The Stigler Center for the Study of the Economy and the State hosted its annual antitrust and competition conference in late April. The following is a transcript of the Judges Frank Easterbrook and Diane Wood’s keynote conversation with Stigler Center Fellow Filippo Lancieri.
Good evening from Chicago to everyone watching us from home and good evening to all of you joining us here. It’s truly a pleasure. As Chris said, my name is Filippo Lancieri. I’m a postdoctoral fellow at the ETH Zurich Center for Law and Economics, and a Stigler Center Research Fellow. And tonight, I’m truly thrilled to moderate this keynote conversation between Judge Frank Easterbrook and Judge Diane Wood on antitrust policy in the US and the role of courts. So before we begin, let me please note that we are on the record and live streaming. And we’ll also post this video on the Stigler Center YouTube channel later, so we’ll have plenty of time for you to ask questions, just know that we’re recording. As usual, the views of Judge Easterbrook and Judge Wood are their own and not those of the Stigler Center at the University of Chicago. And we hope that you’ll join us for more upcoming events, so please check the Stigler Center’s website, as well as our publication ProMarket.org and the Capitalisn’t podcast. Now, let me briefly introduce our speakers.
The Honorable Frank Easterbrook is a judge on the US Court of Appeals for the Seventh Circuit, and a Senior Lecturer at the University of Chicago. Previously, he has served as Deputy Solicitor General of the United States. And he was the editor of the Journal of Law and Economics, and is the coauthor also of “The Economic Structure of Corporate Law” among many other publications that we could cite. And the Honorable Judge Diane Wood is also a judge of the US Court of Appeals for the Seventh Circuit, and also a Senior Lecturer at the University of Chicago. So this is a U of Chicago affair, let’s put it like this. She previously served as the Associate Dean of the Chicago Law School, and she was also the Deputy Assistant Attorney General at the Antitrust Division, Department of Justice with responsibility for the division’s international, appellate, and legal policy matters among others. And I double checked, both of them have taught antitrust law at the University of Chicago Law School as well. So we have the full house of judge, former practitioner, enforcer, and now an academic as well. So thank you very much for joining us today. We’re delighted to host you.
My idea is to start with a couple of questions, really just to get a conversation started. And we have such an amazing room, I want to leave plenty of time for q&a and for interactions. And then I’ll interject a couple more questions as well. So question number one, and it is the motivating question for today. No, not for tomorrow at this conference.
How would you define the consumer welfare standard? Frank, we’ll start with you. And then we’ll move this way.
The Consumer Welfare Standard was loose language written by Chief Justice Burger in a Supreme Court opinion that didn’t involve it. And I think I was in the Solicitor General’s office at the time this phrase was penned. It was kind of cribbed from Bob Bork’s book “The Antitrust Paradox,” which had just come out. The Chief Justice meant it, well, to the extent Chief Justice Burger meant anything seriously, meant it as a shorthand for economic efficiency. But it wasn’t in the, it’s not in the nature of the Supreme Court to draw graphs and put up a welfare triangle and talk in the language of calculus. And so something simple was used. But it was perfectly clear to those of us who were reading the opinion that it had to be understood as a shorthand. And I don’t, I’ve gone off in my career as a teacher and scholar without ever using that phrase, except as a shorthand for “avoiding the deadweight welfare loss of monopoly.”
Would you agree, Diane? Would this be your view?
I certainly agree that it was a shorthand. I would say, from the perspective I gained at the Department of Justice, it was a very unfortunate shorthand because people took it literally. They took it as a signal that the only thing we cared about was perhaps price or choice effects for consumers. And I remember endless debates with the Canadians, for example, who said, “Oh, we have a total welfare standard, you have a consumer welfare standard.” And I tried to explain, actually not very successfully, apparently, that our law did in fact address, let’s say, monopsony. And that had been true since the Mandeville Island Farms case in the 1920s. And so that isn’t really a, quote, “consumer welfare” application. So I think it was an unfortunate turn of phrase that people who were not immersed in the field took, and it has led to some mischief in understanding exactly what the antitrust laws are supposed to be covering.
Okay. And so I think, building on this. Do you think the Consumer Welfare Standard does drive, or is a central point in, US antitrust policy? And do you think it should be? How do you see this, thinking from a more policy perspective on the application nowadays? And what’s happening on the ground nowadays?
Do you mean by that Consumer Welfare Standard, as opposed to avoiding the deadweight welfare loss of monopoly? Of course, judges, professors of antitrust, and economists talk about such things as welfare laws. Lawyers tend not to do that. And judges are just a subset of lawyers, with usually a much broader portfolio than lawyers. So if you’re an antitrust lawyer, you’re a specialist, you might start talking to judges in language you assume judges would understand. That is, something really simple. And that’s about what judges do understand. Something really simple. Judges are generalists. But behind the scenes, if you go back and hear an antitrust case, and talk with your colleagues, you know, I’ve just past 38 years on the federal bench. I can’t recall a single discussion with my colleagues in which the question is, “Well, does this comport with the Consumer Welfare Standard or some other standard?” Right? What the judges understand is that we’re trying to avoid the loss caused by monopolies. Whatever those are. Monopsony cases are pretty rare. And I don’t think anybody has any trouble seeing them as the same thing as monopoly cases. But it’s not. It may be how antitrust lawyers sometimes talk to judges, but I don’t think it’s how judges talk among themselves.
