On September 7, the Stigler Center hosted a webinar to discuss the draft merger guidelines. What follows is a slightly edited transcript of the event.
Brooke Fox: Hello, everyone! Welcome to this discussion of the draft merger guidelines. My name is Brooke Fox. I’m managing editor of ProMarket, the Stigler Center’s digital publication, which publishes academic contributors and debates current events that are at the intersections of economics and government.
Last month we launched a symposium on the draft merger guidelines, where antitrust experts provide their comments and respond to each other’s pieces. This series, which was conceived and has been primarily executed by our deputy managing editor, Andy Shi, has been successful in informing the broader discussion of the merger guidelines, including citations in major news publications like the New York Times and the Washington Post.
As a part of today’s conversation, we’re excited to host this panel, delving deeper into the topics with Harry First, Eleanor Fox, Herb Hovenkamp, and Eric Posner, moderated by Terrell McSweeny.
Before we begin, please note we are on the record. We will post this video on our YouTube channel later. If you have questions for the speakers, you can submit them via the Q&A icon at the bottom of your screen. Views expressed by guests are their own, not those of ProMarket, the Stigler Center, or the University of Chicago. Please check the Stigler Center website for more details on upcoming events as well as subscribe to our Captitalisn’t podcast. Today there was a new episode out with Susan Athey, the chief economist of the DOJ’s Antitrust Division, so definitely go download that. We also hope you’ll explore the full coverage of the draft Merger Guidelines, including the newly released second round comments, on ProMarket.org. All of the relevant links will be seen in the chat function at the bottom of your screen.
And now I’ll briefly introduce our speakers. Their full bios can be found on our website.
Harry First is the Charles L. Denison Professor of Law at New York University School of Law. He’s a specialist in antitrust and business crime, and contributing editor of Antitrust Law Journal and former antitrust editor of Antitrust Bulletin, among others.
Eleanor Fox is the Walter J. Derenberg Professor of Trade Regulation Emerita at New York University School of Law. She’s an expert in antitrust and teaches, writes, and advises on competition policy around the world. Among other honors, she received the Global Competition Review’s inaugural Lifetime Achievement Award in 2011.
Herb Hovenkamp is the James G. Dinon University Professor at the University of Pennsylvania, Carey Law School. He is a fellow of the American Academy of Arts and Sciences, and won the Justice Department’s John Sherman Award in 2008 for lifetime contributions to antitrust law.
Eric Posner is the Kirkland and Ellis Distinguished Service Professor of Law at the University of Chicago. He worked on the draft Merger Guidelines while serving as counsel to the assistant attorney general at the Antitrust Division of the Department of Justice in 2022. He’s a fellow of the American Academy of Arts and Sciences, and a member of the American Law Institute.
And lastly, our moderator is Terrell McSweeny, who is senior of counsel at Covington and Burling. She is a former commissioner of the FTC and has held senior appointments in the White House, DOJ, and the US Senate. At the FTC and DOJ Antitrust Division, she played key roles in significant antitrust and consumer protection enforcement matters.
And now, without much further ado, I will turn it over to our speakers. Thank you.
Terrell McSweeny: Thank you very much, Brooke, and welcome everybody. This is truly a dream team set of panelists to discuss these revised Merger Guidelines that were announced by the FTC and DOJ Antitrust Division, July 19, 2023. It’s worth noting, these are still out for comment. The comment period, I think, expires September 18, 2023, and we will see the final version of these sometime after that point. But there is a lot to talk about in these guidelines. I wanted to start by asking each panelist—and I think we’ll start with you, Herb, if you don’t mind—for a quick take. How big a departure are these guidelines from the last set of guidelines which were issued in 2010?
Herbert Hovenkamp: Thank you. They’re a big departure. I’ll make three very quick points. One is that their declared purpose was to express—in, quote, simple and straightforward language—what the merger concerns are. On that issue, I don’t grade them very high. Yes, I think some of the language is simpler. They avoid terms like market power. I think they do a very, very poor job of capturing the public’s concern. I mean, if you look at, for example, Gallup polling on the economy. Overwhelmingly, it’s not even close, overwhelmingly, the public’s concerns are with high prices, inflation, the cost of energy, cost of food, yada yada. The word “price” never even appears in the first guideline, which expresses the principal concerns here, and it’s clearly true throughout that price is subordinate. By the way, I say this as a kind of a friendly critic. I want these guidelines to be good, and I hope they can be made better.
Secondly, the guidelines are caught in a bit of a time warp because they rely so much on 1960s and ‘70s Supreme Court case law, rather than more recent decisions from the lower courts. I think that was a mistake, because there’s much more in there than the guidelines suggest. Just a couple of illustrations. In 2013, Justice Sotomayor wrote the Supreme Court’s unanimous opinion in Phoebe Putney, striking down statute under the state action doctrine because it failed to articulate a goal of not using merger law to express market power. Every justice in the Supreme Court appears to have been behind the view at that time, that one goal of merger law is to control market power. Gorsuch and Justice Jackson were the only two members who were not yet on the court at that time. In the DC Circuit, cases like Heinz spoke of the purpose of the merger law as controlling market power, just as all of the other antitrust laws do. Attorney General Merrick Garland was on that particular panel. That was quoted in turn by Justice Kavanaugh in the Whole Foods case. And then, of course, there’s Justice Thomas’s declaration, way back in 1990 in the Baker Hughes case, that the purpose of merger laws to control market power. So anybody who wants to convince the court that the guidelines should not be concerned with market power, I think, is going to have a very steep uphill climb. Practically everybody on the Supreme Court has already articulated a contrary view. And I think the biggest thing that these guidelines can do is to start out with pretty much the same, quote, unifying theme that the 2010 guidelines created, which is that the principal purpose of merger law is to control the exercise of market power manifested by reduced output, or more particularly in the guidelines, higher prices. And I think it’s quite distressing, the very attenuated role that the concern for higher prices plays in these guidelines.
