Mason and recently served as director of the FTC’s Bureau of Economics.
Tom Lambert is a professor at Missouri School of Law. And Taylor Owings is a partner at Baker Botts and recently served as senior counsel and chief of staff in the antitrust division at the Department of Justice. So even from that brief overview you can see we’ve got a real powerhouse of experience here.
And we thought we’d begin our discussion today with the vertical merger guidelines, or the VMGs as we might refer to them, to offer some background and provide a bit of a lay of the land before we get into our discussion of the Illumina/Grail case. So Taylor, you were at the antitrust division for some key moments that led to these VMGs. Can you kick us off by providing some background on how these VMGs came to be realized?
Taylor Owings: Thanks, Elyse, and hello to my fellow panelists and hello to the audience members. It’s a pleasure to be here with you today. I’d be happy to kick off the panel with a little bit of background to how we got to the regulatory sort of status quo that we’re at today.
So starting with just some basics on the way that the enforcement agencies and the courts look at vertical mergers in comparison or in contrast to horizontal mergers, the horizontal merger guidelines that are tried and true, have been adopted by courts as stating the proper jurisprudence and task for shifting the burden in a merger case, they really speak to the phenomenon of a merger taking a competitor out of the market entirely. Now, vertical mergers have, of course, a more indirect effect on competition. And at their worst what we’re really looking at to see if a vertical merger will cause competitive problems is if a vertical merger gives one company in a competitive market control over something that everyone else in that market needs.
So it’s, of course, obvious to see why vertical mergers might be a hot topic right now. Much of the concern in the digital markets space is about one company owning something that everyone else in the market needs. Now, while traditional theories of monopolization don’t prevent companies from withholding access to such a needed good or service or charging high prices for access, enforcers do have this prophylactic tool for avoiding getting into that situation in the first place when a company would buy through vertical merger that gatekeeper status rather than build that gatekeeper function organically.
So I think interestingly the D.C. Circuit opinion in the AT&T/Time Warner case was very clear on this comparison — this point of comparison. And it’s a, I think, significant development in the jurisprudence that the court applied Section 7 to that vertical merger case, and it said, quote, Section 7 applies a much more stringent test than does the rule of reason analysis under Section 1 of the Sherman Act. So I think that that sets the stage a little bit for why the issue of vertical mergers might be especially relevant in today’s climate where there’s concern over potential gatekeeper status in the structural arrangement of companies that provide platforms for other business to operate.
So going back to discuss a little bit of the chronology of how we got to where we are today with the vertical merger guidelines, I think that the widely known highwater mark I think of vertical merger litigated cases was in the ‘60s and ‘70s. The seminal Brown Shoe case decided in the ‘60s involved a merger between one of the largest shoe manufacturers and the largest independent chain of family stores in the nation. That is a direct quote from the opinion because the opinion doesn’t do much to explain the power of market shares of the retailer distributer in that case. But it does describe it as the largest independent chain of family stores, which does seem like a lot of qualifiers.
That Brown Shoe case discussed the test for illegality of a vertical merger as resting on the percentage share of the market that was foreclosed when the manufacturer forced its shoes into the retailer. But of course it also discussed in rather vague term that there was a general trend towards manufacturers buying up retailers. And that is what the Court saw as problematic in that case.
Other cases in that era that establish our somewhat thin jurisprudence on vertical mergers were the DuPont case which unwound DuPont’s years’ earlier acquisition of 23 percent of General Motor’s stock because the Court held that that acquisition foreclosed sales to GM by other suppliers of automotive paints and fabrics — and then the Ford Motor case decided in 1972 where the U.S. enjoined the car manufacturer from acquiring a spark plug manufacturer. So other than that, there was one privately litigated case against Volkswagen in 1977, and that got to the Fifth Circuit, not the Supreme Court. And that’s it for litigated cases until recently — until the Department of Justice challenged the AT&T/Time Warner merger in the fall of 2017. And there was also a case that we may get into. I won’t spend time on it in my introductory remarks here — a privately litigated case between Steves & Sons and JELD-WEN — a vertical merger case between a door manufacturer and a door skins — vertically integrated door skins door manufacturer. That was decided recently as well, and the appeal just decided a couple of months ago.
So even though the 2017 challenge to AT&T/Time Warner was a new business — or excuse me, new in the since of it being a long time since a litigated case — and, of course, we’ll get to talking about the Illumina/Grail case following in that footstep as well. It doesn’t mean that the FTC and DOJ weren’t enforcing the antitrust laws against vertical mergers in the intervening time. Indeed, a much cited study by Professor Salop counted 58 enforcement actions by the agencies between 1994 and 2018 alone. And so I think the question for observers is why so many enforcement actions but so few litigated case.
And I think just to set the stage for the discussion I think you can sum that up in probably two points. The first is that vertical mergers have known efficiencies, and for that reason it is incumbent upon the agencies to weigh those efficiencies in deciding on their prosecutorial tact. So as I’m sure we’ll be discussing in more detail later, there is a phenomenon known as the elimination of double marginalization.
That is largely — or that is often considered to be the most certain efficiency created by a vertical merger. So when two independent firms supply to each other successively in a supply chain, each tends to earn a markup that is then past on to the consumer in the price of the final good. Both firms would be better off if the price to the consumer were lower and more consumers bought the product so when the two firms combine they can charge just one markup —that’s the elimination of double marginalization—and consumers get a lower price. So this sort of efficiency is of course the type that’s cognizable under the antitrust laws and weighs heavily in the analysis of the enforcement agencies and is one of the reasons that you see more settled cases rather than enforcement actions.
