Antitrust and Unfair Competition Law

Competition: Spring 2020, Vol 30, No. 1


By Karen E. Silverman1

The 29th Annual Golden State Institute was honored to host a fireside chat with Aaron Hoag, the U.S. Department of Justice Division Chief of the Technology and Financial Services Section, on civil antitrust enforcement and competition issues in the technology space.

MS. SILVERMAN: Thank you very much. Aaron and I worked together for a number of years on bar-related stuff and on casework, so it’s my very great pleasure to have the opportunity to talk to you today.

Aaron started the antitrust division in 1997 in the mergers task force. He was briefly in the front office and then moved over to, I will also call it "net tech," but it’s the technology and financial services section in 2003 and then rose to lead that section in 2016, and for any of you who have worked with Aaron wouldn’t be surprised to witness his rise to leadership in that area.

Over the time that Aaron has been working in the department, the prominence of technology and technology-related questions has only accelerated, and Aaron’s worked with it. So we’re going to have a conversation about that. We are at a moment where antitrust and technology is probably, for the first time in my long career, an interesting social subject. We get asked at every cocktail party that we go to sort of what’s going on. So we’re going to talk about what’s going on.

And before we talk about whether the antitrust rules and laws and enforcement practices are up to the task of technology, which they are, but I would like to ask you to address, is technology different from other industries in ways that are important to antitrust enforcement? How is technology different?

MR. HOAG: Well, first of all, let me just say thank you for having me here. It’s a wonderful time to be back in San Francisco. Thank you, Peter, for having me. We worked together very closely on a trial that we did out here in San Francisco. I see some of my colleagues from our San Francisco office here today too, which is wonderful. We worked with them very closely looking at technology matters.

I also have to add my standard caveat since I’m at a career manager level that these are my own personal views today, so views, my own opinions don’t necessarily affect all the policies and practices of the Department of Justice.

But let me turn to your question, which is kind of a great one and great stage-setting question. I personally tend not to think of technology cases as being in some kind of fundamentally different form of antitrust law than the rest of the antitrust law we practice, but I do think that there are some characteristics that we see in the technology markets that can sometimes come up in non-technology markets, but maybe are a little more frequent there.

[Page 90]

We see certainly as a very popular topic in antitrust circumstance the presence of direct and indirect network effects and sort of the relevance that that can have for the analysis in the antitrust front. There is a lot of focus on two-sided markets, multi-sided markets and platforms, and while you can obviously see those kinds of markets elsewhere in the industry, they are very prominent in the tech space. They do have an effect on the analysis. You see the kinds of entry barriers you see in tech are certainly different in a way than they are in other markets. You don’t talk about entry barriers in terms of how much did it cost to build this plan that makes this product and how many do you need to build and when you have the economies of scale. Because you’re not asking yourself, is it timely and likely sufficient that someone could take this plant to make this many widgets or steel or paper or whatever. You’re asking if someone can invent it, create this product, and get it out there in this marketplace.

So the kinds of questions and the way you approach those can be quite different. I also think when you see some characteristics, the level of dynamism in these markets can vary and vary widely in the tech markets. They have a reputation for being very technology innovative, and there are always new developments and new products and new things coming out into the marketplace, which can result in a lot of change, but in some markets you see maybe that even when there is a lot of innovation, you don’t necessarily see different players in the market share as well. Those are a few of the factors I think what makes it a bit different.

MS. SILVERMAN: What about the one we always say too, is the speed at which technology changes. That can lead us into the next question. Obviously, I would like us to anchor a little bit of where Aaron sees the ongoing debate about how the antitrust laws really map on to from an enforcement perspective, technology markets and the standard by which we typically evaluate competitiveness and largely transactions, but also conduct. So we can talk about the consumer welfare standard, but also just the pace at which technology happens, how does that impact your thinking and what kinds of standard you use?

MR. HOAG: I think it has a significant impact when you’re looking at any particular transaction, right? Because if you’re looking at a particular deal, and you hear the arguments coming from both directions. Both that antitrust law is too slow, therefore, can’t keep up with the pace of innovation, or that innovation is so fast—you used to hear this argument maybe a lot more than you do today, that there’s so much innovation, things are taken so quickly, there’s not even a need for antitrust enforcement. That’s perhaps today’s debate as it was back 20 years ago when I was in law school and people were just using Microsoft and the like.

MS. SILVERMAN: Right. Let’s talk about the consumer welfare standard and how you’re thinking about the sort of core principles, prime objectives of antitrust enforcement and how that maps onto tech questions.