And then one more sentence, and that is, even assume the guidelines get approved, I hate to think of what will happen in front of Congress when they start discovering that we’re actually using merger policy not to control higher prices but to limit something else.
Terrell McSweeny: Those are some thoughtful points, Herb, as constructive criticism of the draft guidelines. Eleanor, I wonder if I could turn it over to you next, for your sort of quick take, high level thoughts on how big a departure the proposed guidelines are.
Eleanor Fox: Thank you, Terrell, and yes, I think the guidelines are a big departure. I’m going to take a different point of view from Herb.
I think one of the values of the guidelines is to move away from looking only at, and first at, result-oriented output limitation, higher price, and to move back to what was Congress’s concern of lessening competition. That’s what the guidelines do. They say lessening competition, eliminating, softening competition is the concern. And if competition is softened, lessened, eliminated by a merger, it’s bound to increase market power. I suspect that the next version of the guidelines is going to mention market power more, and to link it in with lessening competition. I do not think that they move away from, and want to propose a different rubric for, thinking about the merger law. They are concerned with lessening competition, which does create market power; and which normally, if that happens, does hurt consumers.
So it’s not that they do not have consumers in their sights. They open the sights up to supply-side as well as downstream consumers. But I think they do us a service by moving away from a paradigm that has led into very, very conservative law, much increased concentration in the markets, and a very great sympathy for the merging firms rather than a greater sympathy for public interest against lessening competition.
Terrell McSweeny: Harry, what’s your take? I think what we’re hearing from both Eleanor and Herb is that this is a significant departure from previous guidelines, for different reasons, though. I would love to get your contribution.
Harry First: My first take is: All the people in this panel, except for me, have written reams of stuff on ProMarket. I can barely get through all of it. But we’ve got good representation, so that’s fine.
I want to start actually by reading a quote from a very well-known antitrust commentator about changes in guidelines. And here’s what this person wrote: “Recently, however, the Justice Department proposed a radical deviation from the policy of incremental adjustment and change.” This was Joe Brodley, writing in 1987, in response to the effort by the Reagan administration, which had changed from a focus on lessening competition to increase in market power, and changed the guidelines focus to that, and then tried to get that enshrined in the proposed Merger Modernization Act 1986, not passed by Congress, two successive Congresses, and died. But that change in the guidelines completely embedded in our thinking the reorientation of what merger law was about. And to me, these guidelines are an effort to take a page from the Reagan Administration, Bill Baxter’s playbook, and say, “We’re ripping up the old approach, and what we’re doing is, we are emphasizing what was always supposed to be the emphasis, an emphasis on competition,” which, in the language of these new guidelines, emphasizes rivalry, way over exercise of market power. And it emphasizes a concern for concentration.
So I think this is definitely not an incremental change. This is an effort to change 40 years of merger enforcement and answer the question, “So, how are we doing? What’s it gotten us?” And if you look at the economics literature, it says, what has it gotten us? A more concentrated economy. Higher price-cost margins. New entrepreneurs aren’t entering. Adverse effects on labor. Big platforms that are dominating. Are we going to do something about it? Or are we going to continue to spend huge amount of resources? On what? So I think this is an effort to reorient back to the original purposes and language of Section 7.
And Herb is right that the guidelines… I thought I’d do a little word search, to see the emphasis, you know? How it’s changed. So price is mentioned, in the new guidelines, is mentioned 91 times; in the old guidelines it was mentioned 215 times. Market power is mentioned 5 times; in the old guidelines it was mentioned 35 times. By the “old” ones I mean 2010, not going back to Bill Baxter in 1982. Concentration is mentioned 90 times; 45 times in 2010. It gives you an idea of how that’s changed.
I want to also say, we’re not out of step with the rest of the world. If you compare this to the CMA’s revision of its merger guidance in 2021, it looks pretty similar. Maybe a little more mention of price. They don’t mention market power all that much, either. They mention rivalry a lot.
So now, there are a lot of questions I have about implementation, and how well the guidelines carry out these purposes of really making a big change in the way we’re doing merger enforcement. But I think that’s really the effort. I applaud the effort. I think it’s right.
And, as sort of a footnote, I had this discussion yesterday. Will there be more references to market power? I don’t think so. I think that the focus is not necessarily in the exercise of market power, which implies, no matter what you say, the first focus is on increase in price and lessening of output. I know Herb loves that. I don’t mind it. I don’t like high prices. But there’s a lot more in these guidelines that the drafters are concerned about, and rightly so.
Terrell McSweeny: We are going to be talking about some of those changes in more detail in this discussion. But first, Eric, you get the last word on the quick take for the panel. Big departure from the past? And feel free to pick up on some of the themes that the panelists have been talking about in terms of price and market power, and the role that it plays.
Eric Posner: Sure. Well, you know, maybe it’s a significant departure from 2010, but not from 1968 or from the law. I guess, the significant departure idea, maybe we shouldn’t get hung up on that. I think one of the ProMarket (am I remembering correctly?) one of the ProMarket contributions, I don’t remember the author, complained that this was a radical change, a significant departure, and then praised the 1982 version for being giant steps? You know, a big change from what came before? I think the question is whether the guidelines are good.
As Harry and I think Eleanor were both saying, the basic point here is that there’s something close to a consensus that the merger law has been under-enforced for the last 20 or 40 years. And the goal here is for the agencies to unshackle themselves from unnecessary constraints that they’ve imposed upon themselves in the earlier guidelines. These constraints in the earlier guidelines have also influenced the courts, unfortunately, and wrongly. And so it’s necessary to get rid of them to some extent. So the guidelines should strengthen enforcement, and in a way that’s consistent with the statute. I think the departure here is stronger enforcement and an understanding of the law that’s more consistent with what the law really says, as opposed to how it’s been interpreted by some scholars and some lower courts over the years.