And then the second reason, I think, for that trend is that a behavioral remedy — or excuse me, a remedy that is sure of blocking the deal outright may make more sense in a vertical case than in a horizontal case. And indeed, that was the remedy in many of the cases that Salop logged in his study. So that’s the type of remedy where a company like NBC would be required to supply its content to rival distributors — to Comcast even once Comcast owned NBC.
And I would say that the current debate around vertical mergers and the need for new vertical merger guidelines is really centered around the question of how certain certain efficiencies are and whether behavioral remedies like the type in Comcast/NBCU actually work and whether they are appropriate in cases where it’s hard to understand what price should you charge for the — what a competitive price would be for the content in the NBCU example — the content supplied by NBCU. So we saw this debate play out in the arrival at the new vertical merger guidelines. The DOJ had had non-horizontal merger guidelines that were current at the time of the AT&T/Time Warner case. But those were published in 1984 and were widely considered to miss out on much of the detail that I’ve been discussing about what makes vertical mergers harmful, what countervailing factors might make them good for consumers, and the experience of litigating the AT&T/Time Warner case and the resulting D.C. Circuit opinion that embraced much of the analysis that the DOJ was using I think was a spur to the agencies to update the guidelines.
The agencies worked together and indeed put on a series of workshops. Unfortunately, one of the actual live workshops had to be canceled due to COVID. But after receiving all those submissions and conducting one of the live workshops, the guidelines went from draft to final in June of last year.
And one of the most notable aspects of the debate that played out sort of in the live workshops and was ultimately changed as between the draft guidelines and the final guideline was that there was originally a safe harbor in the draft guidelines indicating that vertical mergers would not be of concern to the agencies if the company had less than 20 percent share in the relevant market where competition would — in the relevant upstream market or the related market. If it was used in less than 20 percent of related markets, then there would be a safe harbor. That was excluded ultimately from the final guidelines issued in June.
And we continue to see debate about whether there is more that the guidelines could do to set out more aggressive stance toward vertical mergers in addition to deleting essentially that safe harbor that had been present. And so I would hand it back over to Elyse to guide the future discussion.
Elyse Dorsey: Taylor, thank you so much just for that kind of great overview of all of the different facets of enforcement, not just the litigation but as you mention consent decrees and the guidance and other ways in which the antitrust agencies are kind of telling the world what’s okay or not in the vertical merger space. So Bruce, you were down the street at the rival antitrust agency, the FTC, when the VMGs were released. Kind of what did that look like from your end, and maybe from an economic perspective what did you see in kind of considering what to put out there?
Bruce Kobayashi: Thanks, Elyse, and thanks to the Regulatory Transparency Project for inviting me. Just a couple of corrections, I actually was there when we were thinking about things. But I left even before the draft ones came out. So they’re Andrew Sweeting’s fault, not mine.
So I think at this point going second in any panel is great because the first panelist says a lot of the things that you were going to say, and I get to cross off a lot of things. But one thing I want to sort of talk about was that one of the things that actually was really important—and it may seem like a technicality, but I don’t think it is—is that the vertical merger guidelines do not treat EDM as an efficiency. EDM’s not a production, research — R&A or procurement efficiency. It is actually the alignment of incentives between the merging firms.
And so if you think about a vertical merger as Taylor said, in a horizontal merger you have — based on unilateral effects you have recaptured diversion and upward pricing pressure. In all those models there’s always upward pricing pressure. Vertical mergers you have both upward and downward pricing pressure to the extent that there’s incentives for RRC and EDM. And this sort of recaptured diversion from RRC and the incentive to sort of eliminate double margins are really the same thing. They’re basically the merging firms internalizing a pricing externality.
And so when the economists — and this has not included just the FTC economists. We worked really well with the AAG economists on this project. All our meetings were really collegial, productive, and there was just a lot of agreement that really what EDM is is unilateral pricing effect. And if you’re going to look at EDM and RRC, they ought to be looked at together, not separately and not as an efficiency but as part of sort of the unilateral incentives generated by the merger.
So Taylor called it an efficiency. The guidelines treat EDM separate from efficiencies. It is a unilateral pricing effect. You should do them together. There’s actually a nice paper that was published last year in the Antitrust Bulletin by Das Varma and De Stefano which basically shows that if you do sort of a general equilibrium model, the size of the RRC incentives are going to be determined by the size of the EDM. And if you sort of do these things in sequence and separately, you get a much different answer than you do if you have a model which takes into account the full incentives of the firms. So those two things I think are really important, and I was very happy to see those things go through in the vertical merger guidelines.
Second, a lot of people were complaining about whether or not there was a presumption of legality. There were no presumptions as Taylor said. I saw no reason to put any sort of structural thresholds in. I was happy to see they were out in the final version.
There’s not a presumption that a vertical merger will generate either of the unilateral incentives that are talked about in the VMGs. There could be no EDM and no RRC if somehow the parties contract to align their incentives pre-merger. What you want to do is you want to sort of look at actually see. You don’t want to make things up.
So for example, you shouldn’t just assume EDM or use imaginary EDM. And this is a bit of a dig at the FTC with respect to Illumina/Grail. You also shouldn’t be using imaginary future diversions. So either of those, they should be based on facts.
With respect to — and I’m sure we’ll get to this in the discussion of Illumina/Grail. Usually it’s the case that you have — our models of unilateral effects say we measure the margins. We figure out the diversions, and we can sort of get at least a model-based estimate based on the margins and the diversions of upward pricing pressure in a horizontal merger.
