Below, we have provided the full transcript of our panel discussion Mergers & Acquisitions in the Digital Economy: Striking the Right Balance. Read below to see the timely discussion where a panel of experts deepened the discussion regarding this topic, and how it specifically relates to Australia.
Kirsten WEBB:
Well, hello everybody and welcome to this panel session. Today we’re going to be discussing the fascinating topic of mergers and acquisitions in the digital economy, which is a topic which is much discussed, of course, at the moment, but our focus is going to be on striking the right balance.
The panelists we have here today, to introduce, we have Chris Berg, who’s a senior research fellow at RMIT Blockchain Innovation Hub. We have John Yun, associate professor of law and director of economic education with the Global Antitrust Institute. We have Sean Ennis, professor of competition policy at Norwich Business School at the University of East Anglia, and we have Rob Nicholls, associate professor at the University of New South Wales. And I’m Kirsten Web, a partner at Clayton Utz and I’m the moderator for today’s session.
So welcome everybody. We might start off with a broad question, which is an obvious question, given our topic, which is, does there need to be a re-balance of mergers analysis in the digital economy? I might ask each panelist to give a very short reaction to that question, and then we’ll dig deeper into some of the issues that are raised by that question. So in alphabetical order, we might start with Chris. What do you think, Chris?
Chris BERG:
Thank you so much for having me. Look, I think it’s absolutely a necessity that we rebalance the way we think about competition in this space. I come at from a very particular place, which is the study of blockchain and the crypto economy.
What we find uniformly across policy areas is that many of the assumptions that we’ve applied to traditional industrial era economic organization just don’t apply, don’t make sense in the crypto and blockchain space. I think that replicates across the economy as well, as we move from the sort of factory organization of the economy to these platform network de-hierarchical organizations that we’re now seeing. I think that necessitates a fairly deep rethinking, not necessarily of competition policy, but of the way we structure our thinking about what looks like anti-competitive conduct and what looks like, even things like consumer protection.
WEBB:
Thanks very much Chris. John, what’s your perspective?
John YUN:
Thank you, Kirsten and thank you for having me here.
It’s a complicated question, and obviously, we’re looking for short responses here. So to properly address it, we need to really look at two questions, one of which is where we are and where we think we need to be and whether there’s a difference between those two. I think there’s a lot of policy debates about that, but I’m of the view that I like where we are. I do think modern antitrust and its evidence-based approach and the goal of economic efficiency is something that has been very good for antitrust. Doesn’t mean that the enforcement along that objective is perfect. No, it certainly doesn’t mean that. But at the very least, I think the objective is right.
So then the question is, is that enforcement getting there? Does the digital sector present unique challenges that we need to incorporate and are we deviating from what we consider to be good enforcement? That to me is a harder question. I think the evidence is still sort of being developed.
I looked at a series of paper coming out of Europe by Lantham and Lamech, and there’s a lot of good papers out there, and their conclusion is it’s really hard. They say it could be anti-competitive, it could be pro-competitive. We just don’t know. The evidence is really mixed, and it’s case by case. To me, that’s the conclusion that I think makes a lot of sense here. So, absent extraordinary evidence, I tend to think that, well, we are headed in the right way. I think generally speaking, merger analysis is in a good place.
WEBB:
Thanks, thanks very much, John. Sean, what’s your perspective?
Sean ENNIS:
Oh, this is really a hard question. If you were to back up and say, “Have there been mergers that seem to have been problematic?” I think a lot of people would say yes. But if you go a little bit further and say, “What percentage of mergers have been problematic?” If, let’s say there’ve been 400 mergers and people tend to talk about between five and 10 as being problematic, that’s a really small percentage of the total. I think some agencies have upped their game in this area and so, to me, that shows that there is some ability, at least for the agencies, to change the way they act without necessarily having a fundamental rethink of the broad system. So, maybe I’m in the camp that remains to be convinced, but could be convinced.
WEBB:
Thanks, Sean. And Rob?
Rob NICHOLLS:
I think, ultimately it’s the consumer welfare issue, the driver. Why efficient markets? Because that leads to great outcomes from a consumer welfare perspective. So, is there a problem? Is there something that’s going wrong at the moment that means we should have the sort of regulatory angst that we’re seeing across the world from different regulators? Now, I’m not quite sure that we’ve got to that stage yet. Just like Sean, I’m willing to be convinced, but I’m certainly not going to be convinced if it suddenly looks like a very targeted approach to mergers analysis, which has that “not invented here” aspect to it.
WEBB:
Well, thank you very much for those short answers to a very big question.
We’ve got a range of interesting perspectives on our panel this morning. Let’s dig a little bit deeper into your thoughts on some of the issues that underlie the brief comments that have just been made. The first question, I suppose, is is merger analysis—or Chris who’s looking at most broadly, I think competition analysis in the broadest sense—something different to analysis of traditional industries? What should merger analysis in the digital economy look like, do we all think?
So let’s put that out there for discussion or, Chris, if you want to go broad, competition regulations in the digital world. Maybe if we start with something a little bit concrete.
John, would you like to kick off by discussing the CMA’s move to block Facebook’s acquisition of Giphy with horizontal and vertical theories of harm? Having a look at that, and more broadly, what factors do you think agencies and courts should be considering here?
