View all episodes on our Tell Me How: The Infrastructure Podcast Series homepage
In this final episode, we close the discussion on the relevance of mergers for competition policy authorities and focus on the various ways countries implement policy in the face of uncertainty and how they differ from each.
This podcast series is produced by Fernando Di Laudo and Jonathan Davidar.
Roumeen Islam: This is the World Bank’s Infrastructure Podcast. In this third and final episode on Big Tech mergers and competition policy, we continue our discussion on how competition authorities evaluate mergers and implement competition policy.
Before we start today's episode, I wanted to give you an idea of how important mergers are to big tech. So, let me quote some numbers I got from PC mag, a computer magazine. These numbers are large. Let's take the purchase of Fitbit by Google. Many of us are familiar with this easy-to-use fitness tracker, where Google reportedly paid over $2 billion for it.
And it paid over $3 billion for Nest, a smart home product line. Well, you get the idea:
So, let's continue to find out how competition authorities deal with mergers. Good morning and welcome. I am Roumeen Islam, host of Tell Me How. And today we have with us Professor Michael Katz from Berkeley's Economics Department and Haas School of Business.
Among other positions he has held, he was Deputy Assistant Attorney General for Economic Analysis at the Antitrust Division of the Department of Justice and Chief Economist of the Federal Communications Commission of the U.S.
Welcome back to our podcast, Michael!
Michael Katz: Thank you. It's a pleasure to be back.
Roumeen Islam: So, Michael, let's go on to how the merging parties would try to justify their case. What is it that they would have to show?
Michael Katz: So, if the antitrust enforcers establish their prima facie case, then the burden shifts to the merging parties to say, “well, the presumption may look bad for us, but here's a reason that the merger is really okay.”
And one way they would do that is just to attack relevant markets and say the relevant market isn't defined correctly, or the market shares tells a misleading story, say, because you say we have a high market share today, there's a new technology coming in and our market share is about to drop rapidly. And so, looking at concentration was misleading. Or else they will argue that “well, we can't do anything bad as a result of the merger, because if we tried to raise prices, there'd be a flood of entry and that will discipline us.”
And then finally they'll argue that their merger efficiencies, which we've talked about before as a potential pro of a merger, they'll say “well, if you let us merge, we're going to have lower costs, better products, and yes, we're not competing with each other anymore, but we'll be much stronger competitors against everyone else.”
And that's an argument, for example, that was made by T-Mobile and Sprint when they merged in the U.S. and I should disclose there that I testified on behalf of T-Mobile, that there would be very large efficiencies and that the firms would have so much lower costs and better products that even though they wouldn't compete with each other anymore, they would compete more vigorously with AT&T and Verizon, in particular.
Roumeen Islam: We talked a lot about the process in the United States, but how might this approach compare with that of other jurisdictions?
Michael Katz: So, a lot of other jurisdictions do something similar at a very high level, although the legal institutions are different and some of the details of the process are different, but let's talk about the European Commission.
So there, the Commission itself would assess whether it believed there would be harm to competition, similar to what the U S would do. And then it would ask the question and it would hear from the parties about whether there are any merger efficiencies that could potentially offset the loss of competition. So that net competition would actually be stronger. You'd lose the competition between the merging parties, but you would gain the strength from the efficiencies.
And so, at that very high level, it is similar to what the U.S. does. Again, the legal process is different because the Commission makes that assessment first, and then, if they try to block the merger, ultimately the parties can then appeal that decision to the courts. In practice, there are certainly differences in how the European Commission or other national agencies interpret evidence and to what extent they'll give way to efficiencies and things like that. But at a very high level, the thinking is similar.
Roumeen Islam: But, as is the case with all institutional structures, a lot depends on the institutional history of the countries. For example, when the European Union came into being, and eventually when the European Commission's competition authority came into being, they were also concerned about creating a single unified market.
So, I just wanted to highlight that in different jurisdictions the history matters a lot for the legal forms that each authority takes. So, I guess you would agree with that. Or not?
Michael Katz: No, I certainly do agree with that and
And I would say for example, that
I think , whereas for decades, the U S sort of accused Europe of being out of step and old fashioned, now it turns out that in fact that the European view may have been what the future of the U.S., rather than the old-fashioned view. But yeah, history definitely matters.
Roumeen Islam: I just wanted to talk quickly about all the information and expertise that's required to make all these decisions, right?
This must be a complex affair to decide whether or not a potential merger is going to be anti-competitive. So, could you speak a bit about where does one get data on all of this? I can just imagine some of our client countries, it must be so very hard to get the relevant information and even harder to think about the innovative impact on those markets.
