Scott Wallsten:
Hi, and welcome back to Two Think Minimum, The Technology Policy Institute’s podcast. It’s Friday, June 25th, 2021. I’m your host, Scott Wallsten, President and Senior Fellow at TPI. I’m joined by my cohosts TPI Senior Fellow, Sarah Oh and TPI President Emeritus and Senior Fellow Tom Leonard. Today, we are delighted to have with us economist, Michael Katz from UC Berkeley.
Michael is Professor Emeritus at the Haas School of Business & Department of Economics, where he was the Sarin Chair in Strategy and Leadership of the Institute for Business Innovation. He has also served as the Deputy Assistant Attorney General for Economic Analysis in the Antitrust Division of the US Justice Department from September 2001 through January 2003. He was the Chief Economist at the Federal Communications Commission from January 1994 through January 1996. He’s published extensively on the economics of network industries, intellectual property, telecommunications policy, and antitrust enforcement. Michael, thanks for joining us today.
Michael Katz:
Thanks. It’s a pleasure to be here, and if I could, I’d like to add one more thing about me. That is that I have worked and have ongoing engagements with several different, Big-Tech companies, media companies, telecom companies. So, I just wanted to disclose that. Well, you know, some of them are confidential, so I’m not going to say particular names. I just wanted your audience to be aware of that so they can be appropriately skeptical when they listen to me. You know my position is always: don’t believe what I’m saying because I say it; believe it because when you think about it, you see the arguments are correct.
Scott Wallsten:
That’s always good advice. Today we’re talking antitrust, which is a hot topic. We’re seeing the huge shift in how academics, policy makers, practitioners think about antitrust and what its goals should be.
President Biden just appointed Lina Khan as Chair of the Federal Trade Commission, and even a few years ago, she would have been considered to be fairly outside mainstream thinking, and yet here she is as Chair.
I’m not coming down on her specifically, but you know, what do you think is driving this change in how people are thinking about antitrust and are some kinds of reforms necessary?
Michael Katz:
So, I’m going to go start with the last bit. I think some reforms are necessary, and we can go back and talk about that. I mean, one thing, when you say changing the way people are thinking, it’s changing the way some people are thinking. I think there were a bunch of us who were sort of thinking the same way for a long time, and also still think that reforms are necessary. You get into the sociology or psychology of it, I’ve actually always been amazed at the extent to which people were just fawning over Big Tech and thought it could do no wrong. And I always saw some of the statements coming out of big tech—oh, and something I should add to my disclaimer, some of them have retained me. What I’m saying, it’s not necessarily something they will endorse—but it really was like in the heyday, when GM used to say, “What’s good for GM is good for America.” Big Tech was doing the same. And I think other people have said this, that part of the backlash—and I don’t think this is the case with Lina Khan—but with some of the backlash, I think it is people sort of fell for this, “Oh, we’re wonderful. Everything’s great,” and now they feel duped and are sort of going to the other extreme.
But look, I think the other big thing is people, part of it is they just look and they see how big and how successful these companies are, and that sort of captured a certain part of the public’s imagination. So, I think that’s where it’s become more of an issue politically than usual because you know, usually with the exception of very few antitrust cases, antitrust sort of proceeds in obscurity. So, you think of the Microsoft case, but not very many others… AT&T-Time Warner, supposedly was this huge public case, but I think that’s just because the media industry is interested in itself.
Anyway, but that’s it, well, let’s talk maybe more on substance. I think a lot of us feel that what’s basically happened is antitrust law in America is essentially common law. The fundamental statutes for antitrust are very short and very basic, and that’s actually been a strength because it’s made them flexible and enduring, but it means that, from the outset, Congress has relied on the courts to fill a lot in.
And there’s been this common law evolution. And I think what’s happened though, if you look at the last, I’m not sure of the right number of years, maybe 40, 50 or so, the courts have really moved to the right., And it’s often called the Chicago School and say they’ve adopted it. But I think that the courts are out of line with mainstream economics and themes. I also think they’re out of line with a lot of populist thinking. And so now there’s pressure for Congress to act. And I think some sort of legislation is needed, and I think it’s appropriate. In fact, there are Supreme Court decisions in which the court says, “Look, here’s how we’re interpreting the law,” and if Congress doesn’t like it, Congress should step in. And I think we’re at a time where that needs to be done because, as I said, I think the courts have just gone off in the wrong direction.
Scott Wallsten:
You’re …saying the courts have gone too far to the right, or have a Chicago school view, but then there’s …a populist backlash against that that has gone too far in the other direction, and so what is the middle area that you’re trying to go for? Because it seems if I’m hearing you right, you think kind of both sides have gone too far in the wrong directions or they’ve gone too far in the opposite directions, and that’s left us without a middle.