Judges are looking for ways to decide cases. And so if somebody says, here’s a, let’s say it’s the unusual case where a merger is being reviewed at the court of appeals level. Very unusual these days, but we’ve had some hospital mergers and other kinds of mergers. And the companies are going to come in and say, “Oh, you know, this merger is going to lead to all sorts of efficiencies. We’re going to be able to have one MRI machine instead of two, and they’re really expensive,” or whatever other efficiencies they want to point to. And you look at it, and you realize, “Oh, but there are going to be so few hospitals in this area. Like, you with 75% of the market, and a smattering of others.” That’s the kind of thing where judges might say, “What is going to be the effect of the price that people pay for their hospital services when effective competition has been eliminated by the merger?”
And so I do think judges are going to talk about it. In some ways, it’s kind of a bad market for them to do that, because it’s a market that’s so complicated by the existence of third party payers.
Third party payers.
Exactly. But people kind of get it. You know, they get these bills. The bills are eye popping most of the time, at least until you find out how much your insurance is going to cover. And so you can say to them, “Maybe we should stop this merger. Because when there’s more competition among hospitals in the area…” This has actually been shown by studies that have been done, when mergers were either allowed or not allowed. The ultimate price goes up, no matter who’s paying it. Maybe it’s the insurance company who’s paying it. Maybe it’s the employer who is trying to calculate what the premiums are going to be on the down-the-line policy.
I do think it has some utility in that intuitive sense.
Yeah. Maybe I should modify what I said. I think thinking about, consumer welfare is helpful in the sense that it tells judges who they should be listening to. Right? In a hospital merger case… I’ve never been on one, but I’ve watched a lot as they’ve gone by in the federal judiciary. In a hospital merger case, you would want to listen to what the third party payers are saying, because they’re consumer proxies. And if Blue Cross is yelling bloody murder, you might think, “Well, the main effect of this is not efficiency but an increase in prices.” If, on the other hand, some other hospital, a competing hospital shows up and says, “We’re against this merger,” that might tell you that that hospital is worried that there will indeed be efficiencies and it will drive down prices. You can learn about the transaction by seeing who’s against it.
Okay. I think, building on this, something that came out of the last panel on the Consumer Welfare Standard itself is that, maybe something more useful is to talk about how do courts and regulators and enforcers define what is anticompetitive behavior, anticompetitive harm. And I hear you talking a lot about prices and the deadweight loss of monopoly. So would you say that this is the primary focus of antitrust policy? Because that’s what we can measure. Right? And if yes, why should it be this way? And if not, how would you go about incorporating it?
I think it’s the primary focus of antitrust policy, but not because it’s something we can measure. Well, in principle, we can measure it. But if you look at how antitrust cases are actually litigated, the parties hire competing experts who perform interesting data collections and regressions which, if you’ve actually been a professional like Diane and I have, you can understand, but most district judges and other non-specialist judges don’t grasp as readily. And it makes it very hard, especially when you’re trying to make predictions. Not about what has happened in this market, but you’re trying to make predictions based on propositions like the structure, conduct, performance paradigm, it’s going to be more concentrated, prices will rise. And you know, remember Yogi Berra, “It’s always hard to make predictions. Especially about the future.” So if you’re in a world of predictions, you don’t get anywhere, but it tells you what you ought to be thinking about.
The alternative is thinking about, well, just as Rufus Wheeler Peckham said, small dealers and worthy men. If you think antitrust is supposed to protect small dealers, you think it’s then designed to injure consumers by driving prices up. And you want to have a framework for thinking about antitrust law that focuses you on something that is understandable, at least in principle, rather than maximizing four or five or six different things, which are incommensurable. And you then don’t know what distinguishes a right decision from a wrong one.
But if I get you correctly, you mean to say that you don’t think entities should only focus on prices, but ultimately price is the main one anyway. So that’s what happens ultimately? That’s the final…?
Prices are important to welfare. Right, I’m wearing this tie today in honor of George Stigler whose name this Center bears. It was his favorite tie, an Adam Smith tie. And Stigler, when he taught occasionally at the law school, had his capsule biography that was handed down to the students. It said, George Stigler has written this book about prices, and that book about prices, and that book about prices, and more than 50 articles about prices. He is interested in prices. Yeah, well, this is the Stigler Center.
Fair enough. What do you think?
Here, Frank and I probably have always diverged a bit, because, there were a couple of points I’d make about the antitrust laws. Certainly we agree entirely that the cases about prices are important. And they matter whether they happen to be monopoly cases, or whether they happen to be cartel cases. Those, I think there’s been widespread agreement about that type of case, not just in the US but around the world for a very long time.
But there is another branch of antitrust that has waxed and waned over the years. We seem to be in a wane right now. That is the exclusionary practices cases, where somebody is, in fact, taking some sort of strategic action to push a competitor out of the market. Now, that competitor might not be a mom and pop inefficient small dealer and “worthy man.” Notice that it’s men who we’re talking about.