MR. HOAG: Yeah, I think it continues to map very well. I mean it is a very lively topic of debate. You see folks pushing out there, pushing for saying the time for the consumer welfare standard has ended and that we should all move on to something new and better and greater things. But I see it as being actually quite adaptable and applying quite well to these technology cases.

[Page 91]

Obviously, you can certainly apply consumer welfare. When you’re looking at a traditional market, you’re talking about an increase in price and how it outweighs output, and all those things work perfectly fine, but I think equally well it works when you start talking about markets where there is maybe a zero-price product or markets where the important aspects of competition are actually product differentiation or are quality or we could see markets that could be privacy in becoming obviously much more important, both in the competitive spirit and antitrust.

And so I continue to be optimistic about our ability to capture all the different forms of competition within consumer law standards. I don’t think consumer welfare—I mean, it has kind of a reputation of being only about price. Everything is obsessed with maximizing the price and only capturing competitive harm when it happens in terms of price.

But I don’t think that’s the case. I think we see plenty of markets in technology, plenty of cases that we’ve worked on where innovation is more important than price and where innovation is more important in the element of product differentiation, and yet we still are able to apply all the adjusted forms, the sniff test, and other types of analysis that are I think sort of fundamentally the same and fundamentally rest on the same antitrust competition principles.

MS. SILVERMAN: And the role of—of sort of the traditional economics tools that we use to—I mean, we’ve all used prices and proxy for competitiveness and consumer welfare for years, but mostly because we can measure it, right? Are you starting to see new tools coming in to aid in a broader scope of welfare analysis?

MR. HOAG: I wouldn’t know if I could go as far as say "tools." It’s hard to come up with an economic model that’s around quality and variation, especially if you start talking about conduct cases. Maybe in a merger you can still look at diversion. You can still measure these things, and price can still be a proxy in such a way. But I think what you do see is a lot of creative and flexible thinking from our economists.

We have such a wonderful group of Ph.D. economists at the antitrust division that are embedded in every single team in every single case. They’re on top of all the latest literature and the ways of approaches and the ways of thinking about different types of competition and how you might prove or disprove in a particular case and what the particular facts are, would be necessary for that. So I think we see kind of an increased focus on that in most recent matters.

MS. SILVERMAN: It strikes me that documents, which are always important than becoming that much more important, where you’re trying to assess other dimensions.

MR. HOAG: I think that’s completely right, because there’s such a qualitative element to a lot of these antitrust cases. You’re trying to look at a particular course of conduct and find out is that conduct fundamentally restraining competition, promoting competition— especially if you’re looking at a conduct case, we always say that intent is an element in antitrust law and there is no doubt that that is true.

[Page 92]

And there are plenty of cases where you can sort of make your antitrust proof without any evidence of intent, but it’s not hard to say it’s probable though. I think you see that in documents time and time again, whether it be a merger case or a conduct case where you see companies that are sort of honest in their assessments of the pros and cons in a particular course of action. That I think you can see that directly having an impact on—if that appears to be consistent with the kind of evidence you find in the rest of the industry. Those documents can carry a lot of persuasive weight, and I think would be very relevant.

MS. SILVERMAN: But the one question—as you were listing down the other sort of dimensions of welfare, you mentioned privacy sort of in a parenthetical way.

MR. HOAG: Yeah.

MS. SILVERMAN: Can you tell us a little bit about how, if at all, you’re looking at data policy and privacy sort of policy discussions, of how you’re interacting with those, if at all, over the course of antitrust enforcement? At least in a lot of our work, we see these as not necessarily competing disciplines or tensions, but they are evolving and emerging in parallel and not always in conversation. And so I would love to get your thoughts on that.

MR. HOAG: As I think about both data and privacy, they kind of cut across all different areas that we look at in antitrust, and what I think of them as—antitrust is not about imposing a prescriptive view of what the privacy law should be in the United States. It’s about, obviously, taking scene of the benefits of competition so that competition can determine, to the best that it’s able, that the consumers get the kind of privacy protections that they are sort of—I don’t want to say entitled to, but entitled to in the commercial sense, not maybe in the moral prescriptive sense.