Michael Katz:
Yeah. I can’t remember the names, but remember it was a late-night host who was interviewing a basketball player and talking about how one of the basketball player’s colleagues had made a rap—what was then a CD—and another one was doing something else he wasn’t really qualified to do, and then the host turned to his guests and said, “Well, what are you going to do that’s stupid?” And I sort of feel like the left have to say “it’s our turn to be wrong.” Now look, I really despair of this actually, because the middle ground I think is—it is always said, antitrust is really fact specific and you sort of really dig in case by case, and it’s very difficult to do that—and I do worry that this sort of middle ground just may not be feasible. I mean, what you have instead, you know, I think we can see the extremes of left and right. People are saying, “Well, here’s the presumption, and I’m either going to presume that all vertical integration is bad, or I’m going to presume all vertical integration is good.”
Neither one of those views is correct, but at the same time, if you said to me, “Okay, how can you tell good vertical integration from bad if you’re going to be in the middle?” It is extremely difficult, and there is a question whether the courts are up to that.
Tom Lenard:
Let me just ask you to expand a little on your statement that, because I think this is part of what we’re discussing on your statement that the courts have departed from mainstream economic thinking. Do you want to give a few examples of that or put a little… a few more specifics?
Michael Katz:
Alright, well some of the specifics, I mean, you mentioned the views that are in Ohio v. American Express is one where the courts have declared you have to do market definition if it involves vertical practices, and more generally the courts, have, or have come close to adopting, the view that all vertical practices are good. And I think the economics just clearly doesn’t support that. I’m not of the view that all vertical practices are bad. And maybe, in fact, the majority of time you see it, they’re good. But I think they just built-in way too strong a presumption: (a) that the practices are good, and then (b) this necessity of market definition and particular ways of doing it… The thing on market definition I’m not sure is actually a left versus right issue. I think maybe that’s economist versus the rest of the world.
I do think we’ve gotten to the point where the formalities of market definition are actually in many cases an obstacle to sound analysis, not an aid. Again, I believe you have to do some sort of market definition in the sense you need to know who the competitors are, and you want to understand how competition works in a market, but I think formal market delineation is sort of out of control, and I think the courts… this is a long-standing issue, so, we continue to put too much weight on it, but I think the Supreme Court in Amex, with their declaration that you have to define a two-sided market and then what that means in terms of assessing harm and how competition works, I don’t think that’s supported by the economic literature on two-sided markets. I think they’ve gone way beyond that.
I also think the Brooke Group tests for predation, if you look at it, don’t really make sense. I will say I criticize everyone, including myself, and you have problems on the other side, but as well, but you know, say on the predation, they say, “Well, how are we going to determine if something’s predatory pricing?” Well, one of the things we’re going to ask as well, what was there below-cost pricing and then was there some reason to believe that the firms would be able to recoup it?
Well, first of all, just logically that’s a silly way to test things because presumably if you have an economically rational firm, for whatever reason it prices below costs, it expects to be able to recoup it. So, that’s not a test that in any way distinguishes whether it was predation designed to harm competition and destroy competitors, or whether it was just a benign activity that was to try to build up a user base because user bases can be very valuable in a market with network effects. So, I think part of it is, the test itself just doesn’t… actually, there’s no economic logic underlying it. And then where they’ve also gone beyond the consensus is to basically say, make an empirical claim, which, “Oh, you’d never see predation. There’s no such thing and it can’t happen or almost never happened.” And I don’t think the court has any basis for saying that.
Tom Lenard:
What would you say would be a rational predatory pricing policy or approach to predatory pricing?
Michael Katz:
So, and this is the thing about being in this difficult ground stuck in the middle, I think, and I might have my PhD taken away—but fortunately it’s a D. Phil so they can’t take away my PhD— for saying this, but I think it may be at some point you do have to dig into intent, and you know, economists don’t like that, and why I say that is a lot of the markets we’re interested in, right, we’re interested in markets that are oligopolies. So, they’re going to tend to be markets with big economies of scale, and they’re often going to be things like network effects or data effects or whatever, and the problem you get into in those markets is just the fundamental problem: in a market like that, it can certainly make sense to charge below-cost prices, whatever your measure of cost, at least initially, to try to build up scale, build up your network, build up your data, whatever, and that would be true even if that had no effect at all on your competitors. So, I don’t think you would want to label that as anti-competitive; that’s called investment.