Let me just say something about market power and price. I think this is one of the big issues, and I’m glad Herb brought it up. I see market power and price as different. I do think market power is a central issue. I’m not sure it matters so much whether the term is used or not. I do think price matters, but I think it’s important that the agencies not commit themselves to proving price increases case by case. I think that’s been part of the problem. That’s why the term probably doesn’t appear as much as in previous guidelines. I think there’s a practical reason for that, which is separate from the sort of theoretical reasons why a lot of people think that price is important in merger law.
But I think that’s a good place to stop. Terrell?
Well, Herb, do you want to rebut me? Terrell’s connection’s probably dropped out.
Herbert Hovenkamp: Couple of things. Number one, “substantially lessened competition” or “tend to create a monopoly” cannot refer to lowering the number of rivals, unless those phrases mean something different in Section 7 than they do in original Section 2 and Section 3 of the Clayton Act. For example, tying is condemned under Section 3, where it may substantially lessen competition or tend to create a monopoly. Consistently, ever since the 1936 International Business Machines tying case, the courts have interpreted that as meaning a price increase. And of course, tying in the ordinary course doesn’t reduce the number of rivals.
Yes, concentration is important. But the big question is, is it important for its own sake? Or is it important as a means for detecting something about market performance which has to do mainly with pricing? Any suggestion in Brown Shoe that we should look at concentration for its own sake was pretty much annihilated a year later in the Philadelphia Bank case, which I think is a watershed case because Philadelphia Bank opinion sites seven economists, more than any other antitrust decision up to that time, four of them on it’s 30% presumption. In fact, the four it cited all recommended a lower number. Kaysen and Turner recommended 20%, Stigler recommended 25%. But the point was that these structuralist economists were not talking about concentration for its own sake. No, they believed concentration was a tool for getting at performance. And it has consistently been used that way, pretty much except to the point of these particular merger guidelines, which, Guideline 1 in particular, talking about concentration just up in the air without making any link to price, output, market power, or anything else makes it a pretty toothless tool when you start looking for where exactly are the harms and benefits.
Eric Posner: So let me propose, in the absence of a moderator, that we just continue going in the same order that we did before. So, Eleanor, you’re next. And why don’t we just go around in circles?
Eleanor Fox: Good idea. I want to jump in and say I like this theme now on concentration that Herb raised, and I think rightly so.
And so this is what I would say about the use of concentration in the guidelines. I think they really must link it up to harm to competition. I have the feeling that the drafters thought maybe it was so obvious. But I think that they must do it on paper.
I think they are not leading with their best foot forward by starting out guideline number 1 against increasing concentration, “significant increase in concentration in highly concentrated markets.” I think that Guideline 2, which would prohibit mergers on the basis of lessening the rivalry between the two merging parties, should go number 1; and that Guideline 3, which is on risk of increasing collaboration and coordination, should go number 2. And then some form of Guideline 1 on concentration should come in next. Probably not even as a separate guideline. But to say, in support of the Guidelines new 1 and 2 and various iterations in between, of lessening competition, when there’s significant competition among the merging parties, it’s eliminated, plus increase in risk of coordination. And I also think this certainly relates to Guideline 8, which merger should not further trend towards concentration. So I would do those things.
But also I’m going to take the opportunity to say, I’m also not a fan of the “should not” language. The should not language actually doesn’t work very well, because the guidelines are saying you should not do this when what is “this” does not often have a distinct link to harm to competition. I think the guidelines should eliminate those should not, and that they should be in terms of saying: Here are the possible harms from mergers that may substantially lessen competition. And in each case, which could be a subject of each of the guidelines, link it up as to what are the necessary conditions for it to harm competition.
Terrell McSweeny: Great, Eleanor, thank you for continuing.
Harry First: So I’ll pick up on both what Eleanor said and Herb said. So, one is sort of in some sense a style point, but not quite. I really hate the way these are put together, frankly.
The first thing I thought of when I looked at it was the 9 no no’s, which I know that some of you remember on this call, was the effort by Bruce Wilson during the, this was in the Nixon administration, right?
Herbert Hovenkamp: You’re dating yourself, Harry!
Harry First: Hey, you know, dating is a good thing, they say, but it requires an app. Well, anyway, I don’t need to date myself. It’s pretty obvious.
It was an effort to say, “Here’s what is bad, and we’re gonna tell you these 9 things you can’t do.” I guess I just don’t like being told them. Then, the second thing I thought of was the 10 Commandments. So, you know, we have trouble following those. We’re going to have trouble following these. But what I really don’t like about the styles, I don’t think they’re analytical. And that’s more my problem. I’m thinking, if I were to teach these, how do you analyze a merger problem going by the guidelines, one by one? It’s really not the way to sell these to me. So, the way it’s put together, I’m not a great fan.
I want to go to Herb’s point about concentration. I just disagree. I think concentration has been an important sort of lodestar, and in a way a proxy for things that we would have trouble measuring. We don’t want a concentrated economy because of dispersion of wealth. We don’t want a concentrated economy because we want more democratic control. We want independent decision-makers. We want all these things which are hard to measure exactly, but are social policies. I think there has been, over time, a concern about concentration. And the economics literature then says (and it produces really bad results!), “And here are some of the results.” Now, maybe it’s not going to be in a particular case. So that’s the problem with the litigation case by case. But concentration is a proxy for important things that go beyond price to me.
The next thing is, the statute doesn’t talk about concentration either. And there were cases in the late 60s when Justice Department was going after conglomerate mergers on sort of a concentration basis. And one court said, “Yeah, but the statute talks about competition. You have to link it.” So my link is, we’re talking about competition. Competition needs competitors. You could ask the same question that we asked about concentration. So what’s so good about competition? And then you do get into maybe some more concrete things about what rivalry in general may produce, in terms of innovation, product choice, dispersion, decision-making, same sort of thing. Actually, it’s in the same quote from, I’m cribbing from Joe Brodley’s piece from 1986, but others, certainly others on this call, have written the same thing. So it brings a lot of benefits. But the one benefit is, it is the statutory language.