So, in the U.S. for example, where do you think there's enough expertise to deal with all of these issues and enough information?
Michael Katz: So, And in fact, there is legislation now to increase the resources they have, but still, they have substantial staff, substantial expertise and resources.
I think the big problem we get to in the United States is, as you get to the court, often a judge will see one antitrust case in his or her career, and as has become evident from our discussion earlier, these things are very complex and we haven't even gotten into some of the most complex things where, you know, the models we're talking about to predict the effects of a merger may be very complicated econometric models.
Typically, the courts just lack the expertise to judge those models and, often very complicated economic arguments are made. So, I think it is
And that's not a criticism in the court. It's just, that's not what they do, it's not what it's set up for. In other kinds of cases there may be a problem even getting access to the information in the first place for the competition authorities. But again, I think for the U.S. it's more about the ability of the courts to process all this stuff that's thrown at them.
Roumeen Islam: Yes. And you can imagine that the ability of the courts to process this, even if they had the information in other countries, in many poorer countries would be even more difficult. So, is there anything that can be done about this problem? Is there any way to make policymaking a bit simpler?
Michael Katz: People have talked about trying to do that. One thing is to try to develop simpler bright-line presumptions. And some of this stuff is against big tech, you can see it that way. If you have something where it says, “all right, you're not allowed to merge, if your annual revenues are above some level, then that's something that a court could apply.”
And so, you could start moving to very simple, bright-line rules. The problem with that is, I don't think that any of us, whether we're talking about the courts or legislative bodies or academicians, really know enough about the overall set of effects in the world, or, all the different possibilities to come up with really good, simple rules.
One of the things I think we should do more, where we can, is instead of simplifying the process, is we should help the courts become more expert. And a way to do that would be to have court-appointed experts. In the U.S., for example, the private parties typically, and also the government would bring in witnesses. They would hire expert witnesses, expert economists, and you're sworn to tell the truth, but the fact is, the witnesses tend to act as advocates for one side or the other. I should say that's different in some other countries in Australia, if you're an expert witness, you have to sign a statement that you realize your obligation is to help the court, not help your client.
But in any case, the U.S. is much more of an adversarial system. And I think it would be helpful for the court to have its own expert and the expert would help them wade through this complex economic testimony.
Roumeen Islam: Okay. Now, let's go back a bit to the question of evidence. What are the other types of evidence that you would look at?
If you can't sit there and, if you're not experts, but you want to just generally look at some business evidence to understand whether or not, the market's going to become anti-competitive. What sort of data would you ask for?
Michael Katz: So, in some cases, what you ask for in data, as we traditionally understand it, you look for example, what sort of sales data that firms have. A lot of firms, I think around the world, for example, have Salesforce data, where they track sales opportunities and, in the firm’s own database, it will say things like “okay, here's who we're competing against, here's who ultimately won the deal, here's what we're hearing from the potential buyer about who's our toughest competition or what kind of deal we have to beat.”
And that kind of information can be very helpful then in understanding the nature of competition. So, if you look at the merging parties, Salesforce databases, for example, or it could be some other vendor, do you see them constantly mentioning each other?
And more generally you can look at business documents and do you see the business's strategy documents and things saying, “look, these are the guys we really worry about.” And it turns out they're naming the company they want to merge with. And if you see that, that can be a source of concern.
And you can also do things like—coming back to, again, more traditional data—if you have markets where there's sort of individual negotiations and bidding, you may be able to get the bidding database, and then you can look for things like, “did the presence of the other firm of these two merging firms, make the firm we're looking at charge lower prices?” You can build an econometric model.
So, there's just a huge range of things you can look at. As I say, ranging from just qualitative business documents and seeing what it is they discuss and do the firms involved, track each other and respond to each other, to having data about how they price. Or when it comes to efficiencies, you start looking at data about how they operate, what their costs look like, things like that. It's just, there's just a vast range that can be both helpful and overwhelming. In U.S. merger cases, you can have, millions of pages of documents and terabytes of data involved. And so, it really can be overwhelming.
In fact, I certainly heard rumors and I believe them, of firms trying to bury the U.S. antitrust authorities in data. Saying “Okay, fine. You asked us these questions. We're going to just send you terabytes of data because we think that's going to overwhelm your system and be useless to you.” And that's for the U.S. antitrust authorities, which are among the best resourced in the world.
There definitely can be problems there that antitrust enforcers can just lack the resources they need to go up against some of these really huge firms.