The problem is, if what you’re investing in is a user base, almost certainly investing in your user base and making it larger, it means you’re also making your rival’s user base smaller. It doesn’t have to be that, but that’s, I think in the vast majority of cases, that’s what we’re going to see. And so now we’re at this fundamental tension of saying, “Well, on the one hand, there’s strengthening yourself and helping consumers in a good way, but on the other hand you’re weakening your rival.” And I think it’s extremely hard, in fact I think impossible—there certainly isn’t any sort of cost-based rule, and Joe Farrell and I have a paper trying to look at that. I think that’s a messy position, but you really have to dig into the specifics and try to figure out what the dominant effects are. And the reason I say, “look at intent,” is partly because I’m worried about what happens when a firm is just trying to compete, and the next thing they know it gets slapped with treble damages for predation, and does that chill competition? So, I think you’d need to do something to try to test for that, but as I say, looking at costs alone is certainly not going to do it.
Scott Wallsten:
It’s interesting that you talk about market definition and the test for predation kind of one after the other, and …there is a logic [to the]…market definition by SSNIP test. But you’re saying that that’s become an obstacle maybe because people are trying to slice it too finely or [inaudible]. I’m not sure, but then the problem with the predation test is that it has no logic. And so, it seems like both sides of this have led to not good outcomes.
Michael Katz:
Well okay. But let’s take the SSNIP test and ask what the logic is. So, it makes some sense if you start with the premise, you say, “Okay, I have to draw something at zero-one boundaries, right? You’re either in the market or you’re not. It may well be that the SSNIP test is the best we can do. But the problem is you started with something that fundamentally doesn’t make sense in many markets. And, look, where I say can it be an obstacle, and I think it’s in some ways one of the most painful cases because it was just laid out so clearly, was when DOJ tried to block the PeopleSoft/Oracle merger. What the judge did in his opinion is he explained very carefully and clearly and intelligently why it’s essentially impossible to draw bright-line boundaries when you have differentiated products. But then, rather than conclude “Okay, so we shouldn’t require the plaintiffs to do that because it can’t be done, and surely that doesn’t mean everything’s fine,” and he said “Nah, that means the government loses.” You have a burden that almost by definition can’t be met, and I think that’s a ridiculous way to decide a case. And you know, whether or not he’s right as a matter of law, I don’t know. I mean, I have my own doubts about why he did what he did, but holding that aside, if you say, “Well, he’s right, that’s what the law compels,” then we should change the law, which actually reminds me of another one I’ll say was sort of equally ridiculous or maybe more so.
One of the interpretations of the Amex case and what the Supreme Court said is that a two-sided transaction platform only competes against another two-sided transaction platform–they can’t compete against say any sort of one-sided competitor. And then the most extreme version of that we’ve seen, I think it was Saber v. Fair Logic, the judge said, “Look, both sides of the case agree with the party, these two entities compete. Everybody agrees they do, plaintiffs and defendants. But also it’s true one of them is a two-sided transactions platform, and the other is not a two-sided transaction platform. Therefore, as a matter of law, they don’t compete.” I mean that’s just ridiculous. I mean I don’t see how anyone can defend that.
Scott Wallsten:
And do you think that there’s been too much emphasis on [the] two-sided market as something special? Because … in some ways everything is a two-sided market. A grocery store is a two-sided market. They buy their supplies from the farmers, not the farmers, from a supplier, and then they’re a platform for the people to come buy it. How do those differ from the other things that people are now saying are special because of two-sided markets? I mean, there are things that are different, but how important is it that you take those things into account?
Michael Katz:
As an academic pursuit, two-sided markets are a useful way to think. I mean, it’s a terrible name. I went to Toulouse very early on and said, “I just want to let you know my 12 year old daughter thinks it’s a terrible name, so that should tell you something.”
But no, it is a problem. So I think, yes, they’ve identified some important forces and then you go back and look, it’s like the early articles by Rochet and Tirole, they actually say, “Look, a lot of these issues are out there but here’s a particularly interesting combination of these issues.” They don’t claim this is something never before seen. And they did good work. And what’s happened is people just grabbed on to two-sided markets and went, “Oh yes, it’s fundamentally different. Let’s rethink everything,” and they’ve done it in a lot of ways that I think don’t make sense. And so, yeah, I think way too much has been made of it as being just completely different. Again, so I think it’s a useful lens in certain ways to worry about the feedback effects across the different sides of the platform. But to your point about isn’t everything a two-sided market, I mean, that’s a big problem.