When you talk about efficiencies, it’s usual a sign to the parties that they’re in a lot of trouble. And often efficiencies are treated as speculative, not cognizable, not specific to the merger. Here in Illumina/Grail, jumping ahead a little bit, we actually have — the contract — the merger will eliminate a sizable (inaudible 00:24:47) royalty on Grail’s revenues. So there is cognizable and real elimination of double marginalization. And as I said before, there are imaginary future diversions because there’s no products being sold yet.
The last thing I’ll say is the reason that we wanted EDM out of the efficiencies section is because the efficiencies section says you treat efficiencies like we do in HNG Section 10. And on the ground that means they’re never credited. But there’s a vast literature starting with Coase’s 1937 article; Oliver Williamson; two guys who won the Nobel Prize; Klein, Crawford and Alchian—people who taught me at UCLA—of the sort of way in which vertical integration is used in a pro-competitive way to solve problems like limiting transactions cost and opportunistic hold up. I’m sure we’ll go into sort of innovation theories, but the same types of considerations hold.
With respect to innovation, there’s a great article— it’s like 30 years old— by Tim Bresnahan and Trachtenberg. I think it’s published in the Journal of Econometrics. It’s a theory piece, and it really shows that the same type of externalized effects apply to R&D in a vertical mergers and vertical mergers can solve sort of the uninternalized spillover effects and improve incentives. And so when you’re looking — if you want to sort of say, well, this is an innovation merger, you have the exact same tradeoffs. Right? You have more innovation now and perhaps although even more speculative harms to innovation in the future. And on a present value analysis it’s hard to see how the current benefits are going to get outweighed by the speculative and far in the future costs. So I’ll stop there.
Elyse Dorsey: Thanks, Bruce. I think you touched on in there a lot of kind of the difficult considerations in terms of especially with EDMs and presumptions of how the guidelines were going to handle that and what the agencies were, again, kind of going to be telling to the world in terms of how it is they’re going to look at these cases. Michael, you and your center have done quite a bit of work in this space and I think might have some different perspectives on how the agencies should be approaching vertical merger enforcement. What can you tell us about your perspective and what you saw in the guidelines versus what you might have liked to have seen in the guidelines? I think you’re still muted, Mike.
Michael Kades: Sorry about that. It will happen at least three times in this presentation. First of all, thank you very much for having me. I appreciate it. Whenever I do FedSoc things, I always feel like I don’t live up to the promotion. I’m going to take a more sort of positive as a normative approach here for the discussion and just sort of comment a little bit because obviously the first two speakers have covered the groundwork very well.
The first thing I sort of feel is important to realize is that I always thought that the vertical guideline project was a real challenge for the FTC because even before the guidelines you had clearly a huge divergent of opinion on how to do vertical merger enforcement. And you can see that in the Staples/Essendant case where that was 3-2. So it didn’t surprise me that we ended up with guidelines that were 3-2, and maybe we can talk a little bit about what that means.
The second thing is I find this always when you’re on the inside writing guidelines the words really matter a lot, and sometimes things about where EDM gets discussed, whether you explicitly say there’s a presumption or not, seem to be really important points. But at the end of the day, I think the proof’s in the tasting of the pudding. And it is really in how they fare in the courts. So I think this is very much a project that is not complete. And we just have to wait to see how it plays out in Illumina.
The other interesting — what I really found interesting about this was at least given that I’ve been doing antitrust for too long now I think these guidelines do reflect a change from where we were 20 years ago that, when I was at the agency, especially early on, yes, there was — absolutely the agencies did some vertical merger enforcement. I think it is important to understand that — take that with a little bit of a grain of salt because they weren’t litigated cases. A lot of settlements in merger cases — if all you have is settlements, it’s hard to really understand what the real principles of the law are and that really sort of internally the thought was, yes, vertical mergers can be in theory anticompetitive. They are almost never anticompetitive.
I don’t think you can read these guidelines as saying that at all. I think quite to the contrary they take a much more neutral approach. I think it’s also very interesting to reflect that when we look at the commentators. You had from George Mason Professor Wright, very well respected economist, former commissioner, sort of taking a very sort of lenient approach on how to think about vertical merger enforcement. That’s not surprising. I think what was surprising is that you had a number of Iowa economists, people who — Nancy Rose, John Baker, Steve Salop, Fiona Scott Morton — who said, no, that lenient approach is wrong. The empirical literature has moved in the last 20 years. The theoretical literature has moved, and we need to be more skeptical.
What I see the guidelines as doing is sort of kind of splitting the different, so if you look at where the sort of aggressive Iowa economists were, they were suggesting create some presumptions that the guidelines didn’t do that. But I think reflect in policy view that is somewhere in between sort of where we were on merger enforcement 20 years ago and where sort of more aggressive — the more aggressive economists were advocating for.
But as I said, take the EDM issue. I kind of think that we can have a lot of talk about — and it’s a fair discussion — is it part of sort of the case in chief? Is it something that you just naturally deal with in thinking about the merger, or is it a pro-competitive justification of efficiency? And I’m not going to disagree with Bruce on the economics. But from a litigation perspective, it just seems to me in every vertical case the defendants are going to argue EDM.
They’re going to argue that, and if you don’t convince the judge that that’s not an overriding benefit, it doesn’t really — saying it’s an inefficiency justification versus the government’s got to deal with it upfront or whether it’s a burden shifting production because the parties have the best evidence, I don’t think — I mean, I think it will really come down to what evidence gets brought to bear in the individual cases. And as I said, we need to see where the judiciary is because for 40 years we’ve had no cases that were litigated. We had the AT&T case, and we have the Illumina case.