YUN:
That’s a great question. So let me caveat my thoughts by saying the CMA, they obviously have evidence that is not public and that we’re not privy to, and so they might have very, very good reasons that we’re not seeing for bringing it, and so my comments are with that caveat. After spending time at the FTC, often, the non-public stuff is very important.
So with that being said, for their theory of harm to really work on a horizontal level, let me start there and then move to the vertical, they need a couple of key factors. I think the main thing they need is that Giphy needs to represent some differentiated product from others that are similarly situated.
How do you identify that? I think there’s a couple of ways that you can do that. One of which is, let’s look at their history and their product development. So Giphy, I looked it up, there are about an eight-year-old company, and so they’ve been around a while. One of the things I think the CMA should be looking for, and if they haven’t, it’s useful, is that have they differentiated their product in a way that’s different from what they’ve developed previously? Because an eight year old company isn’t quite nascent and potential anymore. They’re a fairly established company. I would like to see some evidence that they are moving and pivoting into an area that Facebook is competing in. I think that would be really important.
The second thing I’d like to see is how differentiated is Giphy from other sort of GIF type companies that offer a similar product? I think that’s really critical. Are they at a growth rate, the different types of consumers are they closely aligned with Facebook users more than others? I think that’s really the kind of stuff that I want to see, rather than, “Oh, they’re something that Facebook uses and they could potentially become a competitor.” I think we need that seed of competition that could potentially grow, and maybe there is that. I think that would be an important thing that I’d like to see. Very briefly for the vertical theory where they could foreclose the GIF input into others. I think if this were relied on, it would worry me more, because if, for example, an upstart, a Tik-Tok or Snap uses the GIF inputs that Giphy provides, and it’s been a very important driver for some of their features and functionalities that compete with Facebook, and then now there’s a potential for foreclosure, I’d be worried.
So I think that type of stuff would be a lot more viable than the horizontal potential stuff. I think you need to see more evidence. Not that it’s not viable, but I think it’s a harder sell based off of GIPHY’s history, at least publicly.
WEBB:
Thanks very much John.
Did anyone else have any comments on the horizontal and vertical theories and factors that you think might be taken into account in merger analysis here?
Maybe we’ll move to a related question or a more specific or a subset of that broader question, which is, killer acquisitions, and John was questioning whether a company that’s been established for eight years represents nascent competition. Killer acquisitions. Rob, would you like to have a bit of a discussion about the effect of merger regulation or merger analysis which tries to identify mergers that might stifle emerging competition and what effect that might have more broadly, for example, on innovation?
NICHOLLS:
I think one of the key issues for potential innovators—I work in a business school, so I’ve got bunches of students who are the potential entrepreneurs of, well actually, next week rather than next year. A lot of them have ambitions, which are pretty much like this: we will set up a business, we will establish it, prove its success, sell to Google, live with the restraint of trade that Google imposes and use that as the seed funding for the business that we really wanted to get into.
So if you get into a merger’s analysis that says every potential acquisition by the GAFAM or some list of businesses has to be reviewed in a particular way, then there’s a risk. There’s a risk of stifling innovation in that, that exit process is no longer available and it’s less not being able to exit the first time, it’s not being able to go on to the next innovation that’s risky.
I think there needs to be a balance. I think that issue of nascent competition is really difficult to define in a way that can be appropriately enforced by mass competition regulators, and to have a rule, which is “I’ll know it when I see it” is entirely inappropriate because it takes away predictability from businesses, it takes away the potential for innovation.
Actually, I think some of the acquisition of Giphy is really, well, it’s that angst I mentioned earlier on; how did we let Facebook acquire Instagram? Well, because Instagram had 13 people working for them, and actually 13 exhausted people who were really just looking to have a mechanism to patch up their servers to continue providing the service to their users.
So, there is a risk of if merger, there’s a shift to merger analysis that says, “Well, actually we require the onus of proof on the acquirer to be such that there is no possibility that there will be a reduction in competition, let alone a significant or substantial reduction in competition.” That risk is that there will be an ecosystem of innovation which is stifled. I don’t think that that risk is, that risk needs to be at least appropriately considered before you start saying, “Well, we don’t have fit-for-purpose merger regulation.”
I think killer acquisition’s is a particular issue and if you say, “It’s only the GAFAM’s that we worry about for killer acquisitions”, that’s even worse. If you really need a theory of harm for killer acquisitions, that actually can be translated from a theory of harm to regulatory guidance, which is not going to stifle innovation in the digital economy, and to be frank, innovation in the digital economy is where the vast majority of innovation is occurring. The risk is really high.
BERG:
Can I just elaborate or expand and support Rob’s point for a moment there? Because I think that’s really important. That exit to a large competitor, to a dominant player in the market, is built into the capital structure of the original innovators themselves. A large part of the venture capital model is built around that, that ultimately in five, 10 years, we will flip so that we can return funds to our liquidity providers as well. When we’re thinking about regulating it, if we’re going to regulate that exit, we’re also going to be starting to implicitly regulate the entry.
Now we’ve seen over the course of decades that it’s increasingly difficult to go the other direction to go to an initial public offering, for all sorts of complex regulatory reasons, and so the exit to a large competitor looks like a lot easier and more attractive path for a lot of young innovators. Just to reiterate Rob’s point, just thinking about what this looks like from the beginning of the innovation process is really important because these are fixed lifecycle firms, fixed lifecycle startups very often.