And so when we think about privacy, you look at it as kind of an aspect of that non-price competition that people can benefit from. Data is kind of a different one. It’s been important in antitrust for quite a while. I think some of the cases when we first started doing a lot with data, where data was the actual service that was being provided, many of those are kind of the simplest form of cases. People or a company, they license a particular kind of financial data and they’re merging with another company that licenses a similar kind of financial data, and so you do your kind of traditional analysis of the overlap between those two and how substitutable they are for each other, and doesn’t really dramatically change the nature of the antitrust enforcement. But I think where we’ve seen a lot more recently is where data is a critical input to providing some other kind of service. And I think that can fundamentally affect the antitrust analysis, not sort of the nature of the antitrust analysis, but when applied to a particular set of facts, I think can lead to some very interesting results.

I think back in my time in the division, ten years ago or so, we were looking at all agreements that related to the Yahoo search engine.

So at one point—you may have been involved with this, certainly some of your partners were—Yahoo had an agreement with Google where Google was going to take over to some degree, at much debate, while the Yahoo search—

MS. SILVERMAN: We’re not going to relitigate anything.

[Page 93]

MR. HOAG: I promise. I think there what you saw that it was clear that data was an absolutely essential input into operating a search engine, which I think is obvious to everyone, and that a loss of data or a diminution of data or a reduction in the amount of searches that Yahoo was handling raised some real serious questions for the division about how viable Yahoo would be going forward in the future. So you saw from that perspective, you can see where the importance of data and importance of data at scale was a critical factor in figuring out whether a merger agreement, a commercial agreement, was going to be procompetitive or anticompetitive.

But apparently when you looked at the outburst, you looked at—after that, we all fell apart, and I’m not revealing anything dramatic here. This was public at the time. We were prepared to sue to block that transaction.

Then Yahoo did a parallel kind of deal where they did it with Microsoft, right? There you have an operator who was, probably at the time, suboptimal in terms of scale, and their ability to obtain additional data from Yahoo, as I think we concluded and has been said publicly, sort of provided the potential for some real benefits to competition by giving them additional data, additional scale, and may have been more able to become a more effective competitor to Google.

So you can see those coming into play in particular transactions in different ways, but I think kind of on board with the fundamental antitrust principles.

MS. SILVERMAN: I guess even as recently, I think it might have been yesterday, when assistant attorney general Bill, who was testifying on the distinction between data usage and the value of that in a transaction that’s distinguished from sort of consumer data, which really more is the province of data policy as data policy is sort of more is classically thought of. Are you guys exploring those distinctions in the context of primarily transactions or conduct? How does it come out?

MR. HOAG: I think it could be relevant to any particular case. It’s very hard to say—to make general, broad statements about how data fits in, as you know, because what it really does is it comes up in a particular case, in particular a conduct or a particular merger you’re looking at. So you can definitely see, as the assistant attorney general was commenting yesterday, there are certain kinds of data that seem more replicable or more easily accessible to others.

It’s a little harder to imagine in the abstract why those would be likely to be entry barriers where there’s other more specific kinds of data that tied for a particular use of platform. You can see how they might get companies more durable competitive advantages.

Now, obviously, you don’t have to test that in any particular market, in any particular case, to see what is the real impact that data is having here. So it’s easy to make—everyone makes the general assertions about, we see people that come in. They need all this data or you need this much data, but I think what’s important is sort of putting real concrete aspects to it and really figuring out what’s the quality of the data, like how much do you really need? Can you actually quantify what the gap between the data you have and the data that might be at the next level if you weren’t denied access to this particular service or something like that? You could actually either quantitatively or qualitatively describe what difference that would make in terms of the firm’s ability to meet. I think those are the kinds of details that are necessary to make that really fit into a Folsom antitrust analysis in a particular matter.

[Page 94]

MS. SILVERMAN: Without going down the rabbit hole completely, non-horizontal mergers and killer acquisitions and sort of the non-traditional, forward-looking merger enforcement work that you guys are doing. I think in tech, in particular, a lot of the traditional efficiencies, defenses and arguments that we would raise or we’re beginning to suspect don’t have as much life in them as they may have historically.

Can you expand a little bit on sort of what’s your starting place when you think about these acquisitions and what you’re looking for by way of-—sort of looking for where the competitive interactions are in non-horizonal transactions?

MR. HOAG: I think the best way to think of it is, if you have a non-horizontal transaction, you’re almost invariably looking for, is there some form of market power that’s already preexisting that’s going to be reinforced in some way? So I think of that as the fundamental. I suppose one can come up with a stylized set of facts where you don’t have market power on either side in the vertical transaction and somehow get there at the end, but to me, that’s probably the rare case.

MS. SILVERMAN: It’s a different panel.