And so the other thing I think is worth thinking about is we’ve now had a switch at the FTC in terms of there’s now a democratic majority, including two of the dissenters. And I think we — I don’t know if the Chairwoman has spoken specifically. But I think her views on vertical merger enforcement are pretty easy to ascertain. And obviously you have a change at the Justice Department.
So the real question is — I mean, the one question is, A, how long do these vertical guidelines last and, B, whether changing those now — what implications does that have both for merger enforcement and how the courts react to merger enforcement? And I think one can say that this administration will redo the vertical guidelines. They will be far more aggressive. Maybe that’s helpful in litigation, or maybe courts look at that and go this is an area of partisan disagreement. Obviously, the agencies don’t have a consensus. We’re just going to ignore what the guidelines say.
So I think that’s a second really important issue. And I also think there’s a lot of discussion in the press this week about an eminent Executive Order. We don’t know what will be in it, but I would not be surprised if that order actually asks the — directs DOJ and asks the independent agency, FTC, to modify or redo the merger guidelines, maybe both the horizontal and the vertical guidelines. So that’s it — all I have for this.
Elyse Dorsey: Well, that was super helpful, and I think it hit on kind of a lot of the different interesting issues we’re running into here. I think your comment kind of on what is this next administration going to do with the vertical merger guidelines is a particularly interesting one. The FTC recently, I think just last week, voted to withdraw its guidance on the Section 5 statements. So I think there are certainly questions of are they going to take a similar approach to the vertical merger guidelines. And I think your point also in terms of what the courts ultimately do with all of this shifting guidance, putting it out, taking it back is a really important one.
I think one of the reasons the horizontal merger guidelines have ended up being so effective in court over the last several years is because the agencies have put them out there and they’ve updated them occasionally but through some pretty rigorous processes. And they’ve hewn pretty close in their cases that they bring to those guidelines. And that’s allowed the courts to say, okay. You’ve been clear here. This is a clear path, and you’re doing what you say you’re going to do. I think it maybe presents some interesting challenges.
Taylor or Bruce, did you kind of want to weigh back in on anything? Otherwise, I think we can turn to Illumina/Grail. Taylor?
Taylor Owings: Sure. I thought a lot of what Bruce and Michael had to add was really interesting, and for me, one of the things that it brings up is what are we really looking to guidelines to do. I think there’s a tendency to believe that guidelines sort of automatically change the analysis that courts are going to do. But of course, I would just remind everyone that that’s a process unto itself. It’s somewhat taken for granted now that courts will accept the horizontal merger guidelines’ HHI presumptions and use them as a guide to shifting the burden back and forth. But it’s hard to say which came first, the chicken or the egg, on that one.
And you certainly saw — what I think would be great is if guidelines reflected both what agencies intend to do in terms of selecting cases in the way that they construct cases to bring them to court and also reflected enough economic wisdom that a court would be convinced to take them as guidance for itself. I think that’s the real challenge in thinking about whether even poetically minded or sort of policy minded leaders who want to push further and to test cases further, the question is do you really want to rescind guidelines and change them if part of the value that they can provide is asking courts to come along in what is sort of verifiable analysis.
I am as big of a critic I suppose is a fair way to put it of — or at least critical thinker on the question of whether EDM ought to be included in the case in chief. And that comes from my experience watching the AT&T trial play out. I think the real problem with putting EDM in the government’s case in chief and kind of trying to pre-bought whatever the defendants might raise as a positive change in incentives for the merger — I think the problem with it is that it creates a natural tendency for a judge to want everything to be quantified. A judge when confronted, as Bruce explains, with verifiable contracts and the path that have included markups and can see a quantifiable number about what the EDM effect is going to cause, it creates a sort of impossible standard for plaintiffs who would be arguing about the likely effects of a future foreclosure, future raising rivals’ costs effect of a vertical merger that’s always going to be more speculative.
So I think putting those two types of evidence in the same offering by the plaintiff upfront, it sort of unnecessarily seems to goad a court into thinking that the merger inquiry isn’t necessarily speculative and isn’t necessarily a matter of degree for how confident you can be that certain effects will take place in the market in the future. So that’s one reason I think to take seriously the debate about kind of where the EDM discussion goes in the guidelines and also to be aware of what a potential edit to the guidelines would do to bringing a court along and convincing a court of the right way of analyzing these mergers.
Bruce Kobayashi: Okay. I’ll be real quick because I know we’re sort of short on time. I just want to say Staples/Essendant — Michael mentioned it. I mean, that was not a vertical merger. That’s called a diagonal merger. It’s actually treated in the VMGs. There’s no EDM. I can’t tell you how many times I went to the Commissioner’s office and said, this isn’t a vertical merger. So just sort of addressing that.
But it’s actually in — the discussion of it is in the FTC’s vertical merger commentaries. And it really is one of the reasons we put diagonal mergers in the VMGs. And I’ll just follow up after everybody’s been able to get on.
Elyse Dorsey: Thanks. Yeah, no, I think that’s a really excellent transition also kind of into our discussion of Illumina/Grail. As Taylor and Michael, you had mentioned, there’s sometimes a tension between what the economic theory is and then how you translate that to a judge and to a court. And I think that’s part of what makes the upcoming Illumina/Grail litigation so potentially interesting. Steve, you recently authored a piece highlighting the importance of this case, again, noting that it’s maybe not been receiving as much of the attention as it perhaps deserves. Can you provide us kind of a bit of the overview of the case and how it came to be in its current procedural posture?
Steve Cernak: Sure. Let me go ahead and do that. My apologies for dropping off for a couple of minutes. I missed you, Michael. I had some technical difficulties here with some bad weather.