Right. It could be just another maverick kind of firm in the market, somebody who’s annoying somebody, who you think that, if you can just get them out of the way, you can raise prices. And I think those cases matter, too. And so I think it’s entirely within both the historical scope of the US antitrust laws and the policy scope to reach them.
Now, what the Supreme Court likes to say is a phrase, ironically, which came from the 1954 Brown Shoe decision, which not many people of Frank’s persuasion would put up there on the pantheon. But in the Brown Shoe decision, the court came up with this catchy little phrase, “The antitrust laws are for the protection of competition, not competitors.” And the phrase laid dormant until 1977, when the Continental TV against GTE Sylvania decision came down, and the court resuscitated it in quite a different setting, certainly, than the Brown Shoe merger case, and has clung to it since then. You can find quite a few repetitions of that citation. Sometimes they go all the way back to Brown Shoe, sometimes they pretend it began with Sylvania. It doesn’t matter. But I’ve always myself thought, it’s fine to say that. And I understand that if you had a market with a decent number of participants, maybe the antitrust laws don’t really care if it’s distributor A who’s distributing the product or distributor B, because the ultimate effect of choice in the market and price is going to be the same. So if A isn’t such a good distributor and leaves, and B pops in, okay.
But you are not going to have any competition unless you have competitors. And I think that it is very important to find a way of making sure that the exclusionary practices branch of antitrust law doesn’t get lost in the shuffle.
Well, as Diane said, we have some disagreement about this, I think not about that principle. Right? If the taxi industry had figured out a way to dynamite the headquarters of Uber when it was just being created and gotten rid of a very disruptive factor in the market, they would have been much happier, and consumers would have been less happy. The difficulty is finding those cases when another firm in the industry screams foul and cries unfair. Because the most vigorous competition, the thing we want the antitrust laws to protect, injures rivals. Injures business rivals. Injures them seriously. Knocks them out of business. And of course, if you’re thinking about the discovery and antitrust cases, you will inevitably go into the files and see, “We’re going to kill so and so!” Right? But of course, that’s consistent with lowering the price producing better products.
The question becomes, do we have a good algorithm, something that judges can follow to distinguish the results of hard competition on business rivals from those things that will actually be deleterious to welfare in the end. And I’m very skeptical that we have a good way of doing that.
And so if you take this argument to the extreme, it will be perfectly okay for a market to have only one competitor if is the most efficient, and there is no illegal behavior?
Specifically, if it’s a natural monopoly, you would expect only one competitor.
But anybody who says he’s a natural monopoly is almost certainly lying.
What worries me about that, and maybe we’ll get into this in a few minutes, is I think you’re in trouble by the time you’re down to three competitors, to be honest. Because one of the things the antitrust laws have been quite ineffective at doing, and it also happens to be something that the people who write legislation in Congress actually can’t figure out what to do with either, is the problem with oligopolies. Everything that’s been tried has been a flop. And so once you get to that point, you just have natural human behavior to deal with. People can pick up the Wall Street Journal and figure out what their competitors are doing and pricing. They’re not agreeing. They’re not doing a single thing that the Sherman Act reaches there. Let’s assume it’s a 33/33/33 oligopoly, so nobody’s dominant, or, you know, a monopoly, or even attempting. You just can’t get at it. So my feeling has always been, again, and my interest in keeping enough competitors there, that as a structure of the market, you can have successful competition and all the benefits that we expect to get from it is, you’ve got to do your merger policy right.
Because I’m not sanguine about finding some legislation that’s going to solve this problem. It was tried during the 60s, it was tried during the 70s, people have been throwing bills in the hopper even now. Maybe I’m wrong. I’d be glad to read anything somebody gives me, but it’s been a difficult record.
I remember when I arrived on the faculty of the University of Chicago. In 1979, I started attending the industrial organization workshop and the law economics workshop and some others, populated, often attended by George Stigler and Gary Becker and others. And one of the phrases that was very common in those workshops, and backed up by a decent amount of empirical work, was that in markets, there were only four numbers that mattered: 1, 2, 3, and every other number, right? By the time you had four, you had effective competition. Three is risky. Two is a real problem.
I think he sets it too low, actually. I think there’s a number, maybe it’s seven, you know, but four? I’m thinking of the number of cereal companies, or the number of gas companies. These are examples.
That’s an empirical question. I was dumping on empirical work in litigation, in part because a lot of it is predictive and in part because it takes time to do good econometric work. But you have, after a merger has been either approved or disapproved or a given practice has been approved or disapproved, there’s plenty of time, right? You mentioned that I was, at least for six or seven years, one of the editors of the Journal of Law and Economics, founded by Aaron Director, continued by Ronald Coase. They specialized, at least in the early years, and to some extent still now, in long-term studies of cases where there had been mergers or where the court had found predatory pricing. Were aware, right? And you come back twenty years later and look, what have the numbers told you? What you can actually add up in all that time. And that phrase, 1, 2, 3, and every other number, was a way of summing up what was there. Like any other easy summation, it is, of course, wrong at the edges. And it’s, of course, subject to challenge. But it embodies a wisdom that you will never get in litigation as it’s ongoing. Because, you may all think litigation takes a long time, and we think it does, too. But in many ways, it really moves too fast for an efficient, careful study of what’s happening in these markets.