Now, of course, there is the question of, “Well, when do you stop?” What lines you draw. But in terms of focus, I don’t dismiss the concern for concentration. I think frankly, and Eric, you can tell me, you’re next, so you could tell me if the Justice Department thought differently, or the Federal Trade Commission. But this to me—read the executive order, Biden’s executive order—this is a big focus. And it’s been a big focus and a big source of concern. Certainly for a long time, but focused on in the literature since the mid-2010s. And this is an effort to deal with it. And again, I think that’s appropriate.
Terrell McSweeny: One of the things that is a big deal about these guidelines, I think, is significantly lowering the concentration threshold agencies believe should be required. Obviously, that is a big piece of of the guidelines. So, Eric? Over to you next.
Eric Posner: Sure, I was going to talk about that. That is, obviously, probably the most practically important change that the guidelines make. Reducing the practical threshold from 2,500 to 1,800. There’s plenty of reason to think that that’s wise. The 2,500 threshold was totally arbitrary, not based on any evidence or theory. It was just what the agencies had been doing before, probably because of resource constraints. Then we see rising concentration, rising markups, as Harry mentions. So there’s reason to think that there’s under-enforcement, maybe directly caused by that blunder in the 2010 guidelines.
But, Harry, the way I would teach this is: Put it next to the 2010 guidelines. But really put it next to the 1982 guidelines.
What’s going on, in my view, is that the statute is quite vague. It’s radical. It’s probably too vague and too radical. People are getting upset about it, and pressure is put on the Justice Department and the FTC not to over-enforce, and probably on the Supreme Court, for that matter. And the agencies take it upon themselves to reduce enforcement and to assure the business community that they don’t have anything to worry about. There’s lots of ways agencies do this sort of thing. Usually they just stop enforcing the statute, or they under-enforce the statute. In 1982, and then continuing into the 90s and all the way into 2010, they tell the world, “We’re not gonna enforce this statute. We’re not gonna enforce it very strictly.” They’d write it down. That assures the business community that, for a huge fraction of mergers, they have nothing to worry about. It even influences the courts, in effect causing legislative, well, effective legislative change without having to go through Congress, which would never have agreed. In fact, there was some kind of effort in the 80s to try to get Congress to change the law, which would never have worked. That’s not how things are supposed to work, but that’s how things work.
So what’s this guideline trying to do? My view, and I’m not speaking for people in the Justice Department. I don’t even know what the collective brain thinks. It’s too hard. But what I think it’s doing is, it’s cutting back on all this kind of superstructure that’s been added. I think the reason you see a lot of citations to Supreme Court cases is, it’s saying, “Look, this isn’t the law. These guidelines, this accumulation of guidelines and the cases that are based on them? That’s not what the law is. And we, the agencies, we’re not bound by them. And we’re giving you guidance. We’re telling you we’re not going to be bound by what we said back in 2010. So you should be ready.” And when people complain that they’re not receiving guidance, this is too vague? That’s what guidelines look like. Guidelines are just guidance. They’re vague. Look, for example, at the human resources guidance that the Justice Department, the FTC issued in 2016. It doesn’t give you numbers. It doesn’t tell you if the prices go up or the prices go down. It just says, “By the way, everybody, if you engage in no poaching agreements, we might prosecute you.” That’s the sort of guidance that agencies use, because agency enforcement is discretionary. That’s a big part of how our legal system works. So they’re giving some guidance. It’s going to be hard to teach because of the kind of things we don’t teach in law school. We don’t teach discretion. We don’t teach the discretionary choices that agencies make. We teach the law. And if you want to teach the law, you don’t teach guidelines. You teach cases. That’s where the law is located.
Terrell McSweeny: Well, I want to bring Eleanor in, I see her hand is up. But there’s several things going on in this conversation that I think are fascinating. We’re discussing the big change in the guidelines, and certainly there is one. We’re discussing the role of guidelines. And then as Herb led us off, there is some disconnect between the modern Supreme Court case law discussion, and the precedent there, and some of these older cases. And I do see this discussion especially, one of the interesting topics being, how do the agencies navigate both where they think the law should be, what the guidelines are explaining, what some of the older cases support with some of the more current cases and enforcement trends.
Eleanor, I don’t know if you want to jump in on that topic, or respond to something that Eric or Harry said. And I want to move in next to that, and then talk a little bit about horizontal and vertical mergers. That’s another big shift in these guidelines as well.
Eleanor Fox: I do want to respond to something Eric said, but I also would like to take you up on introducing the topic about the disconnect, which Herb first introduced, of the cases cited and the modern cases.
The point I wanted to make in relation to what Eric said, going back to 1982, which is the Reagan guidelines. I want to say there was much more intentionality than I think you implied, Eric, on changing the sense of what merger law—and hopefully they thought antitrust law—was all about. There was the beginning of the neoliberalism movement. There was a huge force that said antitrust is not good except to control cartels. And so what did these guidelines do? From 1968, guidelines and the law thereafter, before and after, which was saying, “We really have to worry about corporate power and concentration.” The 1982 guidelines basically wrote out the whole sense of the dynamic of competition between parties that might be lost by a merger. They basically said, “All we’re interested in is collusion.”
Eric Posner: That’s what I was trying to say. Maybe it didn’t across.
Eleanor Fox: I didn’t really think that this was opposite to you. But the emphasis. The only thing they said is, we’re worried about collusion.
Eric Posner: Yeah, the Justice Department stopped challenging mergers in the late 80s.