ENNIS:
Can I just jump in? I agree with John’s point that there’s a backwards logic to this so that if you change the end game you’re going to change the start of the whole process as well. It’s a little bit unclear how they would finish out, though. I’m not convinced that it’s always healthy to have the sellout to the very big competitor versus going to the market and getting the price as the end objective. I think some venture capitalists are kind of concerned actually about the current situation, because they have so few exit options as it is in the way that things seem to be going. But if you were to look for some criteria that a competition authority might use for evaluating nascent competition, it could be related to the price being paid for the acquired entity compared to its revenues or compared to its size of its labor force. I think if you use those criteria, you will end up with most of the mergers that people have talked about as potentially problematic coming out as having at least a yellow blinking light. So maybe there is a way, Rob, to have such a criteria. I’m not suggesting those ones necessarily, but those might be ones to consider.
NICHOLLS:
I agree. Unfortunately, I haven’t seen yet a concrete proposal that actually says, “Well, we think that there might be a potential problem if acquisition prices N times revenue.” Because it won’t be N times earnings, because earnings will almost certainly be zero, but at above N times revenue, then we should be thinking about it. But even then, coming up with a, well, why is N N and not N minus one or N plus one? So that isn’t a solution. It needs to be a much clearer solution, which says, “We’ve come to this. We’ve come to it rationally, and why have we come to it?”
The other issue, I think, Sean, is you’re dead right. It’s an optionality issue. There’s no requirement to sell out to a big company and an IPO as Chris pointed out has difficulties, but it’s an option. It’s the removal of the option that’s the risk.
WEBB:
Chris, you’re nodding. Do you have any ideas as to how to identify with a greater level of certainty, which of these sorts of acquisitions might be problematic?
BERG:
No, and I’m really concerned that if we try to head down that way, we’re going to basically be restructuring the venture market at the beginning. I spend a lot of time working with companies at the very start of their venture process when they’re getting the initial evaluations, when they’re getting the seed or even angel investing, and any of the proposals that we could put up and we could devise right now, would inherently mean that they trade at a discount because we’re just writing off or we’re at least, we’re making quite risky, one of the basic exit strategies. So I’m deeply concerned about that and particularly because if you think about some of the tech acquisitions that haven’t come to the attention of competition regulators in recent years, there aren’t revenues, there’s certainly not profits. Often, these are just acquisitions to buy a team.
For example, I’m thinking of the acquisition of Keybase, the secure chat app by Zoom. Zoom purchased them at the start of the pandemic in order to basically buy their encryption skillset. There was no revenue, in fact there was nothing that we could have looked at to have some objective function that would be valuable from a policy analysis perspective, but that is a perfectly justified exit strategy and that would have returned to the investors of Keybase exactly what they were looking for. Now, I’m a bit disappointed because I really liked Keybase, but having said that, I’m very concerned about the changing of the capital structure of the industry sort of really strange and backhanded way through competition policy.
WEBB:
Thanks Chris. John, you were nodding. Have you got any ideas as to how to grapple with this question?
YUN:
I don’t, but I do want to affirm Rob and Chris’s point and also raise a statistic I find interesting.
I recently read that of all the exits in the US from startups and venture capital firms, big tech represents 4% of that. To me, 4% is significant. It’s still important and I agree that the ex-ante and ex-post influence each other in terms of the exit strategy and the valuation, and I’m a firm believer of that. But it also puts it in perspective, and saying that big tech is not the one buying up all these companies.
There’s this perception that startups are really just getting gobbled up. It’s really a relatively small percent. That doesn’t mean it’s not significant and important, but I think that statistic is quite helpful, and I think both points can be right where the exit strategy influences the ex-ante valuation. I also think that perhaps it could be overstated how much big tech is gobbling up these small startups. I think there’s a vibrant activity out there that doesn’t involve big tech.
WEBB:
Thanks very much, John. As a lawyer, I’m going to ask now a principles based question, which is, I think back to you, John, actually. Which is in the discussion that we’ve just had, we’ve referred to potential competition and nascent competition.
John, do you think there’s a legal and economic difference between those two concepts?
YUN:
I do, and I’m going to argue that it’s important. In the US, potential competition is a doctrine that arose out of the 1960s and 70s in a series of Supreme Court cases. I don’t want to bore the audience with a walkthrough of case history, but very basically, these involved products such as natural gas and beer.
Beer in the 1970s was different from today. I don’t mean to suggest these are the same things. I know craft and differentiation is very important today, but back then, these were very basic products. What the court was wrestling with is whether a firm who hasn’t quite entered a geographic market or a product space, but is thinking of entering, exerts some competitive influence either currently or very soon in a way that it’s disciplining the market.
So a merger between a potential competitor and an actual competitor can result in some harm to consumers. It’s a very viable theory. It’s a well-developed doctrine. The problem is, is that it represented a very high burden of what it meant to be a potential competitor. Many people have been upset about this, and I think rightly so, but I think it’s appropriate for potential competition cases that are very basic commodities and products.