Let me build on this. So how much do you actually pay attention to academic debates? Because I think that’s something we were discussing today. John Kwoka introduced some evidence, we can discuss the evidence, that maybe six players might actually lead to, already, higher prices. So let me ask this a different way.
I’m all in favor of more data.
But what was the last academic article on antitrust that you remember reading?
I read the Journal of Law and Economics cover to cover when it comes out. There are always studies, right? None that are standing out at the moment. I read the Journal of Empirical Legal Studies cover to cover when it comes out, but there’s less in it about antitrust. And I read the Journal of Law and Economics and Organization, less about antitrust. But, one that stands out? None that comes to mind at the moment.
Diane? What was the last one that you remember reading?
I read the Yale Law article by Sanjukta Paul, and I read some of the books that have been coming out about labor markets and different things when I find the time.
But I want to make an important point here, which I hope that people pay attention to as you are debating what you think the courts can do and what they can’t do. Lots of the summaries that I read for this program caused me to think, these are fine policy debates but they, number one, don’t really fully take into account that, at least Frank and I, are not our own bosses. We have nine other bosses in Washington. And we could be as persuaded as we could possibly be by an article, and if it doesn’t comport with what the Supreme Court is saying, the article’s just so much intellectual fun. I mean, it’s not something that we can do that much about. Sometimes, as you read these articles, you realize that there’s some play in the joints. You know, if the court has left the door open, for example, for empirical evidence, “We think it’s very unlikely that a predatory pricing conspiracy could last this long. But if you can show us some evidence, we’ll listen,” you know, something like that. Okay, that’s, something that we can work with.
So that’s one caution I would put out there. The other one is, the reason we’re talking so much about judges in the United States is because we do not have the same kind of centralized enforcement discretion in the federal agencies that our counterparts pretty much everywhere else around the world have. We have private litigation. We have state Attorneys General. We have the Justice Department. We have the FTC. You know, it’s like, come one come all and bring your cases to the courts. And so you don’t have that kind of opportunity to incorporate whatever the latest economic thinking is into your actual final product law. Maybe the agencies, if they can persuade a court to do something, that’s fine. But the courts themselves, in a hierarchical system, are going to need either to find some room in a Supreme Court decision to operate, or Congress is going to have to step in and say, “We don’t want you doing this anymore.” And I just happened to be in my law school office earlier this week. And I’d just kind of forgotten that I had done this. But I had written a foreword to some little book that was a compilation of all the legislation that Congress had ever passed in the antitrust area, up to the date of that book. So it was big things that you know. Sherman and Clayton Act, Robinson Patman Act, whatever. It was things like the Newspaper Preservation Act, and the Soft Drink Bottlers Act, and the Webb-Pomerene Act, and all these little things that were in there. And so Congress does act sometimes. Whether this is a great thing or not such a great thing might depend on where you stand.
I wanted to build on that. So, Daniel Crane has this terrific article on “antitrust antitextualism.” There’s a hard one to say. Basically he claims that the history of antitrust policy is Congress expanding enforcement and the courts cutting it down. And then most of the changes that we’ve seen in antitrust policy since the 70s, or the 80s, are mostly courts-driven.
Is this something that in any way worries you about some kind of democratic deficit or not? When people talk about this as just part of the evolution of the common law?
It’s certainly doesn’t worry me. I don’t know how many of you have read what I think of as the greatest opinion in the history of antitrust, written by William Howard Taft when he was on the Sixth Circuit, in Addyston Pipe and Steel shortly after the Sherman Act has passed and the Addyston Pipe merger is challenged. Taft goes back and he tries to reconstruct the history of the Sherman Act and the history of the common law of mergers. And he says in the bottom line, “Look, this is really picking up the English common law of competition, but changing it in a fundamental way. In English law, cartels were not enforceable. But now they’re illegal. They can actually be prosecuted.” But that is a deep understanding of where we came from. It identifies the Sherman Act as part of a common law tradition, but making it criminally and judicially enforceable. And that’s basically what judges have done ever since.
What you’re referring to as the modern willingness of the Supreme Court to describe the Sherman Act as a common law statute really began with Justice Brandeis’s opinion in the Chicago Board of Trade case, where he says, “Look, the statute says all combinations, contracts, or conspiracies in restraint of trade are hereby declared to be unlawful.” But every contract is in restraint of trade. Once I agree to sell you two tons of plate steel, I’ve agreed not to sell it to somebody else. The contract is in that sense in restraint of trade. But the contract is essential to trade. And so you’ve got to have some common law overlay on that. You can’t stop with the language. Because if you do, you’re condemning the very practice of competition, right? Just think about any firm, internally, has lots of cooperation and lots of things that restrain trade inside the firm, the better to compete outside the firm. There isn’t really any rule for how big a firm should be. That was the point Ronald Coase made in his famous article on the theory of the firm. At some point it becomes cheaper, more efficient to do something by market. And at some point, it’s more cheaper to do it by command inside a firm. You can read the Sherman Act until your eyes glaze over, but it won’t tell you where that is. And it’s not going to sit still. It will change as economic circumstances change. And if the judges don’t understand that, we have trouble.