MR. HOAG: Yeah, that’s probably a different panel. I think it’s more likely that the kind of cases we would be concerned about that would cause us to look more closely are cases where you have a firm that has either market power or monopoly power. Could be either one is engaged in a transaction that, yes, it may be complementary in a certain way, but is there a discrete or logical chain of events that leads you to think that that actually could result in reinforcing that market power, not just by innovating and making a good product, but by actually withholding access from only people, for example, to the firm that is being bought? So it could be hard when those companies are nascent and are at the beginning stages.

Vertical cases are hard enough when you have well-established firms on both sides, as we certainly saw in one of our trials recently at the department. But I think in the nascent ones, you want to look for just real preexisting dominance or monopoly power and a real likely path for it to be reinforced by the merger; something that’s not overly speculative.

And even then, I think under Section 7, I think you have kind of a hard road. The law is very tight on potential competition or on nascent competition. In vertical cases, there’s probably no real law there, but to me, that isn’t something you want to look at very closely and we wouldn’t be interested in reviewing.

MS. SILVERMAN: But one of the conversations we’ve had in other contexts is whether there is meaningful daylight between your classic Section 7 standards and your incipient Section 2 type case, right? Where is your interest or shift from causally related to the merger to maybe more focused on merging market power? And I don’t know if you—are you guys thinking about those as divergent from a legal standard perspective these days or is that still—

[Page 95]

MR. HOAG: At the minimum, complementary, but I think you could definitely imagine a case where you’ve come to a different outcome, or a case you’re convinced should have been a different outcome, which is maybe or less the same thing. I think on the Section 7 talk side, you’ve always got to get over the probabilistic threshold, and where it’s not precisely defined whether it’s 30 percent or 50 percent or whatever, but still at some level you have to show more likely than not or very likely that there’s going to be a direct reduction of competition as a result of the merger.

If you look at the Section 2 case, my colleague, or acting chief economist Jeff Wilder gave a really interesting speech on this topic recently, and talked about how in the Section 2 case, if you start with monopoly power, then I think as we saw when you look back at the Microsoft case, for example, looking at particular kinds of conduct, you could extend that line of analysis, thinking towards acquisitions, and if there’s still going to require a coherent story, a coherent reason to think that this transaction is one that’s actually going to reinforce the monopoly position rather than just enable them to expand into a new area or engage in activities or growth. They are inquiring in a way they wouldn’t otherwise.

But I think there is some potential there, and I think I could definitely imagine a scenario where you end up with actually different answers. And it’s unproven, of course, but I think when you think about it from a theoretical point of view, it seems to capture a lot of what the concern about transactions is, which is really related to the enforcement of the existing monopoly power, more than necessary is the removal of this particular competitive entity from the marketplace because it’s so small.

MS. SILVERMAN: Still, I want to live long enough to see the first causation case— which hopefully.

MR. HOAG: Hopefully.

MS. SILVERMAN: One more question before I ask you what we can do for you or what you need to see and how you interact with the bar in ways that are useful—well, more useful than not. We’ve talked about sort of this other way to look at consumer welfare through the broader reach. Is it useful at all to also ask the other question, which is are there categories of conduct, interest, concern that just plainly are outside the purview of the antitrust? Because one of the things as we wander through the world balance that there is this tendency to try to answer all questions in the context of whether the antitrust laws could reach this, fix this, address this, make it worse, make it better. What should sit outside of antitrust and not in this room?

MR. HOAG: I think I tend to take a very broad view of what sits inside antitrust, which is probably not too surprising. But maybe the only limitation that I really apply to it is that it really needs to be competition-related. And that it has to be actually related to a theory that involves a direct loss of competition in some form or another.

So I think antitrust is fully capable, so we were talking a little bit about earlier, taking into account all types of non-price elements, of quality, of variety of products, of innovation, of privacy, of idea. I think those issues are all perfectly able to be tackled to some degree or another within antitrust, but there is always a temptation to see antitrust work reasonably well to see if you can solve all the world’s problems through antitrust.

[Page 96]

And as an antitrust lawyer, I tend not to think that’s the case. That doesn’t mean there are the competition problems that can’t be addressed through antitrust, but I leave it to those much higher in the food chain to what extent we need prescriptive policy to address the gap between antitrust and our vision of the world from a more prescriptive point of view.

MS. SILVERMAN: Which is important to be thinking about. All right. So things that keep you up, and we’ll leave time for some Q and A, so be thinking about your questions. But the things that sort of keep you up, most concern these days, and things from the division’s perspective under the private bar’s perspective, what is going right, what could go better?