Yeah, so I think the Illumina/Grail merger is extremely interesting for lots of reasons. First, it’s some fantastic medical technology, so that’s great. But also, it’s got some interesting procedural issues and the interesting substantive issues that we’ve been dancing around here talking about the vertical merger guidelines.
So Illumina is a San Diego based company. It’s a provider of a certain type of DNA sequencing, including instruments, consumables, and reagents. And according to the FTC’s complaint, at least, it is the dominant provider of this DNA sequencing. There’s another smaller U.S. provider. And there’s a Chinese provider who is currently blocked from selling in the U.S. due to patent disputes.
Grail is one of several companies that is developing an MCED, multi-cancer early detection test, that promises to be able to detect biomarkers associated with up to 50 types of cancer by extracting DNA from a simple blood sample. So that sounds great. But in order for this MCED to work, it needs some sort of DNA sequencing. And at least according to the FTC complaint, the type of DNA sequencing that works best and certainly the type with which Grail and all the other MCED developers have been working is the type that’s supplied by Illumina.
Now, the parties announced the acquisition by Illumina of Grail in September of 2020 and claimed that it would speed the adoption of Grail’s MCED development and enhance patient access to the tool. Now, Illumina also told its investors that the transaction would allow it to participate in the high value clinical market. So it seems that perhaps there’s more money in the downstream market.
Interestingly, Illumina actually did found Grail back in 2016, then spun off most of its ownership interesting, and now it’s buying it back— the rest of it that it does not currently own —for about $8 billion. So the HSR filings were made. The FTC challenged the transaction in March. And as Elyse mentioned earlier, it was doing that through the administrative process at the FTC. And that hearing is going to be going forward starting in late August.
But in the meantime, the FTC, as it often does, sought a temporary restraining order and a preliminary injunction from the U.S. district court for the District of Columbia. That’s not surprising, but the surprising move was the parties’ success in mid-April in getting the case moved to the southern district of California where both of the companies are located. And then in late April, the European Commission decided that it too would investigate the transaction, even though the transaction did not meet its usual thresholds. But several of the member states request the Commission to step in.
The parties have challenged the Commission’s jurisdiction, but in the meantime, the transaction cannot close because of what’s going on in Europe. So as a result, the FTC decided that it did not need its federal case to block the closing of the merger prior to its administrative hearing. So it moved to dismiss its case out in California.
The parties fought that, wanted to make sure that the FTC couldn’t come back in and try to block the merger later on. But the court did allow the dismissal of the case and would allow the FTC to refile the complaint later on if necessary. So as you might expect under the vertical merger guidelines, the case revolves around foreclosure.
First, the complaint says that Illumina has the ability to foreclose on the other MCED suppliers, that is they could refuse to sell the DNA sequencing to those other MCED providers. Or they could sell it at a higher price. Or they could sort of slow walk it and delay the sales and perhaps delay the development, delay the approval of the MCED tests by those other suppliers. Also, the complaint alleges that Illumina would have — the combined company would have the incentive to foreclose those suppliers. It’s difficult to tell with the redactions in the public complaint, but it appears that any loss of revenue from the DNA sequencing sales to other MCED suppliers would be swamped by the increase in revenue if Grail is the first one to develop one of these MCED tests and is as successful as the company and the FTC thinks they will be.
Now, part of the complaint relies on some information that the company has provided that thinks that the combined company will be very successful. But in addition, the company has sort of taken a page out of the AT&T playbook and has said that it would enter into a 12 year — so a long-term supply agreement with all the other MCED companies that need — or at least potentially need this DNA sequencing. And it would keep pricing flat to actually decreasing. And the contract makes other promises about not having the kinds of delays that would slow down development of the MCED test.
So one of the questions that this raises is why isn’t that contract enough? What is it about that contract that the FTC thought would not be sufficient, and what is the administrative law judge in the first instance going to make of this? Will they take the close look at it that the judge in the AT&T case looked at as well?
The two other issues that I would flag here, Bruce talked an awful lot about EDM, and Taylor commented on it as well — whether it should be part of the unilateral effects analysis or whether its some sort of efficiency — maybe a special efficiency, but an efficiency. Well, in the FTC’s complaint it’s treated very quickly as an efficiency, not as part of the basic analysis that the FTC is doing. It’s sort of mentioned in paragraph 78 as something that is not present and would not overcome in any of the potential anti-competitive effects that the FTC describes in the complaint.
And then the final point that I’d make, Elyse, is once again the complaint, as it should, talks about the high market share that the combined company would have in the MCED market. The exact prediction is redacted. But evidently, the FTC thinks that the combined company will be very successful. And it bases part of that prediction on some internal documents from the company — from Illumina.
But remember, this is different than a horizontal case and different than a normal vertical case as well in that normally we’ve got some past sales to look at. And those past sales can at least provide some basis for prediction of what the future market shares or future success or not of the company will be. Here, as Bruce mentioned, we don’t have a market. None of these MCED developers has been successful yet. They haven’t developed it. They haven’t gotten approved. They haven’t sold anything.
So once again, I think it’s important for all of us, the company as well as the FTC and then the judges in Europe and at the FTC and potentially here in federal district court, to be careful about what it is that we think about predictions of the market. Those kinds of predictions are difficult enough to make when you’ve got past sales to look at. When you don’t even have any sales at all by anyone, I think those kinds of predictions can be even more troublesome.
So that’s a brief summary of the case, procedure, and substance and I think some of the issues that I think are key. But Elyse, let me turn it back to you so we can talk about it further.