Again, if something like with the Supreme Court has done where they said if you’re a two-sided transaction platform, you’re treated very differently than somebody that’s not, then a question is, “Well, what’s a two-sided transactions platform?” And the Supreme Court itself seems to have no idea. So, in Amex, it said “American Express, you’re a two-sided platform because what do you do, you help merchants and consumers get together and you get paid only when they transact,” which actually is false, but the Supreme Court said lot of things in that case they made up as fact, but predominantly Amex gets paid a percentage of the transaction value when it occurs. Okay. So, then along came a case, I can’t remember if it’s Pepper v. Apple or Apple v. Pepper, a recent Supreme Court case having to do with standing and such, and the Court looked at that and neither the majority nor the dissent had any recognition that the App Store looks just like American Express at a high-level description.
Because what does the Apple App Store do? Well, they collect a percentage of the value of a transaction that happens between a seller and a buyer. So, it’s hard for me, it’s impossible for me to conceive of a world where Amex is a transactions platform for the App Store is not. Yet, if you look at what the Court then said in Apple v Pepper, it’s all completely inconsistent with their view of how two-sided markets work.
So, just to give an example, in Amex, one of the excuses they use for saying the plaintiff’s lost—and I should disclose for those who don’t know, I was the DOJ’s economic expert in Amex—but in that case, they said, “Look, you know, there’s this fundamental failing because you didn’t look at the two-sided price. You only looked at what happened to the price in merchants.” That also is a false statement by the Supreme Court. We did look at the two-sided price, and there’s evidence that the two-sided price went up, and Justice Breyer talks about that in his dissent.
But in any case, they said, “Look, it’s critical, you have to look at the two-sided price because one side going up could be the other side going down; you’ve got to look at the net effects. You just can’t look at one side in isolation.” And yet in Apple v. Pepper, there actually is a statement—I can’t remember if it’s the majority or the dissent—that says, “Well, look, consumers will have their theory of harm, and sure, app developers may have another theory, but they’re completely unrelated. You know, they’re not coming to a common tool of funds or anything. They’re just two separate things.” Well, that’s just completely inconsistent with what they said in Amex where they said there’s essentially only one price and only one common fund and nothing else.
So, I think this line drawing and this attempt to use labels has actually just made a mess of things, and it’s not even one where it’s sort of left versus right, because I think plenty of people now on the right have said, “Well, wait a minute. Things like excluding competitors because they’re not two-sided markets. That’s a problem.”
Scott Wallsten:
And this is one of the things where there does not seem to be a right-left divide, particularly maybe there are different reasons for what they want, but they want the same thing. If you were Lina Khan and you are head of the FTC, what would this mean for what you do now? Does this mean you try to bring cases or enforcement against companies and try to use different standards? Or is it … at a higher level, you’re still trying to change how we think about the questions?
Michael Katz:
So, a different version of your question, is if people said, “So, you know, what difference is it going to make that Lina is the Chair?”
Scott Wallsten:
Yeah.
Michael Katz:
I think the answer may be not that much of a difference unless we have legislation, because I think a lot of the things that she would like to do, as I understand it, with limited exposure to her writings, they’re not going to get through the courts. And I think a lot of people’s view on a bunch of these big cases against big tech is ultimately, they’re going to fizzle by the time they get to appeal because of how the courts are. And particularly since I’m involved in some of them, I won’t comment on the merits, but I think that assessment probably again is shared by people on the left and right, the difference would be whether you think that’s a good thing or a bad thing, but I think everyone would agree it’s going to be hard-pressed getting through the courts. Now, if you asked me, what would I do as head of the FTC? I think one thing I would do, since it looks like they will get more resources, is to try to do more… economists always called for more retrospective studies, more industry studies. Do more of the things that the Federal Trade Commission is empowered to do that the Department of Justice is not.
But in terms of bringing cases, a lot of people said, “Oh yeah, the agencies have been scared away and they don’t bring cases they otherwise would.” But a bunch of it is just even if you brought the case, it wouldn’t make any difference.
Sarah Oh:
For legislation, what do you think about bright-line rules? Like some people are advocating for thresholds for market share or lower Hart-Scott-Rodino acquisition threshold. That’s one way to legislate bright-line rules, but who knows what the line is? What kind of legislation would be a middle ground without being overly burdensome?
Michael Katz:
So you mentioned a couple of different kinds of thresholds. The one about lowering the Hart-Scott-Rodino thresholds, that I would be in favor of. Just because, fine, let’s give the agencies a chance to look at more things, particularly given all the concern about acquisitions of potential or nascent competitors. But, you know, in the case of, “we want to have more bright-line thresholds,” say for mergers, actually, I’m against that.