Now let’s move to nascent competition. This is a doctrine that came out in the 1990s, I think for the first time by the DOJ with their Microsoft case. Very briefly that case was involving the operating system of Microsoft and the threat that Netscape as a web browser posed, not for Internet Explorer, but for Microsoft’s operating system Windows in of itself, in that it could develop functionalities and overlaps that could chip away at the monopoly power that Microsoft had over the OS system market.
That actually has come true to a way, right? Because look at Chrome and Chromebook. The functionality of web browsers has influenced the market for operating systems in a meaningful way. The point being is that nascent competition represents, often, a product that is differentiated technologically, but could develop in such a way that it could represent a threat to an incumbent in the future. It generally isn’t a product that’s not in the market or anything. It could be in the market, or it could be in an adjacent market, but it’s really about product development and innovation in a certain manner.
So why does that matter? I think to me, that burden should be lower. We shouldn’t be using the potential competition cases to inform us about nascent competition cases, because then you’re going to require on the plaintiffs, these arguments that we need to see documents at the entry would have occurred. We need to see firm commitments. We need to see business plans. You see this in certain cases: FTC vs Steris. FTC brought this as a potential competition case, when to me it’s really a nascent competition case.
Long story short, I think these are two separate doctrines, and I think the problem that agencies and courts are having right now in the US is they’re using a very high standard of potential competition to inform nascent competition cases, when I think these are quite different.
WEBB:
Everyone’s nodding. Anyone want to add to that? Rob, you want to speak?
NICHOLLS:
I actually want to go off on a deviation, which John mentioned craft beer as being distinct.
Craft beer actually looks like the sort of innovation that you see in the digital economy. In countries like Australia, where that potential competition doctrine isn’t nearly as strong, craft beer is just another beer in the market for beer. Now, talk to any beer drinkers, and they would entirely disagree with that analysis, but it leads to that issue of if you’re going to have a doctrine, if you’re going to have an enforcement approach, you need to be very clear as to how it works and you need to be able to have it be more general than just, “Oh, it’s this part of the digital economy that we’re really worried about.”
Potential competition could arise in the financial services sectors, as Chris’s experience shows. It could potentially arise in the pharma sector. So saying that mergers analysis needs to be specific to the digital economy, I think is problematic. I think that I’m just jumping on an example that John gave to actually make that point.
WEBB:
Thanks very much, Rob. Chris, you were nodding vigorously then.
BERG:
Only to fully support everything Rob said.
WEBB:
Well, Chris, while you have the floor, so to speak, we have been talking about the issues that have been raised by some of the high profile big tech mergers that everybody knows about.
But there are some interesting measures in the crypto space, I understand, not having any expertise in that space myself. Would you like to talk about whether, or start off a discussion about whether token mergers raise any antitrust concerns, and if so, what are those concerns and how could that be addressed?
BERG:
Look, it’s a really interesting issue. In the crypto economy, we’ve got a lot of parallel institutions and a lot of assets that look kind of like traditional assets, like shares and various things, but also function in a very different way and imply different governance rights and participation by the community over the nature of the firm or protocol or application or what have you. We’ve got these decentralized networks. The key thing to understand about these crypto and blockchain technologies is that, for the most part, if they are correctly decentralized, if they are properly decentralized, then there isn’t a single agency, there isn’t a single mind that makes decisions about what to do with the protocol or project. There isn’t a management, there isn’t a hierarchy that makes decisions about, well, we’re going to go in this direction or we’re going to go in that direction. It is, in an ideal state, developed by the community. The community tends to both own and use these protocols themselves.
With all that, it looks often a lot like companies, and we’ve got instances of planned mergers and acquisitions by different projects. There’s a project called the Fei protocol, that is proposing to merge with the Rari protocol and proposing to do a sort of token swap, and we won’t get into the details about that, but it’s the users and governors of these two protocols who are debating and trying to decide this.
This creates some really interesting problems just to look at it from a traditional mergers acquisition space. Now, competition regulators haven’t gone into this yet, because, A, it is very difficult. And B, it is pretty marginal of course to the rest of the economy. My prediction is that it’s going to get more and more important and interesting, but the key thing that I think is interesting from our perspective is when we’re blurring the boundaries between who owns the firm and who uses the firm’s products.
If a community decides to merge with another community, even if that might materially harm the competitive landscape of the crypto economy. Well, does that raise competition concerns? I don’t have an answer to that. I’m not sure that you can have an answer to that because the presumptions are so different from the traditional sort of crypto economy, but it is a really interesting thing to think about, and I suspect it’s going to become more and more significant as time goes by and more people are being onboarded into cryptocurrencies and blockchain.
WEBB:
Thanks very much, Chris. Has anyone else thought of that? Yeah, Rob has, of course.
NICHOLLS:
Sorry, I don’t want to monopolize things.
I think one of the key issues that flows from that is the issue of gun jumping. So, in traditional mergers analysis, one of the issues is unless and until your merger is cleared, you can’t operate as if the merger has occurred. Actually, if you’ve got two communities which are owners and users that are negotiating a potential merger, actually that negotiation is, for example, going to get to adjustments in rates between the exchange values of the tokens. In effect, even those discussions create some of the effects of the merger.