Elyse Dorsey: Yeah, thanks, Steve. I think you kind of highlighted a lot of what we’ve been discussing so far. It seems like it’s going to be presenting some issues in the Illumina/Grail case as well. And Tom, you also had a really thoughtful piece that you wrote recently on kind of some of what’s happening here, the legal theories, and maybe some of the shortfalls. Can you kind of talk to us about what you see in this case?
Thomas Lambert: Sure. Going fifth, you really have very little to say that hasn’t been said already. And that was a really terrific presentation of the case by Steve. I guess I’ll just make a couple of quick points, and that’ll give us some extra time for Q&A.
But one is just sort of about the business realities of this case. It seems to me that if you think about it, Illumina did everything right to actually create this multi-cancer early detection test. It created Grail in 2015. It created it as a subsidiary. It provided the bulk of the capital for the company, so it had a controlling interest.
And then a couple years later it didn’t sell its interest. It caused the company to raise money basically by selling more shares. So they raised a billion dollars, which of course diluted Grail’s share so that it fell down to 14.5 percent.
But holding the company as a separate company was a good thing. It allowed the company to go out and raise equity financing. It allowed the managers of the company to really focus on one thing, not distracted by other things that would be going on at Illumina. It allowed the company to attract really talented managers. The first CEO of Grail was a very prominent former Google executive. And Grail succeeded.
And so now it just wants to recombine, and the FTC is blocking that. It seems to me this is an unfortunate thing to do. It will limit options for companies in the future that want to basically innovate and do it by creating a subsidiary and letting that subsidiary raise equity financing. So that’s just kind of a big picture thing.
But going to the actual case, I guess Steve discussed it in great detail, so I’ll just make three quick points. One is that the case does track pretty closely the vertical merger guidelines. It’s a straightforward input foreclosure claim that’s based on the combined company having both the ability and the incentive to foreclose rivals from the necessary input, very straightforward. The Commission pays lip service to possible pro-competitive benefits. It mentions EDM and also efficiencies, but it gives them very short shrift. There are really three sentences in the complaint that mention EDM and potential efficiencies. So we are paying lip service to the guidelines, but I think the big issue here is in an actual implementation of the guidelines.
So I see two things that kind of worry me here, and they’ve both been mentioned. So I’ll just briefly say what they are, and then we’ll turn to Q&A. But there’s sort of an asymmetry in speculation in this case. The Commission is allowed to speculate about what’s going to happen and diversions and all that sort of stuff. And it’s necessarily speculating because there is no MCED market at this point. So we’re sort of just assuming if there’s a market, this is what the incentives would be.
On the other hand, when it comes to efficiencies and EDM, the parties are held to a very exacting standard of proof, and the Commission says, look, you can’t prove these efficiencies. You can’t prove the elimination of double marginalization, and so this is on balance in anti-competitive merger. It seems to me the EDM is pretty obvious in this case. Illumina is the sole supplier of these NGS platforms that are necessary input. It’s charging a significant margin most likely.
If Grail’s test comes to be, it will be the only MCED test out there. There should be a significant margin there. There’s going to be double marginalization. This merger would eliminate the double marginalization. That seems so obvious to me.
In addition, the efficiencies are significant and also pretty obvious. Grail is a small company that’s been focused entirely on product development. To get to the next step, it needs regulatory approval. It needs to establish relationships with third party payers. It needs to get regulatory approval across the globe. It needs to figure out how to manufacture in compliance with the FDA’s very stringent rules for manufacture of medical devices. There’s hardly any — I can’t imagine any company out there that has more experience in those areas in the genetic space than Illumina. So Illumina could very quickly help Grail bring its product to market.
And those to me seem like really obvious efficiencies that the FTC is just blowing off. So I’m concerned with this sort of asymmetry. We can speculate, but we’re going to hold you, the parties, to a high standard of proof.
I’m also concerned that the Commission has set an impossible standard for allaying anti-competitive concerns with some kind of an agreement or really ultimately with some sort of a conduct remedy. There’s a paragraph in the complaint that’s, I think, really troubling. “Any existing or potential supply agreements between Illumina and third party MCED tests cannot offset the likely anti-competitive effects because these agreements cannot account for each and every current and future method by which Illumina may foreclose, raise the cost of, or otherwise disadvantage Grail’s rivals.” That’s a really super high standard.
Steve mentioned the agreement — the option contract that’s been offered by Illumina to its oncology customers. It is a very thorough contract that seems to alleviate every single one of the anti-competitive concerns that the Commission has raised. So bottom line, I think that if this case goes to court that the FTC will loose. EDM is big. Pro-competitive efficiencies seem pretty obvious. There’s a lot of speculation about anti-competitive harms. And then we’ve got this agreement that seems to alleviate any anti-competitive harms that do exist.
I’m frankly a little bit concerned that the FTC might view — or at least a majority of the commissioners might view a loss in this case as a gain in the larger war. I mean, the FTC is on a losing streak right now, Qualcomm, 1-800-Contacts, Facebook. And every time it loses, what we hear is, this shows that we need new laws. We need Congress to step in and give us more power, change the standards, or whatever. And so I’m a little bit worried that this case will be brought, will be lost, and then will be spun in that way to get some legislation that I think would probably not be good. Thanks.
Elyse Dorsey: Yeah. Thanks, Tom. That, again, kind of highlighted a lot of the challenges the FTC is facing and kind of broader the more kind of regulatory environment that we’re in. As you highlighted, there’s kind of the aspect of it where these are medical devices. There’s a lot of medical regulation that happens. On top of which we’ve got this — there are a lot of hankering for legislation lately that might come into play.