But the judges are less and less appointed for their economic expertise, right?
Not at all, actually.
So does this worry you? Is this something, like, such a fundamental part of economic policy is driven by judges sitting deciding specific cases?
Or for that matter driven by the antitrust division? Diane mentioned the Robinson Patman Act, which I assume most people in this room think of as an unmitigated disaster, and which the Justice Department has refused to enforce now for probably close to 50 years. The FTC? Well, it was representative Patman who thought the FTC would be the big enforcer of the Robinson Patman act. And for a little while it did. And it finally gave up in the 70s after, I don’t want to give myself too much credit, after I managed to lose a Robinson Patman Act case in the Supreme Court, naming as the respondent the Great Atlantic and Pacific Tea Company, whose behavior had given rise to the act! If by the 70s, you can’t convict the great A&P Tea Company of violating the Robinson Patman Act, it’s toast.
Do you agree, Diane? Do you think courts are well equipped to drive these changes in policy?
Well, I want to be careful in what our starting point is. If our starting point is, as it probably should be, that the antitrust law that we have, for better or for worse, is the 1890 Sherman Act supplemented by the 1914 Clayton Act, then you do have some rather delphic language. And I’m a big fan of Addyston Pipe and Steel too, actually. I think it’s a very fine opinion, and I’ve cited it. And actually, I find in it, also, the caution that the English common law had against overly expansive covenants not to compete. A narrowly tailored covenant not to compete was fine. That preserved the value of the business you were selling. But if it was too expansive, it interfered with a person’s right to ply his trade. And so there was a sense of market access even in the Addyston Pipe and Steel case.
So if you’re dealing with the Sherman Act, and that’s what you’ve got, then the corollary is that the court is going to put some meat on the bones, and that’s going to be informed, probably, by a lot of things. It is going to be informed by the economics of the day. It’s going to be informed by the industries of the day. The trade practices. How did those industries interact with each other? We hear a huge amount these days about, “What are we going to do with the giant tech firms? You know, the Googles and the whatever?” Of course, I don’t know what they’re doing to themselves right now. Perhaps laying people off last I heard. But in any event, yes, the economic theory matters. But the way competition is occurring matters. At what scale competition is occurring also matters a lot. Is it some little town in the middle of nowhere, where it’s just an isolated little spot? Is it, as is true with many things, the entire globe? That’s going to also factor in. So we have this very elastic, kind of, “go forth and think” statute. And that’s what the courts have been trying to do. If that’s what you bought, then sometimes they’re going to see things in the light of one economic theory. New economic theories can come along. I’ll give you the example, in the field of predation, of let’s just say, skepticism that predation really could ever work. But that was against an economics that hadn’t yet been really grappling with the insights of price theory and other kinds of things that showed that things that on the surface didn’t look like they could be economically feasible actually could be economically feasible.
I’ll just conclude by saying, the problem that I have with some of this, especially with this “just throw anything into the hopper you want,” you know, all your economics, is we have made the law almost impossible to enforce. I mean, you can’t really bring a promising antitrust case without a tremendous investment in the best economists that you can find, and the best empirical studies. And a lot of people just can’t afford it.
Diane and I agree completely. We need simple roles.
Yeah, I’m with you on that.
We judges are responsible for too much complication in these roles. But we’ll also point the finger at our bosses.
Still more questions, then I open it, so please prepare your questions and then I move. So I think, building on this, so Michael Carrier has a study showing that, basically, plaintiffs always lose rule of reason cases, no? They lose like 96% of the judgments. Is this something that worries you at all? Or is it something that you think is just, that’s how the law should be? And it’s fine?
I would say it depends which cases they’re bringing. Because I remember an opinion Posner wrote, the Valley Liquors case, if I’m remembering correctly. But he was saying, you can bring the rule of reason into the realm of things that are subject to sensible prediction if you realize that, if you’re talking about the rule of reason, if the firms have virtually no market power, nothing they do is going to really affect anybody. So you can actually get rid of a lot of rule of reason cases that way. And if you then put the other overlay on, say, one of my favorite interest cases, the BMI ASCAP case. Just because people have gotten together doesn’t necessarily mean they’ve cartelized. You have to ask, what are they doing? Well, in that case, the court decides they were actually, through the package license, creating a new product. And it was a very efficient, desirable new product that people wanted. So it would have been folly to have condemned that, quote, per se. So it eventually goes back to the Second Circuit, other things happen, but it was an interesting rule of reason case.
So I think the rule of reason depends what you mean by it. If you just have, like, a giant cauldron, and you have all sorts of things in it, and you pull one thing out and it says affirmed, and another thing out that says revert, that’s no good.
But I mean, the assumption of this will be that basically 96% of private litigation is just bad litigation. Is this how you see it?
You need to see, what’s the market that they’re operating, I would say.