Again, it’s a case I was in—the Sprint-T-Mobile merger—and I testified as a witness retained by T-Mobile, and a colleague and friend and coauthor Carl Shapiro, was retained by the states. He wanted—the plaintiffs in that case wanted—to rely on a presumption based on concentration. And I said, but I don’t think you have any basis for that in this market and we put in some econometrics that looked at concentration differences in local markets and said, “Look, it doesn’t change the behavior that you guys claim it does.” I’ve talked to Carl about this subsequently. He says, “Well, but you know, there are all these problems econometrically,” which I agree with, that there are difficulties in it. But my comeback is, “Okay, but you don’t even have any bad econometrics. You have nothing other than to say that mine aren’t good enough.” Now I may be setting myself up for the second half of this podcast, by the way, but I think this is, as I say, where I struggle because I worry a lot about the thresholds because I just don’t think we know enough to know where to set them.
That said, I think it would be useful to have legislation to say to the courts, “Well, you know, you’ve set thresholds that are too high or too low, depending on what sort of things you’re looking for.” I guess I want there to be a presumption against presumptions, perhaps. Stop saying you know in advance what the right answer is. And then have to sort of dig in. And, then if you dig in, it’s dig into what? And this is where what I’m proposing falls apart potentially, because at the agency level, they can do extremely sophisticated analysis. They have the time, the resources, the training, the people, but a court just doesn’t have those.
And so that’s why I will admit I’m sort of at a loss about what to do other than trying to undo the bad presumptions, such as the one that any vertical practice must be good. I haven’t seen a lot of really good presumptions other than when there’s sort of this negative.
Scott Wallsten:
[W]hen we were talking earlier, you mentioned concerns about using antitrust as regulation, and that sounds like what you were saying now. I mean, it’s hard because courts don’t have the kind of resources and expertise necessary to ever do anything like that, unless I’m misinterpreting what you were saying. So, I don’t want to put words in your mouth, unless they’re funny.Michael Katz:
Well, I would say the problem is we’ve said to the courts, “Okay, we want you to engage in this really fact-intensive exercise and weigh all these things.” I guess what I’d say as part of this is I think courts should have much greater reliance on economic experts who are retained by the court to try to help increase the expertise. Even with court-appointed experts, I think there are problems when you get to economic analysis that’s so sophisticated that only an expert has any chance, an expert PhD economist, has any chance of understanding it because then we’ll get all these fights about, well, is this the right expert for the court, et cetera.
Scott Wallsten:
[D]id you say experts retained by the court?Michael Katz:
Yes.
Scott Wallsten:
So, that’s very different from the way things normally happen.
Michael Katz:
Yes, I mean, there are cases where judges have done it, and I’ve talked to a few, and they were quite pleased with it. Now I think it’ll help cut through some of this stuff. And as I say, there are certain things where it may be so sophisticated, it’s just a problem because then people will start saying, “Well, whatever the choice, the court’s choice of expert is going to predetermine the outcome because it’s all a judgment call.” And there are always going to be things like that. But I would hope that, if the court retained an expert, the expert would tell it things like the fact that prices have been falling over time does not prove the market is competitive. That’s just bad economics at the freshman-economics level. I bring that example up because that’s another thing where… the Supreme Court was vague about it in the Amex case, but the Second Circuit, which was the court that first heard the appeal of that case said things like, “Well, output has gone up since the 1950s.” I think they said, “Prices have gone down. So, that shows that whatever American Express is doing must be competitive.” That’s just complete nonsense as a matter of economics. And I would hope at least on things like that, that any economic expert the court may have could explain why that’s wrong. I mean, you might hope that that would come out in the adversarial process and what goes on in the courtroom. But…
But can I just say one thing when I talk about antitrust? There’s this thing about what you know, is characterized as a Chicago school, even though people in Chicago aren’t like that anymore.
There’s the courts’ being in sort of the old Chicago school and then versus progressive thinking about it. But there’s also this thing about what’s the difference between more middle-of-the-road people, which I would put myself as, or maybe it’s traditionalist.
And I think, you know, part of it comes down to, I think Lina Khan and I both would say that antitrust is about protecting competition. And I always say, when people say “like does the US have a consumer welfare standard or a total surplus standard?”, I say that debate is missing the fact that it’s not illegal in the US to do things that harm consumers. What’s illegal is certain forms of conduct that are viewed as not being competition on the merits or are viewed as harming competition. So, we have this consumer welfare overlay coupled with this harm to competition. So, I think she and I, and people like her and people like me, would say fine, it’s about protecting competition. But what we mean by that, I think turns out to be really different. And I think certainly in some cases I would characterize the progressives’ views more as protecting specific competitors, and I would say, in fact, that does risk killing competition.