So we’ve got a few minutes left. I want to make sure we open the floor up to the audience. So a reminder, if you have questions, you can use the raise hand function, and we’ll call on you and have our administrator unmute you so that you can ask your question. In the meantime, Bruce, I think you had some other thoughts you wanted to share with us once everyone kind of got their opportunity to weigh in.
Bruce Kobayashi: Thanks, Elyse. Tom covered most of the things that were on my checklist. Elyse, as — I mean, Taylor — sorry, Elyse. Taylor expressed the lawyer’s view of this that we don’t want to give away our Baker Hughes advantages. But I think the concern and the reason we really did want EDM taken out of sort of the — we’ve never seen an efficiencies bucket and put into unilateral effects is because it is a unilateral effect. And when we at the FTC, at least when I was there, looked at vertical transactions, we did both. We took a serious look at trying to measure both, and then we tried to figure out the net effect.
Remember, horizontal mergers, if you’re just doing UPP, it’s always up in sort of the standard unilateral effects model. The vertical mergers, things go both ways. You can have raising rivals cost that harms sort of the non-vertically integrated downstream rival of the now merged firm. Yet, because you’ve created a more efficient vertical structure, that has downward pressure. So you could harm the competitor and you could not harm consumers.
And so the problem is is that — Taylor started with Brown Shoe, but Brown Shoe is where the Court said antitrust is not about the harm to competitors. It’s about harm to competition. When we’re thinking about the downstream market and what’s going to go on to consumers, vertical mergers are more complex. It’s just that it’s the fundamental theorem of antitrust raising its head.
When you combine complements, it’s generally good, absent some other factors. And when you combine substitutes, it’s generally bad. And these vertical mergers with the possibility of input foreclosure have both, and so you have to do both. And they’re a little bit harder to do. But that’s just the nature of the beast. And what economists want to do is they want to figure out whether this is going to be good or bad, not make broad pronouncements about let’s not let firms get to big or dominate or have some European style dominance regime. So once again, I think what the guidelines do is they’re not sort of — they were just sort of a way in which the agencies tried to describe what we do. And it’s not much guidance in my view, but it is a description.
I want to say just a really short thing about the contractual promise by Illumina. There’s a maybe well-deserved skepticism of behavioral remedies. Certainly, they’re almost, per se, non grata in horizontal mergers. They’re used a lot in vertical mergers. They’re mostly sort of firewall based, based on sort of not using information when you sell your inputs to your competitor about what you’re going to do with it.
There are other ways in which these behavioral remedies are problematic. And you think about ABI/Modelo. You say, okay. We’re going to sell it at this price. And it sort of brings up the same problems as you have when you’re thinking about doing duty to deal cases. You always have to figure out at what price and what the terms are and what if conditions change and all that.
And so some people say antitrust has this bright line of separation between price regulation and antitrust. It’s been broken sometimes. But in general, it’s true because doing price regulation is not really within the institutional competence of antitrust.
And the one thing about this thing, this thing is an MFN. And what it is is it says, look, we’re going to give everybody, including people who don’t have competitive downstream products to something that Illumina has downstream — we’re going to give you their treatment. And so it sort of becomes a relative — you have a relative marker about what market competition is going to absent RRC incentives. In that way, you sort of solve this problem about changing conditions. If conditions change, then the contracts to sort of unrelated downstream products will change. And you could use that sort of as a reference point for enforcing the 12 year contractual remedy that’s proposed.
Elyse Dorsey: Thanks. Yeah. Again, I think this is an interesting case. I’ll also kind of take my privilege as the moderator to kind of open up the floor to ask you all — we’ve been talking a lot about how vertical merger enforcement has kind of developed and maybe started changing over the last few years. Do you see this case as kind of more evidence of that? Do you see it as the turning of a tide? Is that overstating it? What are kind of all of your thoughts on what this case means for what we’ll see over the next few years?
Taylor Owings: Well, I’ll chime in. Oh, go ahead.
Michael Kades: Sort of ironically I think the most interesting fact about the Illumina case, having Bruce and Tom state how awful it is, is this case was voted out 4-0. That includes both Commissioner Wilson and Commissioner Phillips, both of whom would be sympathetic to every single concern both Tom and Bruce have raised. It doesn’t mean the commissioners are right, and they’re wrong.
But I think it suggests to me that the complaint, which is a litigation document — it’s not a decision. You allege conclusions. But it suggests to me that there must have been evidence that convinced two people who are fairly skeptical of overenforcement in vertical cases that this was a case to bring, even though you’re dealing with a market in the future. That’s uncertain — even though there seems to be sort of obvious evidence of EDM.
And so I think what we need to do here is sort of — I put a lot of faith in the fact that there’s a viable theory here that the harms of foreclosure, although uncertain, could be massive because the market that’s going to exist is going to be huge, in the tens if not hundreds of billions of dollars. And I think this reflects a Commission that — the shift here is maybe less ideological on merger enforcement and more a sign that there is a consensus at the Commission— maybe the only place there’s a consensus at the moment— that enforcers have been too concerned with overenforcement and not concerned enough with underenforcement.
So that’s my take on this case, that this is an aggressive case. But the fact that it’s 4-0 makes me think that there’s got to be a reason why this was 4-0 and that we just need to see where the facts develop and whether they can persuade the ALJ or eventually the court of appeals.