Eleanor Fox: That’s right. And they certainly weren’t challenging monopoly power. So it was all about collusion. And they use the word collusion, not coordination, which makes one thing, well, what was the level of collusion that was needed? So these were intentionally very, very conservative guidelines. And that, they say, and it’s true: Who writes the first draft, wins. Well, neoliberals wrote that first draft of last 40 years, and every change had to be based on that draft. Like, you had to justify why you were making a change. And some very good people with bigger ideas came in and did change it little by little, but it’s that first draft that still comes through. It’s neoliberal antitrust. The agencies really, and the administration, and Biden and his executive order, really want to change the sense of political economy from neoliberalism to something more like, as Biden said I think last week: We want to build from middle out and bottom up, not top down.
Terrell McSweeny: Go ahead, Eleanor, and then I want to bring Herb back in on this.
Eleanor Fox: I was just going to shift to address your question really quickly, Terrell, on the disjuncture of the cases. Sure, it’s true the agencies would like to make big use of older cases, which sometimes have clauses that are cited again and again today, like even Brown Shoe, and certainly Philadelphia National Bank holding is cited today. But there is a disjuncture with that and the modern cases. I think they would be well-served by pulling some thoughts out of the modern cases that would definitely constrain them in court.
There’s a real question of who and what these guidelines are for. It sounds, if you read the guidelines, it sounds like we’re going to go into court and we’re going to say, “If these facts are true, a potential competitor is eliminated, you’ve met your burden of proof”—I mean “we,” sorry—“we the agencies have met our burden of proof, now we shift the burden to the defendants.” Well, they’re not going to get there. So they have a lot of wishful thinking about what would happen in terms of how they present presumptions and rebuttal, and they really should be engaging with modern law. Thanks.
Terrell McSweeny: Herb, I see your hand is up, and I’m wondering if you’re going to respond to Eleanor on the disjuncture.
Herbert Hovenkamp: On the first point, yeah. What kind of big change can the guidelines make that the Federal courts will actually buy-in? I think the important first question there is to distinguish questions of fact from questions of law. The HHI thresholds are questions of fact, right? There’s nothing about them in the statute. There’s not even all that much about them as a matter of law in the case law. I totally agree with everybody who’s spoken on it, I think Eric first, that the guidelines are correct to raise the standard back to 1800. In fact, based on recent literature, I would go a little bit further than that, depending on the propensity of a market to collude.
The thing about keeping the focus on questions of fact is that they will be made subject to the standards for expert testimony in the Daubert case. That means we look at methodology. We look at well-focused, accepted empirical studies. And there I think the empirical studies are pretty convincing recently. It’s not that we’re going after concentration for its own sake. It’s that we are under-enforcing because, number one, the standards are too lenient, and number two, the Federal courts don’t even follow the standards that have been articulated. So, if you can push home the fact that these are fact questions that are driven by expertise, and then look at the methodological literature that we’ve had, which makes a very, very compelling case that there are plenty of competitive threats that fall well below even the 1800 threshold. Okay, that’s my first point.
My second point, and I’ll be really brief about this one. I think the guidelines did a huge disservice by coming out with this kind of promiscuous version of unilateral effects. I think whoever drafted them didn’t understand that the reason unilateral effects analysis has been so successful is because it is very econometric. It is focused on elasticities and responses. And as a result, it’s an excellent tool for analyzing the predicted price effects from mergers. Well, these guidelines, they have an equivalent to unilateral effects, but they want to include in all kinds of things, like attractive features and so on. I don’t know how you model that. It is going to be a nightmare for economists to model unilateral effects with a potpourri of factors that relate to all kinds of things. I’ll tell you what I think they will do. They will reduce everything to cash value, because the only way they’re going to be able to make unilateral effects work is to translate them somehow or other into prices and demand responses. And then, of course, we’re going to be right back where we were under the 2010 guidelines.
Terrell McSweeny: Harry, your hand is up first, and Eric, I’m sure you want to jump in on this as well. Harry, are you going to respond to Herb about the “promiscuity” of the unilateral effects?
Harry First: I don’t want to fully take that on, except to say, as Shakespeare said, first thing we do is we kill the economists. Economists are good for certain things. But you’re right, they’re, you know, once we move out of price, it’s… So, I used to say to the economist (we all know who it was) in Staples for the FTC that it was a bargain with the devil. They had data that would show price would go up. And then the court said, “Okay, show me the data.” And once you go down that rabbit hole, you’re in tough waters when you come to markets that are just emerging, where you don’t have good price data, when you’re looking, as Herb said, at non-price things. “Well, let’s make it into price.” And so I have nothing wrong with that. And unilateral effects has been important and useful. It just may not be the only tool.
I want to just say one thing about the concentration levels. And lauding them for returning them. There seems to be one guideline which says, I think it’s Guideline 8, mergers between a thousand and 1800, if concentration’s increasing and the delta is more than 200, that’s illegal. Bring the case. If you’re really serious about this, you want to deter these mergers, and you’re concerned about concentration? Bring the lower bound case because they haven’t brought cases in this range, and let’s see what happens. Now, on the “let’s see what happens,” I think we all on this call assume: death. Darth Vader sits in all the courts and is just going to kill enforcers. Maybe yes, maybe no. But whenever I hear that, I think about Columbia Steel case in 1948. Justice Department brings that case to prevent a steel merger, vertical steel merger, and loses before the Supreme Court. Two years later, we get Celler-Kefauver Act. So there’s precedent for shaking Congress. Now, will that work here in this? I mean, who in the heck knows! I mean, who knows who’s going to be on the Supreme Court by the time (if!) these cases ever get to the Supreme Court. So there are a lot of ifs out there. I wouldn’t moderate the guidelines figuring we’ll never get anything at any court to go along. I would try to convince the court.