And it’s a surprising number because it implies that the plaintiffs lawyers are shooting themselves in the foot. Right? Many of these lawsuits are brought on contingent fee. And if you’re losing all but 4% of your cases, how are you going to even cover the costs of running your firm? What the economics of litigation would suggest, what you really want to do is, look not at the percentage of wins and losses, but what’s the average settlement value of the cases that are settled? And of course, we know many, many more cases are settled than are litigated to judgment.
Oh, many more. And the other thing certainly to remember, too, I don’t know what this data show, but I would want myself to draw temporal lines between, let’s say, pre-Sylvania, post-Leegin, you know?
That’s new data. It’s about the 2000s, I think, from 2008.
Yeah, I mean, because a lot of cases for a long time were being brought in the vertical area. And once the Supreme Court sort of put that to an end with the Leegin Creative Leathers, I would be surprised that people were bringing too much.
And so and my final question. Then I will open it, and Steve already is jumping to ask questions. So I think what comes out of this conversation is that, you seem fairly confident that the economic scholarship is ultimately making its way to the courts. Well, in a way, there seems to be a disconnect between many areas of the evolution of economic knowledge. Think of, I don’t know, vertical restraints, where there is almost a per se presumption of the legality of it. Very few cases are won on purely vertical grounds. And the notion of foreclosure that comes out of the scholarship overall.
How long do you think is the lag between what happens in the advancement of economic scholarship and knowledge, and the time the courts process it? Or do they?
Well, historically, antitrust law long preceded economic knowledge. There wasn’t a good theory of cartel behavior until John Robinson created it in the 1930s, long after the Sherman Act. Around the same time, Frank Knight and friends are founding what’s now often referred to as the Chicago School of economics, and questioning many of the Supreme Court’s decisions. It took 40 years for that to influence the decisions of the Supreme Court. Right? If there is some other movement, I mean, I’m not going to Thomas Kuhn’s extreme of suggesting that the only way you get change is for the proponents of the old view to die because, given the lifetime tenure of federal judges, that can be a really long turnover period. But there is a long incubation period when new theories really have to prove themselves in some way, have to commend themselves to people entering the field.
It’s pretty slow. And you also have to ask, how does it actually reach the desk, first, of the district court judge? Which obviously neither one of us is. And it does so largely because you’ve got good antitrust lawyers who themselves are expert in the economics, and they know how to communicate. So they know how to take the reports that their experts have created and distill them in some way that’s going to reach at least the, well, one hopes the judge. Maybe the judges and law clerks. The judges and law clerks who were more recently in school, perhaps exposed to this thinking in a way. And then further lag before it gets into this.
But remember, I always like to remind people of this, 1.5% of all civil cases filed in the federal district courts are resolved after full proceedings and a jury trial. One point five percent! So, as Frank just said, they’ll be thrown out on 12(b)(6), they’ll be thrown out on summary judgment, they’ll be settled. Something else happens. But they very rarely actually go to the final.
Diane mentioned law clerks, who can in principle bring new knowledge. But many of you have been to law school yourselves. And law schools don’t teach people how to do a multivariate regression. Right? Even the antitrust class that Diane and I taught introduced a little bit of calculus so you could understand the concept of inelasticity. Right? But actually teaching how to do a regression, that whole process of gathering data, of forming the equation, of forming a hypothesis and testing a hypothesis, is taught in grad schools but not taught in law schools. So even among the younger lawyers, we don’t see people who are bringing those skills to the table, either as judges or as law clerks.
Unless you deliberately tried to hire them. Some of them are there in the pool.
Thank you. So let’s open for q&a. They asked me to remind you that they cannot comment on pending cases. So please don’t ask questions about those. Steve, please. Just wait for the mic so people can hear you.
Thank you. Interesting talk. Speaking of Addyston Pipe, I’ve done a lot of thinking recently about out of market benefits, and whether they should be permitted to offset harms in the relevant market. So we know from Philadelphia National Bank that out of market benefits don’t count. In, you know, outside of mergers. Addyston pipe, the harms and benefits were in the same market. And AmEx, it was all in the same market, in the same two-sided platform market. Same in Leegin, it was all within the same market. So I’m wondering, you know, Bork, I think sort of would have counted out of market benefits in his version of the ancillary restraints doctrine. So I was wondering if you guys could talk about that?
Well, I’ll say a few words about it, which is roughly, courts have had a dickens of a time dealing with claims of offsetting harms and benefits, even in the same market. Judges have been very reluctant to say, “Well, we think your merger is going to produce some efficiencies in production. And it’s also going to produce monopoly prices for consumers. And we’re trying to offset them.”
Enforcement agencies have been more willing to consider those kinds of things in deciding what cases to bring. I think judges have rightly been very skeptical of trying to perform that, even within one market. And I would be highly doubtful that we could make things better by looking outside markets, too.
Doesn’t Leegin tell you to do that?