Roumeen Islam: All right. So, could you name a few reforms? You've mentioned one already, which is to have court-appointed experts and maybe look at, I guess this is not a reform, but you also mentioned that there should be many kinds of alternative sources of data that you might want to look at, but what sort of reforms do you think merger policy would benefit from?
Michael Katz: As you said, I think having court appointed experts, informing the process. In terms of the data, the reform I'd like to see is less attention on formal market definition. It's using the same information that would go into market definition but focus more directly on the question of how strongly do the merging firms compete with one another. What does the evidence say? What does the evidence tell us about whether there are other firms that they would continue to compete with after the merger? So, get less hung up on how do we put things in this market definition box of in or out.
And just, as I said, more directly say, “what does the evidence tell us about how strongly these firms compete with each other and how strongly they compete with third parties?” Another thing I would do – it is not specific to the U.S. entirely, I think it's an issue that comes up in other countries as well – is that it's very hard to block a merger where you say, “okay, the firms don't compete that strongly against one another today, but we're worried that in the future, they would compete with each other very strongly and the merger is going to stop that from happening.”
And that's the sort of concern people have, as we talked about, with Facebook's acquisition of Instagram and WhatsApp, the concern was not, “oh, these firms are competing really strongly.” They weren’t competing really strongly at the time of the merger with Facebook and social networking, they weren’t.
The concern was, "oh, by buying them up, you've stopped them from emerging as competitors.” And so, I think , to block a merger on the grounds that it's going to harm potential competition versus actual competition.
Roumeen Islam: I was going to say that basically what I hear you saying is that -and, of course, in the developing policy world, we feel this every day- that
And then on top of that, you need judgment and experience in order to figure out what the right answer might be. So, I understand that. Now, one thing that I realized we hadn't really talked about, and that was the vertical mergers, because most of what we've been talking about are horizontal mergers.
So, if you could just say a few words quickly about vertical mergers. And are those always legal? Because they could lead to company structures that lead to reduced competition in the future, right? So, I'm just wondering, are they always legal or not?
Michael Katz: All right. So, let me step back just in case there are people listening who are wondering what the difference between a horizontal and a vertical merger is.
It turns out vertical mergers are not totally well-defined. But So, you could think of, Google search service, and then Microsoft Bing search services. They're both offering search services, they're competing against each other. Or two automobile manufacturers.
Another merger I was involved in was with AT&T and Time Warner, where Time Warner was creating content and then AT&T was distributing that content.
So, they weren't competing with each other directly. One of them was supplying product to the other. And the sorts of concerns that vertical mergers can raise—and in fact were raised in that particular case—is, “Well, if I have, say, control, own, the content and the distribution, when I compete against other distributors, am I going to withhold the content from them?”
I say, “wait a minute, why would I let you use my content to then compete with my distribution arm?” So, there is this concern that there can be this harm to competition, and it's been a long-standing concern. I was in the U.S. Department of Justice, I guess, in 2001. And we used to look at vertical mergers and worry about them.
And I think in the end we didn't challenge any, cause we thought it was just too hard, but I think
Roumeen Islam: Thank you, Michael. So, we've actually covered a lot of ground and I want to thank you for all of that and for revealing to us how complex it is to actually do competition policymaking right.
But thank you very much. Actually, is there anything else you'd like to add before we end?
Michael Katz: So, I would say two things. First of all, I would like to thank you because I've really enjoyed this, and I am grateful for the opportunity to speak to you and to your listeners.
But the other thing I would say is, “look, it is extremely complex.” As you said, there's a lot of matters of judgment. And that's why I think And there are international networks and their basis for doing this is for the different agencies to learn from each other and to share expertise.
Roumeen Islam: And to collaborate, generally.
Michael Katz: Yes.
Michael Katz: Thank you.
Roumeen Islam: Well listeners, what did we learn today? Firstly,
The history and culture of a nation influence the presumptions held, the interpretation of evidence, and the final outcome. Secondly, But developing countries, in particular, suffer from capacity constraints. To deal with complexity countries may adopt rules of thumb or bright line assumptions, such as just limiting the size a company can grow to. Even if such methods have shortcomings. And sometimes courts appoint experts to help them.
Thirdly, Actively participating in international networks is a way to do this.
Thank you and bye for now.
You can find more information about the podcast on https://www.worldbank.org/en/news/series/tell-me-how. If you've got questions or comments, we'd love to hear from you. You can also find us on all popular podcasting platforms.
This episode was released in October 2021. Don't forget to subscribe and thanks for listening.