How do you deal with gun jumping is sensible policy. We’ve used it traditionally in mergers and mergers analysis. I think it starts to raise some really difficult issues that look sectorally specific. That takes away from my argument, “No, no, all this needs to be across all,” but I think we do need to start to recognize that once you start to get into these, the complexities, let alone the complexity of the merger analysis itself, then there are going to be issues that flow from the traditional way of dealing with mergers, such as the prohibition on gun jumping once the negotiation process is in place.
BERG:
It’s an interesting point because it would be very difficult to prevent those sorts of discussions, and if you wanted to prevent them, who would you be preventing from those occurring?
The question always becomes in these crypto and blockchain conversations is it’s all well and good to say, “Yes, this regulation applies in this place.” And we can make all sorts of arguments or you might need a license to do something, or you might be prevented from doing something else. It’s very hard to figure out how to enforce it, though, not least because you can’t necessarily identify who has been driving these things, but also because once they’ve been done, so once the merger has been effected between these two protocols, if it is, it’s unwindable. You can’t un-unwindable. You can’t roll it back. We see this across the crypto economy, just as a digression, a lot of people have argued that certain products on the blockchain should require various types of financial services license, in Australia particularly. That’s all well and good, but what if you were to withdraw a license for failure to comply with its terms? Well, what would you do? The thing’s still on the internet; can’t be taken down. It does create this field does create a whole lot of interesting regulatory challenges.
WEBB:
Well, on that note, we might move to our second topic, which is having acknowledged that there are a range of regulatory challenges in theories of harm or in competition policy or underlying analysis, let’s move to whether we think that different rules are needed or are justified for tech businesses or for particular parts of the digital economy, or overall. We might go to Sean to open up this topic.
Sean, do you think it’s just to have a different merger standard only for big tech or only for particular participants in the digital economy?
ENNIS:
I think that it’s commonly accepted in some countries that you might have a different standard for abuse of dominance based on size. I definitely understand the argument. But for mergers, I’m a little more concerned, to be honest, because I think it’s absolutely possible to come up with some economic indicia that would distinguish one group of firms from another. It might be the GAFAM center, it might be GAFAM plus, but you could come up with the indicia. You could have a policy rationale that these are the ones that you really care about, but I’m not sure you get the right treatment, and in particular, a just treatment and a fair treatment if some companies are singled out in merger analysis.
When you have smaller ones who are doing substantively the same type of deal from all the, in terms of gaining, perhaps gaining similar, if you could measure market power precisely, let’s say they were gaining the same amount of market power as the GAFAM plus, why should the GAFAM plus deal be stopped and the others not be stopped? I’m at a loss to figure out what the difference should be. Maybe that’s because I was raised working on a lot of smaller merger cases in which a lot of the concern was to set a standard that would apply to a broader set of merger cases, and so you wanted to establish the principles that would be generally applicable. I think this is moving away from having generally applicable principles, and that’s a concern for me.
WEBB:
John, what do you think? I think there are explicit calls in the US to bifurcate antitrust into two rules, for big and for everybody else. What do you think about this?
YUN:
Yeah, there are calls and you know, my position, I really couldn’t say it any better than what Sean just said. I call it discriminatory antitrust, that may be a little provocative, but we do engage in some of that, right?
If you have market power, or in Europe, abuse of dominance, that means there’s sort of a different set of presumptions and burdens placed on you, and that’s perfectly fine. That’s a market-based approach to analyzing firm behavior. We use a different lens if you have market power relative to if you don’t and that makes a lot of sense.
I think the bigger concern, however, is when you use discriminatory antitrust, not based off of market-based reasons, but size or identity. I think there you run into some issues, and as Sean said, you could have a firm that’s a medium-sized tech company that is the market leader in a certain area engaging in acquisition that would be under a rule of reason, but then you would have a Google or a Facebook who could be a distant second or third and looking to catch up to the incumbent who’s leading the market and that would be certainly a more scrutinized, if not outright banned, and they would have to vertically integrate in order to compete in an effective way. I think that’s really becoming more regulatory than enforcement of antitrust laws. I think that worries me and if we move in that direction, we need a good evidence that that regulatory approach would lead to better outcomes. I don’t think we’re there yet.
BERG:
Sorry, to jump in. Sort of taking that approach also just badly over politicizes the antitrust process, because Sean was talking about GAFAM plus. Here we have, I know it’s not your phrase, Sean, we’ve got a selection of companies and we can just add different ones at a given time. If you are the marginal company. If you are, say, IBM, wanting to stay out of that list of targeted firms, you’re going to invest a lot of funds in influencing the political process and people are going to invest a lot of funds bringing you into that political process as well.
I think it does exactly what we don’t want from antitrust policy, which is re-politicize it; take it away from this base level of analytics that we’ve been talking about here today and push it entirely into the political process. Very dangerous, as I see it.
WEBB:
Rob, what so you think?
NICHOLLS:
I agree and I think it’s that important distinction that yes, if there is abuse of market power or issues which are associated with market power, those are completely different from the mergers analyses.
For example, you might have, let’s take a traditional business, you might have pipeline operator which has pretty much a monopoly in one state. A pipeline operator which has pretty much a monopoly in another state. If the two merge, well, there is no lessening of competition, certainly no substantial lessening of competition, but you wouldn’t prevent the merger just because they’re two big pipeline companies. I think you need to think very carefully about saying, “Well GAFAM is different or GAFAM plus is different.” That politicization, which Chris worried about, I think one of the worries is that the plus bit turns out to be, “Yeah, well you’re a plus if you’re a US domiciled multinational and you’re not plus if you’re a European domiciled multinational,” and that becomes problematic in the analysis of mergers on a global basis.