Taylor Owings: I agree, Michael, that one of the important things to keep in mind about the complaint— and Tom read a few quotations from the complaint— the complaint is not reflective of the decision making that went into the decision to sue. And so going back to the concept of how involved a complaint or a case in chief should be in laying out the exact way that EDM works in this case, I am sensitive and sympathetic to an approach that would mention it in the complaint but allow the defendants to be the chief propagators of that evidence. And I think taking of — trying to take the conclusion from the short shrift giving in the complaint that it wasn’t weighing heavily on the minds of the commissioners and in the staff’s analysis whether EDM was verifiable, whether there potentially were examples of the elimination of double marginalization even in the current relationship with the companies — I don’t know the facts of the case obviously well enough to say whether that’s present here.
But the whole point of Baker Hughes burden shifting — going and putting my lawyer hat on, Bruce. The whole point of Baker Hughes burden shifting is that it’s not until the third step that you’re doing the net analysis of whether the benefits from the merger outweigh the harms. And one of the points I was trying to make with saying including EDM in the case in chief tends towards an over quantification, I think that the quantification, the which is higher of these foreclosure effects, or which is more damaging, which is heavier if you will, the potential foreclosure effects versus the cost savings to consumers from EDM — that quantification I think really ought to be saved until the third stage of the Baker Hughes burden shifting. And I think it’s proper for the FTC to just mention the concept that will become part of that case without taking it into it’s case in chief.
And then speaking a little bit on the agreements — the sort of preventative options contract, I think that in AT&T/Time Warner we made a motion in limine that the equivalent contracts shouldn’t even be a part of the case for liability and should only be considered as part of a remedies case if liability or if a violation of Section 7 was found. And I think that’s — the Court didn’t agree with that. The Court did not grant the motion in limine. But I remain convinced that that’s the right analytical way to approach these types of preventative options contracts that are offered. It really is a stop gap measure, and the question is just is that stop gap measure effective enough.
And that’s the question that the Court should be considering after it’s decided that there’s likely to be harm from the merger. I mentioned at my very outset that one case that might be interesting for the audience to do a comparison with is the Steves vs. JELD-WEN case, which was a privately litigated case about — a vertical merger case that was litigated — I think it started six years after the consummation of the merger. And that’s an instance where the plaintiff was a recipient of one of these allegedly preventative supply contracts that was going to protect it from any harms or disruption in supply once the firm vertically integrated and had an incentive to only supply itself with door skins or raise rivals’ cost for door skins that rivals needed to make their doors.
You saw in that case — it’s an interesting case because it really gets to the heart of what we’re talking about with the ability to predict what sort of effects will come from a vertical merger in the future. You have these options or supply contracts in place, but there’s plenty that a vertically merged firm can do to skirt the spirit of that contract. And those were all the allegations that Steves put in their complaint to unwind the merger. And they were ultimately successful in that.
Now, that brings up the dovetailed point that maybe the best way to deal with mergers of this kind is to look and see what happens after consummation. But then, of course, you have the problem of unscrambling the eggs and businesses wanting certainty. So it really is — I really think that the best way to sort of split the baby in this case is to deal with options contracts as a remedial provision in the remedies part of a case if liability is established.
Elyse Dorsey: Yeah. Again, some really interesting points. We’re in the final couple minutes here. I want to make sure I give Tom and Steve a chance to kind of weigh back in. Any responses to the last couple of rounds or any final words for us?
Thomas Lambert: The final thing I would say to your question, Elyse, is yes, the mergers that I’ve been involved in — the merger investigations that I’ve been involved in over the last couple of years all seem to have — the staff is looking for a vertical issue. So where there might not have been a vertical issue seen in the past or the staff would have just quickly gone over that and wouldn’t have spent much time on it in the past, I think we’re at least seeing the staff incorporate vertical issues as part of their investigation on a much more frequent basis then we were a few years ago.
And I think the Illumina/Grail is sort of an outgrowth of that. It’s much more obvious here, but I think even in other cases where I think the dominant effect would be horizontal the staff is also often seeing vertical issues as well.
Steve Cernak: Yeah. I would just say I agree with lots of things that were just said. Michael, I too was surprised that Commissioners Wilson and Phillips signed off on this thing. So maybe they know something that I don’t. And Taylor, I totally agree that you can’t take a complaint as representative of the full analysis that the Commission has done. It’s obviously a litigation document.
I will say — and I’m also skeptical of the sort of contract solutions to the problems, but I did look at this one pretty closely. And it’s an awfully good contract. It’s subject to audit. It’s subject to baseball style arbitration, which the D.C. Circuit loved so much in the AT&T/Time Warner case. And given that the pro-competitive effects of this merger seem to be obvious and this looks like an awfully tight agreement, I think that this is a loser for the Commission.
Elyse Dorsey: Yeah, yeah. I think as everyone’s kind of made clear today it sounds like there might be something really compelling here, but the FTC certainly does seem to have its work cut out. So I think this will be a really interesting case to continue to watch. I want to thank you all so much for joining us. This was a really interesting discussion. You hit on so many different points. It’ll be really interesting, I think, to also see what the Court ultimately finds compelling. Nate, I think that’s all we’ve got for you today.
Nate Kaczmarek: Well, that was certainly a lot, and we certainly covered important ground. We’re grateful to the panel for a great conversation. Look forward to future opportunities to have the benefit of all of your expertise. Our audience, we welcome feedback by email at RTP@regproject.org. Thank you all for joining us today. Have a great day.
Bona Law PC
Director, Markets and Competition Policy
Professor of Law
Antonin Scalia Law School, George Mason University
Wall Chair in Corporate Law and Governance and Professor of Law
University of Missouri School of Law
Baker Botts LLP
Antonin Scalia Law School, George Mason University