Final point, sorry, which we’ve talked about or talked around, which is a question of, who’s the audience for these guidelines? I think this is maybe a little lost, and I think it’s very important, because I think it’s led to some missteps, maybe, in the guidelines. The first audience are the courts. And I don’t think you get the courts by citing cases, to make them read cases. They need principals. The second audience is staff. If you want to bring these low concentration cases, the staff has to understand you’re doing it. Eric can probably talk better to that than I. And the third is defense counsel, counsel for parties in merger. I think those are the three audiences. It’s not the general public who has to understand them. I don’t know where in the heck that idea comes from. Who in the hell’s going to understand these? I can barely understand these! So it just seems like, you know, some politically misguided effort. I’ve heard Jonathan Kanter and Lina Khan address these groups. Give us a, I mean, okay. But really, the guidelines are not addressed to those groups. It’s a technical exercise. And you’ve got to move the ball. And the first group you have to move is to convince the courts. That is what the 1982 guidelines over time did. And now it’s up to the new enforcers to convince them otherwise.
Terrell McSweeny: Harry, that’s a really important point, and we’ve been discussing it, right? There’s some big changes in these guidelines. There’s a bit of a disconnect between the modern case law and the proposed changes. And then I think Eric’s point—and, Eric, you had your hand up on this, I want to bring you back into it—is, one of the things that is going to need to happen in order to convince courts, especially in these lower band cases, or in vertical cases, or in potential competition cases, or in cases where you’re showing quality and innovation effects (which were, by the way, in the 2010 guidelines as well) is some evidence. Right? That in fact the challenge is warranted to the merger. I think, in litigation, the courts are going to be looking for something more than references to guidelines or very old cases. Eric, what’s your take?
Eric Posner: Harry sort of stole my thunder about a lot of these issues. But let me just add, one of the dangers, of course, is that the guidelines will be cited against you, which happens actually quite a bit. And if you use what the economists want, you know, like basically a cost-benefit analysis. But unilateral effects? You know, the way it’s written down: You have efficiency on one side of the greater than or less than symbol, and some other squiggly figures on the other. And those have to be estimated. And if it’s not done, or if there’s a controversy among experts, the court’s going to say, “Look, I can’t tell. You lose! It’s there in the guidelines. So if you’re trying to make an argument that there’s other stuff going on, well, tough luck.” That’s what I mean by tying your hands. When the statute and case law refer more broadly to just substantial lessening of competition. Why is it now the burden of the government to show that there’s this actual price effect or efficiency loss? That just makes it more difficult. That just reduces the scope of cases that the government can win.
The thing about Philadelphia National Bank also that surprises me. The fans of it. Philadelphia Bank is an incredibly crude, formalistic opinion, which basically says that if market shares are high enough, the government wins. It doesn’t say anything about trying to figure out what the price effects of this merger are. And the HHI threshold doesn’t require you to determine the price effects. Although if you have a strong, you know, the sufficiency rebuttal, maybe the defendants can. But that’s not in Philadelphia National Bank. It’s not clear that that should be required. If you have an economic theory, and if you have evidence that very large firms merging will most of the time (or even some of the time) generate high prices—even if all you care about is high prices, if that’s all you care about, is consumer harm—it’s still not necessarily the case that you want to require the government to prove in any particular case that prices are going to go up. It could be that it’s just too hard, on a case by case basis, to do that.
So, even if you only care about consumer harm, you may want the merger guidelines never to mention consumer prices. But (and Harry said this earlier on I think, Terrell, when you were off in cyberspace by yourself), it’s not even clear that consumer prices is all that we should care about. There are other reasons why market concentration can be harmful. Political reasons, social reasons, what have you. And those effects cannot be captured just by looking at the price effects of mergers. Also related to that are quality effects, very hard to measure, or can be measured only in a very small subset of cases like hospital mergers, which has a ton of data about quality. So again, if you limit your scope, you’re going to miss a large fraction of harmful mergers. These are all reasonable, overlapping factors that could justify limiting the relevance of looking at price effects in merger litigation, even though I think everybody agrees that the impact of mergers on consumers is important, is something that the agencies should be concerned about.
Terrell McSweeny: Herb, I want to turn to you. I also want to remind our audience that the panel will be taking questions. So if you have one, send them through.
But I want to circle back on something here. I think, as a panel, we generally agree that addressing concentration is important; that there has been under-enforcement in, especially, the merger space; that these guidelines are a big departure in order to address that. But I am hearing some real disagreement over the role of econometrics, and what evidence beyond concentration is required to justify lower thresholds and enforcement. Am I wrong about that, Herb?
Herbert Hovenkamp: No, I think that’s a good and valid question. But a couple of points here. So-called unilateral effects now account for over half of second request merger evaluations in the agencies. That number is growing. I think it will continue to grow, and I’ll tell you why, and that is, analytically it gets you to a result much more quickly.
Number two, unilateral effects formally doesn’t depend on any kind of market definition or any measure of concentration. In fact, the measures are purely econometric, looking at responses to cost changes or price changes. So, the majority position among economists at both agencies (and generally) is: You don’t need to define a relevant market in order to do unilateral effects. Now, as a practical matter, the agencies get stuck between that technical position and the Brown Shoe requirements that section of the country and line of commerce require a product market and a geographic market definition. But a lot of the reports say, clear, we’re defining a market because the law says we have to. But we really don’t need to. And I think one of the prices we’re going to pay for a revisiting of… I suspect nobody in this group wants to abolish the law of unilateral effects. But it is important to realize that they really are not concerned about concentration. They are concerned about substitution rates between best place and second-best place competitors. Whether few or many. And the reason the system works well is because of its focus on substitution and prices.
Terrell McSweeny: I think it’s worth noting that that’s true when we’re thinking about horizontal mergers in particular. But one of the other big changes in these guidelines is really, for the first time, trying to eliminate some of the distinction in enforcement and approach between vertical and horizontal mergers, at least from the agency’s perspective, and a new skepticism of vertical mergers in particular.