WEBB:
Well, I think everybody’s agreed that having specific merger rules that apply to entities just because of their size is problematic for the range of reasons we’ve discussed. Everybody’s in favor of a more market-based analysis. That’s been referenced through most people’s responses to the market dominance or abuse or misuse of market power regimes that exist anywhere in the world.
What do you think about a notion that you have different merger rules for someone who has market power?
NICHOLLS:
You could potentially have a reverse onus of proof, but the risk, so that is you have to show that the merger will not lessen competition to whatever threshold your jurisdiction requires. The problem is that that has the effect of slowing down merger processes in a dynamic environment.
The risk is that unless the enforcement, the actual mergers analysis by the competition authority, is very efficient, that the mergers themselves will never occur, and the economic benefits and the consumer welfare benefits will never flow because the signal is, “Ah, well, you’re big, so we’re going to really look hard at you and we’ll spend months and months and months doing it, so please don’t bother,” which is the effective signal that comes from this. I think we’ve seen in other jurisdictions, perhaps the issue that says, well, if there is a sensitive sector, then potentially there is an expectation that merger clearance flows in a different way. But if you’re going to take that sort of approach and have exceptions from a normal or usual process, are the resources to the competition regulator to ensure that that process is expedited, or at least follows a timeline which cannot be, does not have clock stops from the regulator’s side, but might have clock stops from the merger parties’ side.
WEBB:
Anyone else have a view? Sean has a view.
ENNIS:
Yeah, I think there have been some proposals to both flip the standard of proof and reduce checks and balances on the competition authority, and I think if you put those two together, that’s a very dangerous combination, because effectively you might end up with the administrative authorities, the prosecutors, having both a stronger ability to stop deals and less control over themselves to make sure that they were acting in a reasoned way.
I’m really concerned by that type of development. I wouldn’t want to see that, because I think that administrative authorities tend to try to push the line when they can. They may feel like if given this kind of liberty, they might just push the line too far. It might be that there’s a low risk with pushing it too far. That might be the argument they would use. But I think it’s really important to have a backstop that’s demanding accountability from the authorities at all times when they do act and so that must be maintained if there is any change in the standards.
WEBB:
John, what do you think?
YUN:
No, I don’t have anything to add. It’s a terrific response, I’m just listening now.
WEBB:
Well, looking at this from a different perspective, we’ve referred to GAFAM, of course, because everybody does. But let’s take a look at Spotify as an example of the issues, I suppose, that we’ve been tossing around. Tell us what you think about whether or not Spotify should face the same standards of scrutiny, having embarked on a number of acquisitions as we’ve been discussing as a possibility for Facebook and Google, et cetera. What do you think about that?
YUN:
I think Spotify is a really good example of the dangers of what we’ve been sort of discussing in terms of having a different set of rules based simply on the identity of a firm.
I just did some looking into Spotify and they’re a very big company. They’re not part of the elite of the big tech companies, but they are the clear number one leader, in streaming music subscription services which is a big market and Apple, Amazon, and Google are certainly big players, but they’re playing catch up. I looked at Spotify’s acquisitions over the past couple of years, and it was eye-opening how many they’ve made. Just to give you a broad list, it’s Findaway, Podz, Betty Labs, Ringer, Megaphone, Anchor, Gimlet, Cutler.
I would read their actual press releases, and Spotify was very unabashed at saying, “Look, we’re buying this to jumpstart our entry into podcasting, our entry into audio books, we want to be the premier destination for everything audio in the digital economy.” They have a very strong ambition, and good for them.
It actually has worked. Statistically, they were behind Apple and Podcast, but after these acquisitions and obviously other investments they’ve made, they are now, according to Spotify, so this could be in dispute, they are the number one destination for podcasting globally, currently. Here we have acquisitions that have allowed Spotify to compete with Apple in an adjacent market, but some could recast this. What if this was Apple instead of Spotify? I think our lens would be totally different. I think the public pressure would be completely different on the agencies and the courts and our perception of these acquisitions as quote, “Good,” for consumers would be very different.
That’s my concern. Here we’re swapping the identity of a firm simply for actual market-based analysis. I think to me, Spotify is a good example of a very important company, that’s a market leader, that’s using acquisitions in a way that many would accept as pro-competitive, but then we drop those principles when we switch out the identity and I think that’s what worries me rather than, than having a market-based sort of fact-based approach to these types of deals.
BERG:
It does worry me sometimes in the mergers and acquisitions arguments that we don’t fully recognize that these mergers can be extremely pro-competitive, because of course they are taking not just off Apple, but Spotify is moving into audio books. Audio books has been dominated by Amazon, and it’s Audible thing, and it’s actually very pro-consumer as well. If you look at the shape of the podcasting industry, as it was when it was dominated by Apple, it was actually very difficult to sort of expand audiences outside the Apple ecosystem. It was very directed at one particular side of the cell phone markets of the iOS of the OS market and so forth. This has actually massively and quite radically transformed podcasting.