I don’t know that Leegin tells you to do that. So I think the theme that we’re both in our own ways talking about, even though my rules might be a little different from Frank’s, is we need rules that can be administered by ordinary human judges. And I would say, in following that thought along, that maybe the Supreme Court has been a little too enthusiastic to throw away some of the per se rules that it had. They’re not perfect. No per se was ever going to be perfect. But you could work with them and give judges much greater guidance. And if you had a rule that said, “Stick with the in-market benefits,” that would be something that the courts could administer.
It becomes quite complicated to do those trade-offs. I guess I’m not sure which part of Leegin, you know, insists that you do that.
Any more questions? Now’s the moment to ask. So, feel free. Please.
This, I think, picks up on Steve’s question and your responses. You both talk about relatively simple rules. And I wonder if that has, or will lead to, emphasis on either preventing over-enforcement or preventing under-enforcement? That seems to be one of the debates now that, you know, Judge Easterbrook’s famous article, “Limits of Antitrust,” sort of did emphasize simple rules. The courts apparently have focused on over-enforcement. And I think you need complex rules to sort of balance both errors. Minimize the cost of both errors, so to speak. So I wonder where you come out on that.
When a judge condemns a business practice, whether it’s a merger, or something that looks like a cartel or an exclusionary practice, the judge doesn’t think that’s over-enforcement. Right? Just by definition, if the judge thought it was, the judge would have decided the case otherwise. If you’re thinking about over-enforcement and under-enforcement, type one and type two errors, you need to think systemically about what rules you want judges to carry out. And you can put a bias in the rules that makes it harder to catch things, to prevent over-enforcement.
You mentioned my article, “The Limits of Antitrust.” I argued there that if a judge makes a mistake and does not condemn some practice, competition will eventually undermine it. may take time. Think about the rocket business. The federal government wanted the two major US rocket producers, Boeing and I forget what the other one was, to form United Launch Alliance. And the price for the government dramatically went up. The Department of Defense actually wanted that. It took a long time for SpaceX to enter the business and bring the prices back down again. But that did in fact happen. So there was an error of under-enforcement, and a market response. Delayed. Market responses are always delayed. But if the alternative had been something like, “You can’t have any mergers in the rocket business,” that legal rule is very difficult to displace.
There’s already been mention tonight of Leegin, which overturned the Dr. Miles medical case after, what, 103 years? Where it had been, I think, pretty well established that Dr. Miles was not doing economic good for people. And even the dissent in Leegin, Justice Breyer’s dissent in Leegin says, “Well, you know, the majority’s economic theory may be all right, but we’re in favor of stare decisis.”
Markets move slowly at undermining an incorrect failure to condemn something. But they move relatively fast compared to an incorrect condemnation, which is in the world of 103 years. Or, if you try to do something about it, you’ll be put in jail. I would rather make one kind of error than the other error.
I don’t see it quite that way. Because I am very worried about the length of time it takes to fix the under-enforcement error.
I think of the Kodak case, for example, where the Kodak people had clearly a monopoly and were charging monopoly prices on film until, literally, digital. You know, your phone came along as the camera. It was a good century. And I think antitrust enforcement is not that inefficient. And so if we had preserved better competition, I think there was a lot of room to protect consumers from those prices.
And by the same token, I would say, sure, the Dr. Miles example is 103 years. Sylvania? Oh, we took 10 years to overrule Schwinn. It’s like lightning speed by these standards. And so I’m not sure, just because it’s a wrong decision, sometimes Congress acts quickly, you know? Look at the Ledbetter case. I know it’s not antitrust, but…
I wouldn’t want to limit it to Dr. Miles. I mean, think about Brulotte against Thys Co., where one of my colleagues on the Seventh Circuit wrote an opinion needling the Supreme Court, saying that, you know, “The work you’ve been doing in antitrust tells you, you really ought to overrule this case.” And maybe five years ago, they wrote an opinion. Elena Kagan, another Chicago professor, writes an opinion saying, “Well, you know, all the economics suggests that Brulotte was wrong and injurious. But it’s a statutory case. And patent law is not as loosey goosey as antitrust law. So we’re not overruling it.” Now, I think there are an awful lot of examples of what people will concede is erroneous over-enforcement that we can’t, it seems, do anything about.
Frank, have your views changed at all since you wrote “The Limits of Antitrust?”
I don’t want to say I’m like the Bourbons who’ve learned nothing and forgotten nothing. But no, I’m worried, within the limits of antitrust, about what we’ve just been talking about, about the disproportionate costs of error. That markets will undermine monopolies faster than any procedure deals with legal errors. And I still think that’s true. I’ve watched more legal errors. You know, I watched with delight when Dr. Miles was overruled. And I was disappointed when Brulotte against Thys Company was not overruled. I watched with delight when State Oil against Khan overruled the per se rule against fixing maximum prices, which was an outrage of a rule. Right?
So legal rules can change. But I still think there’s this differential stickiness. And that should lead to some accommodation in how we phrase antitrust rules.
So maybe you’ve all got a challenge for the next empirical study. You can check on Judge Easterbrook’s examples and see how it comes out.
I’m strongly in favor of empirical work done by other people for me to read.