But I think a huge source of debate though, is this thing that we have collectively—lawyers, politicians, economists—have never really defined what we mean by “competition on the merits,” yet it plays a critical, critical role when we assess things, but we’ve never been able to define what it means. And I think that’s a big part of what we’re seeing now is that people say, “Well, what competition means is we’ve got to protect little guys and have a lot of firms,” and other people say, “No, no competition is sort of this, not a free for all, but you know, you get to use the advantages you have and there are just a some set of restrictions on what counts as going too far.” I think it’s important because we need to be clear about what it is we’re debating. I think that is what a bunch of us are debating: it’s what constitutes competition.
Scott Wallsten:
It seems like there’s so many parts of this where it’s almost impossible to have answers or guidelines. I mean, the question of acquisitions, either acquisitions are a great thing because that’s why people, that’s why people enter the market and become entrepreneurs because they want to get bought out, or it’s preventing competition or, you know, it’s blocking a future competitor by buying them out. And sometimes it’s the one and sometimes it’s the other, and they can both be true. How do you ever decide how to take that on? Not, that’s not the question. Is there a way to think about it?
Michael Katz:
I think the problem in some way, I was going to suggest a way, except I’ve already undermined it. It’s even harder. Because you could try to distinguish whether the firms is… There are a lot of different reasons, I think for big-tech acquisitions, because those are the ones who’ve gotten a lot of attention.
So, in a bunch of it, I think you really do want to keep having merger because what it is, it’s a way to reward people for pieces of, you know, complementary intellectual property. They’re not developing… these are not firms that are going to go on to be competitors, okay. And on some level, it’s a form of exclusive licensing, and I think you do want to allow those mergers. So, then you’d say, “Okay, so why wouldn’t I have a rule that says that it’s just, you know, this little piece of something, of course it’s like any other input, but what I’m worried about is when you’re a firm that looks like you actually could grow to be a rival.” Now, I think there’s merit to that, except the problem you’re going to run into is then when firms are deciding what to do, they may say to themselves, “Look, the odds of being a successful rival are really low and there’s a much higher payoff to being bought up, and so I better make sure I don’t look like a firm that could become a rival.” And that you could actually start shaping what you do to make yourself a more desirable acquisition target, not in the maybe traditional sense of, “Oh, I want to be the kind of firm that looks attractive to buy because I can generate returns. I want to be a kind of firm that is attractive to buy because the deal will get through the merger authority,” and so there’s a problem there. But I do think, ultimately though, we are going to have to try to do something like that. Say, look, that there are a bunch of acquisitions where it’s fine, it’s how you get IP, and there are other ones where it’s not fine because these really are firms that have the potential to be competitors. And that’s really hard, but that’s an area where I think we need specific reform.
And I think there’s a fairly broad consensus on this that we do need to make it easier to challenge mergers based on potential competition. In thinking about that, for example—and this was proposed years and years ago, I’m going to forget by whom—but this whole thing of looking at adjacent markets, you shouldn’t just ask, is the acquisition target in the current market, but are they in the adjacent market and then you have to have criteria defining adjacency. But I think that that’s something that I’m not sure we define in terms of markets; I think we might do it more as is this firm accumulating assets in another market that turn out to be transferrable to the one in question.
Scott Wallsten:
But I mean, that seems, that seems doable when the firms … being acquired are of some decent size, but Instagram was just a few people, right? And Facebook’s acquisition of Instagram either blocked a competitor or it made Instagram the success that it is — one or the other. It was either great or it was terrible. Back in that time, how would you have applied this standard to that?
Michael Katz:
So I think this is something people generally criticized and said, “Look, there’s zillions of potential competitors so how are you going to pick which ones count.” I think one of the things is you would look for the growth trajectory. I’m not going to say growth rate, because if you have one person and you go to 10 that’s a big rate. So you’d look at the growth trajectory and see: is there a reason to think that this firm is starting to take off and that you might be able to tell—you might get it wrong—but even when they’re very small. And I think that there are people who would say that was the case with Instagram.
I mean, there’s a reason that Facebook was worried about them. And my sense is that actually the firms in the industry can do more to tell who really is a threat fairly early on than a lot of people give them credit for. Because they do look for ones that are starting to get the feedback—the cycle going, or to get the expectations going their way. Because in these markets with network effects, that’s going to be critical in these markets where consumers are making purchase decisions by being forward-looking and asking, “What’s hot? What’s going to grow?” That’s the sort of thing we’d look at.