MR. HOAG: I’d start to say half-jokingly like what could go better. It’s always better if everyone returned our calls when we called them for our investigation. I really strongly encourage people to do that. That could be quite helpful. It does surprise me how often we’re doing some merger investigation and we get the list of the top customers. We call 30 people, and 20 of them don’t call us back sometimes. And it’s really hard to figure out to determine quickly whether you have a transaction that—in trying to determine whether it’s likely to be anticompetitive or not, if you can’t get the responses. Even if the response is, oh, we don’t care about this at all because there’s 14 other competitors, that’s a perfectly fine 30-second response, even if you don’t want to sit down for an interview.

But I will say more seriously that we are always interested in hearing from people in the bar and from the companies in Silicon Valley and elsewhere about their markets. We invite people to come in and give us presentations and I’ll extend an open invitation for people that come in and think they have things we should know about as we’re looking at these markets. We are more than happy to hear a variety of views, whether you have a specific set of complaints about a particular company or whether you just want to come in and tell us about trends that you see across the industry. We’re actually pretty open. So we are always, both in our office in D.C. and for folks here, working with our colleagues in San Francisco, always happy to do that and happy to benefit from the knowledge that’s out there in the industry as a whole. So we welcome that.

MS. SILVERMAN: Great. With that, we have time for a couple of questions before we yield.

MR. ALIOTO: Enjoy it very much. My name is Joe Alioto, and what I would be interested in is your view as to whether or not any of the mega mergers have resulted in an undue concentration of economic power in most of the industries in the United States.

MR. HOAG: Well, this is where we get to the boring part of the presentation where I say that I unfortunately can’t comment on specific matters, sort of—

MR. ALIOTO: You personally; forget the DOJ.

MR. HOAG: Fair effort.

MS. SILVERMAN: Other questions?

[Page 97]

MR. HOAG: The only thing I will say is obviously, there’s been no reticence about the fact that both us and the FTC are taking a very broad look at all these technology markets. So to the extent that someone has a concern or a complaint of a particular transaction in the past has resulted in either an undue concentration today or then resulted in conduct resulting from that, that might violate the antitrust laws. We’re very open to hearing that, and we’re doing a lot of activity in these markets right now.

MS. SILVERMAN: So, interestingly, Aaron worked on the Microsoft case and decree. Do you want to talk briefly about—I mean, that sort of is now a historical starting place for how a lot of the market power analysis sees, and you have some thoughts of what the real takeaways are from that . . .

MR. HOAG: So I worked on the case after the trial was over and when the case had already gone up to the court of appeals and come back down and settled, and we were left with the remedy. There was a lot of debate and argument of whether the remedy was adequate, but I worked for, gosh, eight years or so on the actual enforcement of the decree, and I think what it leads me to—maybe a couple of quick conclusions.

One was the belief that these kinds of antitrust remedies, even though ultimately we obviously weren’t granted structural relief in that case, I think it did have a major impact on the way the industry developed. I mean, you can look back, I saw Bill Gates say last week that he thought that the antitrust investigation case is the reason why we aren’t all using Windows phones.

MS. SILVERMAN: Cost in the phone market.

MR. HOAG: Yeah, of course, what he meant was that we were distracting them unfairly and persecuting them. But I maybe take a little different view of it which is, if you look at the fundamental restraints that were put in place in that decree, I think that they prevented a lot of nefarious conduct that very well might have taken place. I don’t know if the industry would have developed the same way in web browsers and sort of the overarching use of the Internet in a generally neutral and open manner. If it had been the case, that Microsoft could have tied it all back in and continued to make it part of Internet Explorer in their own unique domain.

Also sort of led me to believe that you need to think very carefully about what some of these remedies look like. Because there are certainly some parts of it that were much more complicated and I think therefore were less effective, sort of overall, there were parts of the remedy that were both a challenge to implement, but also I think because of the challenge, almost never really had the effect that one might have hoped, certainly not as much as the direct conduct prohibitions that actually, I think, guided Microsoft’s conduct. I’d like to think at least a little bit opens up that industry, some competition.

MS. SILVERMAN: Other questions, comments, before we let Aaron go home? Thank you, guys, very, very much.


[Page 98]



1. Karen E. Silverman is a partner of Latham Watkins LLP. She advises on emerging legal and governance issues arising from the business applications of AI and related technologies, as well as the antitrust regulatory aspects of strategic M&A transactions.

Forgot Password

Enter the email associated with you account. You will then receive a link in your inbox to reset your password.

Personal Information

Select Section(s)

CLA Membership is $99 and includes one section. Additional sections are $99 each.