I used to be a podcaster so I’m really, really engaged in this, and I’m super pleased that as both a consumer and producer of podcasts, things are just more accessible, and I think that’s going to happen for audio books as well. That’s an unambiguously, pro-consumer merger story and unambiguously pro-competitive merger story.
WEBB:
All right, just before we move from the topic of whether we need different rules for different types of tech businesses, we’ve spoken a lot about the different merger rules, but, Chris, frontier technologies, do you think that there are different rules necessary to address any potential antitrust problems that might arise in frontier technologies?
BERG:
Look, thank you for indulging me, I’d like to talk about blockchains and cryptocurrency again. Again, I think a lot of things operate in a different, in a lot of different dynamics, but there are some interesting examples of what we’re talking about a lot is digital infrastructure and shared digital infrastructure across multiple firms.
For a lot of the use cases that we think are most exciting about blockchain, say tracking products across a supply chain, it’s very important that you get, if not the whole industry on board a shared piece of infrastructure, certainly a large part of that industry. So a lot matters about, well, we’ve built the shared infrastructure and we’re all going to sign up and we’re all going to use this standard or this smart contract platform, or what have you. How are we going to govern that smart contract platform? Who are the participants in governance going to be?
We’ve seen a number of really substantial infrastructure plays by for example, IBM. There’s shipping; the shipping company Maersk coming together and building out a trade lens product that they’re trying to onboard everybody onto. They’ve found that very difficult because it’s been seen as a sort of pro-Maersk play within the shipping industry. At the same time, if we’re going to get the gains that we’re going to have to standardize. We’re going to have to share infrastructure.
I predict in the next couple of years, that’s going to come really to the forefront of competition and analysis. Certainly anti-competitive behaviors. Some competitors might be working through the governance of these networks to keep out some of their competitors as well. Really interesting area, a lot’s going to ride on what constitutes shared governance, what constitutes genuinely decentralized. But I think everybody in this call’s going to have to tackle this issue sooner than you think.
WEBB:
Yeah, well, that sounds like a huge topic. Rob, go.
NICHOLLS:
No, I was just going to say, Chris says shared governance, somebody else says cartel conduct. It’s a very difficult issue to deal with.
WEBB:
We might move away from that topic, which we obviously would need years to talk about, back to merger analysis and remedies, and the sort of enterprises we’ve been discussing operate across boundaries, of course, there’s, and we all come from specific national regimes, which have some similarities and have some differences.
What do we think about the notion that we might need some kind of global approach; some kind of realignment that might be driven by who? We might start with, we might start with Rob and then move to everybody else to see what they think about it. A non-national based approach to merger analysis or, Chris, broader analysis of these sorts of enterprises.
NICHOLLS:
I think there’s the potential for having a globally coordinated approach. Even though, as you say, Kirsten, that you’ve got jurisdictionally specific legislation, there is the underlying rationale for why there needs to be a merger analysis and merger control is broadly common, at least amongst the OECD countries and almost all countries which have some form of competition law or antitrust laws. So the potential is there. One approach might be to use the leverage of the OECD, but I think the risk of using the OECD is who it excludes, rather than that the OECD isn’t the right place to do it. The OECD’s competition analysis, their support of competition regulators is great. But it is a formal international organization.
I actually think that the more likely way in which you’ll get at least de facto standardization, even if it’s not a formalized standardization, is through organized, I was going to say organizations like the International Competition Network, but there is no organization like the ICN, so I think the ICN has the potential for doing it. The issue there is the ICN can’t make rules. It can’t, but it can get to an approach where actually, if you look at most of the people, most of the competition authorities that engage in dawn raids, you’ll see that actually the learnings that they have, have been shared through the ICN, more than they have through the OECD or any other body. So, if there’s going to be a level of harmonization, having an informal body such as the ICN, seems to me a preferable approach. It will allow jurisdictionally specific legislative changes, if they’re required, without having the, “Well, we’re not going to do it because, just because the OECD says it’s a good idea.” So that actually sounds much more like a blockchain approach. The consensus that you can get through the ICN has the potential for leading to much more efficient outcomes, because the economic rationale can be agreed, but on an informal basis.
WEBB:
Sean, what do you think?
ENNIS:
Yeah, this is a tough question to come. Just for you to know it before I answer, I worked at the OECD for a long number of years, so I’ve thought about this from that perspective, also. Just quickly for Rob, the OECD has bodies associated with it that don’t have exactly the same members as the OECD and the OECD competition committee has an association status, which is equivalent to membership of that committee. So, there are ways that it can work. There are ways that WTO could potentially do something that’s very different from its current process as well.
I think if you really want to have a guarantee of consistency, you do need something more than promises of cooperation. You need an enforcement mechanism that would somehow have laws and procedures that were uniform everywhere. That would be a suggestion that you might want to have some global powers.
On the other hand, I think if you look at digital markets, in particular, there may, there is an argument that remedies can be much more easily customized to particular geographic areas in the digital space than for other types of products. Just as one example, it’s not a merger example, but if you want to disenroll from a service that you’ve signed up to, that’s being provided over the internet, if you’re in California, you can do that often, much more easily than elsewhere.