But also, and this is where we’d have to change the standards. But if you went back, and you did that and you said, “Oh, you know, now we appreciate this is more important than the FTC did back then.” I don’t think it would make any difference to what you would do if you were the FTC under those rules. I can’t say what the name of it was, but I was involved in something as consultant with the FTC looking at a merger, and we had reasons to be concerned. It had to do with taking something online, but the biggest firms at the time were all bricks and mortar. And we said, “There’s just no way we’re going to be able to win claiming there’s a separate online market even though we’re worried about this.” And to this day, I think we’re absolutely right, and it would have been a waste of public’s money to try to bring a case. So, I think we need to have a change in the standards that’s tilted towards the enforcers in this. And yes, there are going to be some errors, but they’re already errors in the other way. It’s a long winded way of saying, I think there are things we can do. They’re all highly imperfect, but you’ve got to decide how you balance type-one and type-two errors, and I think we’ve gone too far in one direction, particularly on those. I mean, partially because it seems like that’s likely the only way we’re going to see … Step back. If you guys say what we’re looking at is a world of Schumpeterian competition, and therefore we shouldn’t be worried about seeing really high market shares of incumbents today because they could be swept away by someone with a trivial share, then we need to be worried about mergers that acquire people with trivial shares because they’re the ones who could take it away.
Some people try to use the Schumpeterian view of competition as a way to say, “Well, antitrust doesn’t really have to worry about market power and let’s stay out of it.” I think it’s actually the opposite when it comes to things like dealing with nascent competitors. The Schumpeterian view tells you, you’ve got to be more cautious about harm to nascent competitors and more aggressive with antitrust law.
Scott Wallsten:
But then at the end of the day, are you still pessimistic that these will happen given the nature of the courts?
Michael Katz:
I don’t think it will happen without legislation. And what I’m pessimistic about with legislation is I think this focus on Big-Tech legislation as opposed to sort of generic legislation that applies across the economy. I think that’s a mistake. I understand, both politically, but also even economically, why people will say, “Well, wait a minute, but these firms, they’re so quantitatively different, we feel they’re qualitatively different.” I understand why they’re doing that, but ultimately, I think it’s a mistake. I think we need reform, but it should be economy-wide reform.
Tom Lenard:
I mean, one of the changes and this is, I don’t think this is necessarily something economists have a lot to say about, but it’s changing the burden of proof. It’s not the government doesn’t have to prove it’s anti-competitive, that the merging parties have to prove it’s pro-competitive. Are you in favor of something like that?
Michael Katz:
So, I think the burdens are screwed up, and I actually think economists should have something to say about it. But I think the… so I can say a couple of things. I do think we’ve gotten… we’re in this state where it is a problem that a lot of the economics of it—the analysis—is sufficiently complicated, and a lot of these issues are too; so whoever has the burden is going to lose, and that’s a problem.
That shouldn’t be. Because that’s basically down as in, “Fine we should just have some simple rule and who needs a court, we’re going to have such strong presumptions. Off we go or something.” Part of the problem is I’m not sure what the real standards are, and I’ll say the following. On the one hand, I think it appears like we simultaneously have it that it’s too hard to establish the prima facie case of saying that a merger is going to harm competition.
But I also think it’s too hard to establish efficiencies. It’s incredible. I mean, some people have said that they think that the Sprint/T-Mobile case I mentioned before is the only case where substantial efficiencies, actually the claim of that, is what drove the outcome. Now, it’s really hard to say because I think part of it is the judges are worried about appeal and such. They tend to like pick which side wins and then say the other side is completely wrong. And we see this in a lot of things where, like with Section II, there’s supposed to be a balancing under the rule of reason, but essentially no cases (when you get to the decisions) talk about balancing. I mean, there were a few, but it’s like a handful. But I think a lot of us suspect actually judges do the balancing, but then once they’ve decided which side should prevail, they then sort of just pile on to say that side was right, and the other side didn’t present evidence.
So, as I said, I think it’s sort of messed up with both sides. So, I guess I would be happy to say, if you want to merge, you have to show that it’s good, but then we need to make it a lot easier… or not easier, but I think we need to take more seriously the efficiency claims. And I’ll also say, I’m talking about the agencies, meaning the FTC and DOJ. But when you get to the FCC, it’s just a routine thing for them in a merger they don’t like to say, “Well, we couldn’t judge the efficiencies with certainty. Therefore, they get zero weight.” And again, that’s bad economics, bad decision theory, whatever.
Scott Wallsten:
Is there a system that you can point to that you think does a better job than we do? Whoever has the burden of proof has the harder job obviously. The FCC has to approve mergers, but the DOJ and FTC can challenge them. So, you’ve kind of got the two different burdens of proof right there, but one is not inherently better than the other and they can both get things wrong. But is there some system of country who’s gotten this better than we have?
Michael Katz:
I don’t know that anybody has. I don’t know that much about other regimes, but I think that as things go, the US does a pretty good job. And that’s why…
Tom Lenard:
So, you don’t think the Europeans, to put it specifically, do a better job?