So it’s such a privilege to hear from both of you tonight. And I was struck, Judge Easterbrook, hearing about both, so, Coase, of course, and in “The Nature of the Firm,” whom you cited, really talking about, as you said, command displacing competition. And you also mentioned your Adam Smith tie, I think. And Adam Smith was very concerned about corporate privileges and sort of had this idea of a utopia of owner operators. And sort of, the corporation itself was a concern to him. Senator Sherman was really concerned about control. So, all three of these people are concerned about control and domination. And Senator Sherman, I think in the debates, said something about, you know, the problem with Standard Oil was leaving the business and the transportation of the country to a few men sitting at their council board.
And so I’m wondering, to hear from both of you, whether a concern with control and domination, displacing competition, but potentially also just displacing dispersed decision making, which could be facilitated potentially by other areas of law, like labor law, is an appropriate concern of antitrust law? And also how working with other areas of law, if so, could play a role?
Well, I’ll answer that in a generic way. Labor law has goals different from antitrust law. And there are other bodies of law that also have different goals.
One of the problems if you tell somebody like me or Diane or the Justices of the Supreme Court to use antitrust law to achieve multiple goals is that those goals are often incommensurable. I don’t know how much of one you get, to sacrifice some of the other. In general, when you have an agent—and I think of judges as agents of the legislature in carrying out these rules—when you have an agent, you want to give the agent a clear maximand. And if you give the agent multiple maximands, the agent is really in charge. The agent can do anything he or she wants and say, “Well, I’ve achieved more of this, and I’m sorry if you think there’s less of that, but that’s just how it is.” But I don’t think you want a system in which judges can rationalize any outcome in an antitrust case by having too many things they’re trying to achieve. Because then you can never say a decision is wrong.
In addition to the fact that you want to be able to say judges have made an error, you really don’t, in the end, want people who you can’t vote out of office to have the power to do anything they want. I was asked earlier whether I was troubled by the idea of antitrust as common law. And I’m not troubled by it, as long as there is some understood goal that the judges are supposed to be getting at. But if it’s just “maximize the total welfare of society, however you think that should be measured,” you’re in deep trouble.
I guess what I would say is that there’s a distinction, in my mind, between the motivations of the people who voted for this law—Senator Sherman notably, you know, he and his colleagues—and what the coverage of the law itself is. People in Congress have all sorts of reasons for voting for laws. And one of them may well be that if we break up these big monopolists, not only are we leaving room for more people to compete in the market, but there may be this ancillary benefit that the oil business of the country, or the railroads, or whichever trust you want to focus on, won’t be essentially a privately run public utility where the interest of the public isn’t going to be foremost.
So I’m sure that this easily could have been one of their goals. And it’s one of the reasons why I think we shouldn’t be too quick to say that it was the Borkian notion of just high prices, low output, et cetera, motivating the law. I think that’s just too narrow.
But I have great sympathy for the idea that you need clear goals. And in my experience—and I know I’ve given you several international examples but that’s what I did when I was at the antitrust division—the countries such as South Africa, such as Canada, even the European Union, that give you goals that seem to point in different directions, don’t have the same private right of action that we do. And an enforcement agency in the executive branch, or an independent agency as the South African Competition Authority is, and then goes to the tribunal, they can balance these goals. They’re not subject to the objection that Frank is making because if the government changes, there can be different people on the commission. And the system seems to work in a way. And then the judges can affirm anything that serves any of the goals.
Both Frank and I share, I think, an antipathy for, like, 15 factor balancing tests, wherever pop up, because they’re just maddening. No one ever gives you weights on the factors, and you just sort of could flip a coin and be just as principled about it. So in the world that we live in, we probably are doing ourselves a favor at least to keep it in the economic realm. Even though once you’re there, you haven’t finished the debate. You have to figure out, what exactly does economics tell us that we need to control?
I’ll conclude with our final question to finalize. I think it’s a fair summary to say that if you go to the, sometimes in the rules of this discussion on antitrust under-enforcement, the lawyers first point the finger at the economists, and the economists point their fingers at the lawyers, and then both sides agree it’s actually the courts to blame. And so I wonder, how would you respond to that? Would you agree on that?
Courts then point fingers at the lawyers and economists.
I don’t want to accept the premise that there is under-enforcement. It is very difficult to know the optimal amount of enforcement. And it’s not clear. I’m perfectly free to recommend other people to do empirical studies to figure out whether more or less would be helpful, and if so, in what part of the law.
I worry about under-enforcement. And I think, as I said earlier, our move away from the kind of rough, at least, presumptions that this set of things are not troublesome because there’s no market power and this other set of things you need to look at much more carefully, maybe attach quick looks or per se rules or whatever to them. I think that would help. Because when the reason that enforcement isn’t happening, if this is true and I certainly don’t know, but if the reason is the transaction costs of just putting the case together to give this giant mass of stuff. And remember, after Twombly, you even have to have quite a bit to make a plausible claim in your complaint, and heaven help you if you’ve gotten to the summary judgment stage. Of course, these days it’s easy, it’s just a link to wherever you’ve stored things. But there’s a lot of material.
So, if you’re just being drowned with the motions that you have to go through, we don’t have a good system. We don’t have a system that’s looking at the merit of whether this practice should or should not be tolerated.