One of the reasons is that there is a specific law in California that makes the companies have a better process of disenrollment, and that’s the type of thing to think about on the national side. Another one is that if you have national remedies, you don’t move the whole world towards one specific outcome, which might not actually be the best one. You have much more opportunity for experimentation and discovery among regulators before you move, and then perhaps some evidence gathering, before you move to perhaps a common view of what the best outcome is.
WEBB:
John, what do you think?
YUN:
This actually reminds me of when I was growing up in Savannah, Georgia. My science teacher said, “Oh, you better get ready, we’re moving away from this Crown English system to the metric system. It’s going to be outdated and we need to harmonize and science has been held back by this dual system.” Well, here we are today, we’re still in pounds and feet and inches and all that and we seem to be getting by and certainly language is another thing.
I know those are quite different, but, ultimately I do think there is value in having different differentiated jurisdictions trying to get at the right answer. I know there are problems as well, but I think a lot of learning can be developed from each sort of system that’s out there.
I do think that there are problems with harmonization because if you’re a company, they’re not so excited about the marketplace of ideas and all these different rules. It might be intellectually stimulating, but as a practical matter, getting a deal through across all these jurisdictions, or defending a practice in multiple jurisdictions is exhausting and it’s costly and it’s resource intensive.
So, initiatives that Rob and Sean mentioned, the OECD, ICN, I think those are important. How important should they be? Those are obviously different questions. But I do like those initiatives, I do like the sharing of ideas. I do think that there is some sympathy that we should have towards moving to some harmonization. I don’t know how, I haven’t thought about that problem deeply, but I’m certainly open to the thought that we do need better harmonization.
BERG:
Also the politics of this is quite difficult, so it’s interesting viewing this from an Australian perspective. In Australia, there’s a move, certainly on some sides of politics, to assert a sovereign control over the large tech giants. So there’s this drive and we got this from the requirement that Google fund newspapers and Facebook fund news outlets and so forth, but more generally there’s this idea that for things that affect Australians, we should be able to have Australian public policy makers make those choices.
Now that’s all well and good and that sounds really wonderful if you’re an Australian policymaker and maybe it sounds great as an Australian consumer, but often it looks almost comical in practice. I’m thinking particularly of the ACCC, Australia’s competition policy regulator, trying to impose its supervision over Google’s acquisition of Fitbit.
Fitbit, an American company, Google or Alphabet, an American company, and the ACCC was trying to hold it up and check that it was okay under Australian law before it went through. This has got a very much sort of a “we warn the czar” type vibe to it in Australia, and it just really emphasizes to me how in these globalized markets…
Yes, it might not be possible to harmonize competition policy, and we might not want that as intellectuals interested in seeing discovery and innovation in this space, but at the same time, it looks kind of silly in practice if all these sovereign nations are trying to impose their own individual sovereign regulatory frameworks over tech mergers.
WEBB:
Thanks very much, Chris. Well, we only have a couple of minutes left. We could continue this conversation for a really long time, obviously.
I’d like to invite each of our panelists to make some, whatever closing remarks they’d like to make. Might go back to alphabetical, but reverse, so might start with you, Rob.
NICHOLLS:
I think the key message that I have is the one that I’ve made before. It’s really important in any changes to merger control which are looking to be sector specific within the digital economy relating to a small number of businesses within that digital economy, to be certain that any changes don’t stifle innovation. Innovation in the digital economy is the driver of innovation and the driver of productivity in national economies.
WEBB:
Sean?
ENNIS:
I would just say that I think a lot of tools do exist for competition authorities to deal with digital deals. There’s a question of how large one has to be as a country in order to successfully address some of the international deals and to not to be ignored. But I think there’s a lot of opportunity for behavioral remedies that could be reasonable on a national level. The tools are there. They can look at in a very fact specific way at deals. I think some of the deals people talk about were ones that one might’ve been concerned about at the time with the right facts. Just what matters for me is having the appropriate process and a fair ability of both firms and make sure the regulators or competition authorities get the right information and have an opportunity to input before there’s any final decision.
WEBB:
John?
YUN:
My final theme is really that we need to be focused on dynamic incentives rather than static ones. I think this harkens on what Rob said in that innovation is the key driver of wealth in societies. This view that we need to look for short run gains, which I think we often fall into a trap of, and it gets heavily regulatory. It gets to be breaking up companies and various things like that. We need to really look into the dynamic incentive effects of these various policies, and I think that will drive us into the right direction.
WEBB:
And Chris, the final word.
BERG:
Completely agree with all that’s been said. The only thing I would add is that we’re innovating at such a rapid pace, not just on the types of things that we do in the digital economy, but also in the way that we organize economic exchange. We know this. We have now spent the better part of a decade or more understanding and being involved in the gig economy. These restructures of corporate organizations. Looking at the technologies that are coming through and the technologies that we’re going to be considering over the next decade or so, that restructuring of economic organization is just going to become more and more evident and more and more significant, and it’s going to force, I think, a rethink of what do we mean by consumer harm, what do we mean by market dominance when we look at it from a regulatory perspective.
WEBB:
Thanks very much.
Well, thank you all for such an interesting discussion. We’re going to have to leave it there, we’ve reached our hour. A whole lot of interesting ideas and thoughts that can be further developed. A discussion that could go on for a very long time, but we’re going to call it a day now. So thank you very much to each of our panelists.
Thank you.