Michael Katz:
I don’t. I mean, it’s interesting because when I was at DOJ back in 2001, the DOJ was sort of actively fighting against the European Commission, and in fact, when I was in the Antitrust Division, I went over to Europe a few times because I was sort of viewed as the person in the middle who might help bridge the gap.
And now the US is sort of moving much more towards them. I think lots of the US system is pretty good. As I said, I think the big problem is that the courts have sort of gone off too far in one direction. As I said, I think it’s time for Congress to try to reset it, push things to the middle, and that is messy, the way the US does it. But yeah, I don’t know of anything that’s better.
Tom Lenard:
Well, I’ve been looking at all the legislative proposals. That’s not obvious to me that any of those are better.
Michael Katz:
Let me just say the legislative proposals that have come out. The Big-Tech ones…
Tom Lenard:
Well, they’re not all Big Tech. Klobuchar’s is generally applicable and some of them are you know..
Michael Katz:
I mean, I haven’t looked at Klobuchar’s in a while, but I think that’s much more the way to go though than the Big-Tech ones. I can’t remember. I’m sure there’s a bunch of things in Klobuchar’s I disagree with, but as I recall some of this resetting, I think about things and trying to undo bad presumptions. And I think that, as you know, I’m in favor of that program.
Tom Lenard:
Then there’s of course, the burden of proof. I’m not sure, I can’t remember for which class of cases, but for a lot of cases.
Michael Katz:
Yeah, I worry about that. I mean, part of this thing, let’s talk about mergers and I’ll bring up another one I was involved in with AT&T-Time Warner, which did get a huge amount of attention, I think mainly because of Time Warner, both the media liked to cover themselves and then this whole issue with, wow, this is pretty wild—a conservative, Republican administration going after a vertical merger. How did that happen? Does that have anything to do with CNN?
But you know, in the end, they’re undoing the merger, and AT&T says it wasn’t a mistake. Maybe it was. But I think what’s turned out to be the case with that is that the vertical synergies the parties claimed turned out not to be as big as they thought. And then all the harms the government claimed, “Oh, you’re going to just be raking it in. You’re going to have so much market power,” that was also false. And look, I think a lot of mergers probably don’t do what the parties thought they were going to do and they’re less beneficial to them, but I’m also a big believer that we ought to let people make mistakes. And I actually think that’s a lot of what we see in merger activity. It’s not that they’re particularly anticompetitive. It may be that a lot of them don’t generate efficiencies. They may be empire building, but I think we should just let people go out and make those mistakes. That’s not what antitrust is trying to stop.
Sarah Oh:
One thing the new FTC or DOJ antitrust chief can do is perform or rewrite the Horizontal and Vertical Merger Guidelines. I mean, that’s a guidance document that is out of the executive branch. What do you think of that?
Michael Katz:
So, I don’t know. I mean, go back and think about 2010. You know, the huge thing was moving away from saying how important market definition was and doing some sort of upward pricing pressure. And then the DOJ, FTC, if they’re going to be in court, they’re going to be arguing about market shares and concentration in relevant markets. So, I’m sort of torn.
I mean, on the one hand, the Merger Guidelines, and you know, the rest of the world looks to them too. I mean, they’re good. They spell a lot of stuff out, and I think they’re very useful for helping you think about how the agencies think about things. But I think we still have this problem that how they’re translating into what the courts do. Now, they clearly do have a big impact on the courts. Now a lot of courts are looking at the hypothetical monopoly test and things like that. And, in fact, I’ve complained sometimes publicly that the courts seem to be under the misapprehension that the Merger Guidelines are statute, and the courts give them too much weight. But I don’t know, I think the Merger Guidelines, they do have an effect and with a lag, but I think there is this problem that there’s a difference, a big difference, between what the agencies do internally and what they do in court.
And maybe they… maybe that would be the thing to do— to rewrite the Guidelines with more thinking about, well, what does it mean for when we actually get to court? And you know, there’s some of that; they redid talking about what sort of thresholds, or what HHI thresholds and stuff like that, which I’m talking about that. But I don’t know. I still think that, I guess I come back to, that’s not going to be enough, particularly because I think a lot of places where the courts have run amok have to do with exclusionary behavior. And so, I think that’s a problem, it’s just not going to speak to that.
Scott Wallsten:
So, with that, we should probably wrap it up. We’ve been talking for almost an hour, I guess. Michael, thanks so much for being with us. We really appreciate it. It’s always fun talking to you.
Michael Katz:
Thank you. I appreciate the chance to talk to you, it was fun.
Tom Lenard:
Thanks.