So I did want to turn to that topic, if we could, because I think it’s a really interesting issue in these guidelines. One of my threshold questions for the panel, and we’ll start with you, Eleanor, is: What do you make of the change, especially, that it’s the first time that there’s this presumption around certain vertical mergers in these guidelines? And historically, are the efficiencies of vertical mergers overvalued? Are the agencies right to be bringing together horizontal and vertical merger enforcement guidelines into one place, as they are here?
Eleanor Fox: Could I first make one statement reflecting on Herb and then address your question? So, the one statement, it’s about unilateral effects, and that the thing we’re really concerned about is diversion rates and substitution rates. I would say this, and it fits with what Eric said before. The thing we’re really concerned about is: Here are two head-on competitors, and we’re losing a dynamic of competition between them. There is a quality of that dynamic of competition. They want to get rid of it because it’s too much of a thorn in the side. And the only thing that we had to measure it are diversion rates. It relates to price, but there’s something about quality that is lost when two very close competitors merge.
Coming to your question about horizontal, vertical, and then sometimes to conglomerate, the agencies do want to lessen the boundaries, and mention that there are a number of cases in which there may be horizontal and vertical effects in the same merger. For efficiencies, as you just said, Terrell, there’s no particular weight given to what used to be (at least in prior guidelines was) a presumption: Vertical mergers are efficient. This presumption was partly that vertical mergers eliminate double marginalization, and, number one, therefore they’re efficient, we presume they’re efficient. I think that that is a wrong assumption that they were efficient. And here I would rely on Steve Salop’s work, which I think is really good on point. One of the things he says is, there are very narrow circumstances in which elimination of double marginalization actually helps consumers, and it has to be passed on to help consumers. And there are very few cases in which it will be passed on.
I think this actually is one of the big points of what agencies are trying to do in the guidelines. They want to get rid of that upfront presumption. Also, I would say this on horizontal. Prior guidelines, recently prior guidelines say (I’m exaggerating), “We love mergers because they’re generally good for society and consumers. They do a whole lot for society and consumers.” When in fact the agencies have said, the heads of agencies and other fora have said, “Look at all the data that has been coming out about the harms of mergers, and how a huge percentage of mergers are inefficient. A huge percent fail. The claimed efficiencies don’t very often pan out.” So the agencies are trying to get rid of those efficiency presumptions for both horizontal and vertical, and they allow efficiencies to come in on what they call a rebuttal. Of course we know, in both the prior guidelines and these guidelines, there’s a very limited space in which agency’s proof of efficiencies can actually come in as cognizable efficiencies, and then they must improve competition. And now these merger guidelines on efficiencies make it clear, too, there must be a pass-on of the efficiencies for them even to be counted.
I think that it’s generally a good move that the efficiencies of yore have been discounted, and that the agencies would like to proceed with making out a prima facie case without tipping into the efficiencies. And letting them come out, of course, when the merging parties say, “I have efficiencies and they’re good for consumers.”
Terrell McSweeny: Well, I want to bring in a question from some of our audience that is kind of relevant to this, because I think it is tricky. On the one hand, as Eleanor is saying, the guidelines are embracing kind of a broader view that addresses some of the questions that have been raised about efficiencies, and whether mergers are really generating them.
But the question that we’ve gotten from the audience is: Do the guidelines suggest an alternative harm to competition, distinct from higher prices?
And I guess a broader question is: There are thousands of reportable transactions every year. The agencies can’t challenge all of them out there. What should they focus on? And do these guidelines really address those key areas?
I’m going to turn it over to Harry. Looks like he wants to jump in.
Harry First: Let’s see, what should they focus on? So, one footnote on efficiencies. Which is, what’s an efficiency? I would like the final guidelines to say that elimination of double marginalization will be treated as an efficiency defense, subject to defense having to prove it’s magnitude in short, just like any other efficiency defense. Actually, I’m a little puzzled that it’s not in the guidelines, not mentioned at all. So that’s sort of a footnote.
Sorry, Terrell. The question?
Terrell McSweeny: The question is, do the guidelines suggest an alternative form to competition distinct from higher prices? In either static or dynamic context?
Harry First: Yes, I think that they do focus, as we’ve talked about, on innovation, on potential competition, and on the process of rivalry. So it’s not just higher prices. The question, I think in the end was, so how do you pick them out? Out of all of the mergers, right? That’s what I wanted to answer.
If you step back and look at second request percentage, the percentage of second requests over time has been pretty constant. Around roughly 2% or something like that, but pretty constant over time. That says that each year, year in and year out, there’s the same percentage of bad mergers. Now, that just can’t be right. Some years there must be more bad mergers, and some years there must be fewer bad mergers. It just doesn’t make sense. But you look at the data that, this is in the Billman Salop article, and I’ve looked at the data. Why is it the same? And the answer is something that Eric mentioned. The agencies are picking something. I’m not quite sure how they’re doing that picking. And I would urge the agencies to think more directly about a deterrence model for picking, and what kinds of mergers you want to deter. That’s why I would go after mergers with low HHI thresholds. High deterrence value.
We’re obviously not stopping every illegal merger. So how is that sorting being done? I don’t know. I’m not inside there. Eric was inside there, he’s going to tell us now. But that’s a real question. And when the guidelines say, “Oh, we’ll be reasonable and use discretion,” that means, “Well, we’ll let some illegal ones go. We won’t challenge them.” So which ones?
This is not new to these guidelines. It’s a constant problem.
Terrell McSweeny: Eric, I’m going to bring you in on this. But there’s also, I mean, there’s a lot in these guidelines that is new, and describes some very specific scenarios, as well as how it relates to digital markets and platforms and some other things.
So, Eric, I’d love to bring you in on that. How do they, in a very broad set of guidelines and a very expansive view of the enforcement mission which is articulated in these guidelines, how are they selecting cases and how do they make decisions in this context?
Eric Posner: I think maybe people are a bit unrealistic about what they expect an enforcement agency to do, in terms of telling people in advance how they decide what cases to bring.
If you think of every other enforcement agency in the country, US Attorneys’ offices, the SEC, they don’t issue guidelines where they tell everybody in advance exactly what kind of cases they bring. They’re constantly changing their priorities. One day they’re worried about (take the US Attorneys’ offices) drug cases, and the next day, human trafficking, and a month later, insider trading. So the resources are going toward one type of case, and then they’re going toward another type of case. And they’re responding to some sense of public harm. They’re not counting up. They’re not looking at necessarily prices, or some narrow accounting of harm to people. When you look at what the agencies are doing, they might be worried about platforms. They might be worried about certain types of international activity. They might be worried about harm to workers. That’s something that appeared on the radar screen about 5 or 10 years ago. But they can’t really embody all this into guidelines, in the sense of saying, “We think the following types of mergers are going to be the most important.” Something new could happen at any moment.
The guidelines are more on the nature of providing some methods that they use for distinguishing between types of behavior that raise concerns and types of behavior that are innocent. And of course, if you’re a small firm, you almost never have anything to worry about; and if you’re a big firm, yeah, you should be careful. I mean, that’s helpful because it tells 98% of percent of the country (or 99% or larger) they have nothing to worry about. Then there’s certain types of transactions that will raise eyebrows. If you really want to make a prediction, you can read the cases. That’s what lawyers usually do if they want to give advice. And then these guidelines will help you a little bit, in giving you a sense of what the priorities of the agencies are. But I can’t tell you, and I don’t think anybody in the Justice Department could tell you, even the leadership, what… Well, I think they probably do when they give speeches. They say, “This is what we’re worried about at the moment,” and I think that’s probably as much guidance as they can give. But they really can’t be that precise. I think there’s something a little bit eccentric about the merger guidelines. As I was saying at the very beginning, I think part of the complaints about the vagueness of these merger guidelines reflect unrealistic expectations that were created by the earlier guidelines, which were probably too detailed. They reflect a kind of economist’s vision that I don’t think reflected an understanding about how much discretion agencies actually need if they’re going to do their jobs properly.
Terrell McSweeny: Well, we are almost out of time, and I wanted to give everybody about a minute to have little wrap-up remarks. My main question is: What do we think we’re going to see next? Do we think there will be significant revision to these guidelines after the comment process? Do we think they’ll probably remain pretty much the same? And if they do, are the agencies going to struggle with convincing courts of this vision, given the shift, Eric, that you just mentioned and, Herb, the case law that you cited at the outset?
So really, it’s two questions. And I want quick responses (this is hard!). What next? And how impactful will these guidelines be on actual enforcement?
Herbert Hovenkamp: I expect the revisions, after the note and comment period is over, to add a little bit more discussion about prices and market power. I think Harry said that a while ago, and I totally agree with that. Based on my own former comment about those issues being questions of fact, I think that is the one place where the guidelines can ramp up enforcement considerably and get buy-in from the judges. Because the standards for expert testimony are pretty favorable to the Justice Department and the FTC’s position, if they can show that the dominant literature supports them. And it is. Dominant literature right now supports much more aggressive concentration standards than we are currently enforcing. So if they want to win cases, I think that’s what they have to do.
Terrell McSweeny: Harry, over to you.
Harry First: Now we’re in for incrementalism, maybe some changes, but I don’t really see big changes. People, when they write drafts and put them out in public, I don’t think they say, “Oh, you know what? Let’s tear this one up.” Unless they just abandon it. So that’s just my guess. Will they continue to struggle? Will they struggle? It depends on whether they push the bounds of some of these guidelines. If they’re cautious in enforcing. They’re obviously struggling right now. A number of the guidelines, were they adopted, would make the mergers they’re struggling with illegal. Now, are they going to write a brief saying, “Oh, by the way, we have new guidelines, and you know what? Microsoft’s acquisition of Activision. It’s illegal. See the guidelines.”
Terrell McSweeny: Just to be clear for the audience, the guidelines are not legally binding.
Harry First: Yes, of course, they’re not legally binding, right. They’re not even binding on the Justice Department, as they remind us. So, and how are we going to know? Let’s see, today’s Wall Street Journal: Big paper industry merger in the works, creating a 20 billion dollar company. Let’s see what happens.
Terrell McSweeny: Eleanor, over to you. What next? And are they going to struggle, given where the case law is? In under a minute. And, Eric, be thinking about your last comments.
Eleanor Fox: What next? I think there will be revisions. I think they will be substantial revisions, but keeping their vision.
I think that for market power they’re going to still focus on lessening competition and simply ride off that to say lessening competition harms market power.
Will they have a hard time convincing the courts of their vision? Oh yes! But I agree with Eric’s very insightful points about the fact, and I think that Jonathan Kanter makes this point very often, that we’re going to, he says. You have to do it if you can, but structure the facts, get good cases, structure the facts, get good experts. That is the hope to move the law to make it more aggressive. There’s not a huge amount of hope to move the law to make it more aggressive without legislation, which is probably not going to happen.
Terrell McSweeny: Thank you, Eleanor. Eric?
Eric Posner: Agreed. There’ll be revisions, but the overall vision will be retained.
And the courts? I mean, the merger guidelines can’t and don’t purport to change the law. They reflect the agency’s priorities within the agency’s understanding of the law, and the courts will do what they want, what they understand the law to be. I think that goes without saying.
Terrell McSweeny: Well, I want to thank all of you for joining this discussion today and for everybody that’s been watching it. We certainly covered a lot of territory historically and in the law and in policy as well. I can’t imagine a better team of experts to talk about some of these issues. So thank you Herb, Harry, Eric, and Eleanor. Really appreciate everyone’s time, and we’ll see what the next version of the guidelines holds.
Articles represent the opinions of their writers, not necessarily those of the University of Chicago, the Booth School of Business, or its faculty.
