Below, we have provided the full transcript of the interview with Geoffrey Manne, Distinguished Fellow at Northwestern University’s Center for Law, Business, and Economic Development, recorded on August 25, 2021.
This interview was done as part of the Antitrust Brainstorming Board created by CPI with the support of the CCIA.
Thank you, Mr. Manne, for sharing your time for this interview with CPI.
A video of the complete interview is available HERE.
Do you think the current antitrust framework works for consumers?
Geoffrey MANNE:
I think there’s very little evidence to suggest it’s not working. For example, it’s often claimed that US antitrust should be more like EU antitrust, but it might be worth noting that the European Union hasn’t significantly outperformed the United States since the turn of the century. GDP was once higher in the EU than in the US, but since then, the US has outpaced it. Over the same time period, R&D intensity has increased dramatically in the US, venture capital funding has increased in the US. Economic growth is a multivariate phenomenon, it’s not as though antitrust policy plays a significant part or that we know exactly what role it plays, but those sorts of numbers should at least dispel the myth that implementing a more European-inspired antitrust regime would single-handedly lead to economic success.
I think we have to accept that there’s very little we really know about whether the current framework works for consumers as compared to theoretical alternatives. All the numbers are fuzzy, all the causal claims are weak. But any honest scholar would have to admit we certainly don’t have sufficient evidence to suggest that any particular change in the antitrust laws would improve conditions for consumers. It’s often pointed out that merger enforcement has been lax over the years, for example, and that this has harmed consumers, based on the work of John Kwoka, for example. But Vita and Osinski’s thorough assessment of that evidence basically shows that out of the cases that are assessed to try to support that claim, maybe one of them could possibly show that antitrust enforcement was lax. That’s not a strong basis for claiming that our merger enforcement standards should be changed. Similarly, you hear all the time claims that concentration is too high or markups are too large. And this is all based on studies that exhibit serious failings.
For example, proponents of this view look at national numbers instead of local numbers where competition actually takes place, and that gives a false impression. So Esteban Rossi-Hansberg and coauthors look at the local numbers and find that concentration is actually decreasing. Sharat Ganapati looks at industries where the national concentration numbers do seem to be increasing, but finds that the cause is increased productivity, beneficial growth in firms size, in productive industries, and I’ll quote here, “Expands real output, holds down prices, raising consumer welfare.” These are problems to be avoided.
Could it conceivably be even better? Sure, but we don’t know that for certain, and we certainly don’t know what specific changes would actually make things better. I would just add to that finally, that the political reality is such that even if we thought there was some reason that we could improve economic outcomes for consumers by changing the law or increasing regulation, the political reality is such that it’s bound not to be as optimal as scholars would like it to be. And the outcome may not in fact be an improvement, but a politically-motivated real harming of consumer welfare.
Do you believe the vertical merger guidelines need to be changed?
MANNE:
The previous vertical merger guidelines before they were recently changed were out of date, but I also think we can all agree that they weren’t being used by any courts or really by the agencies. And it’s not clear that they were any kind of impediment to vertical merger enforcement. I guess more to the point, recent claims that we should be enforcing more against vertical mergers and vertical integration, I think are woefully misguided. I do think that’s underlying a lot of the impetus for reform these days. And I think there’re a couple of loud voices that are pushing for that. But while I think you can probably identify some competitors or some inputs or downstream companies that may be harmed, I think that it’s extremely difficult to show that actual consumer welfare is harmed. And of course there’s copious evidence dating back decades showing that vertical integration is almost uniformly good for consumers. And importantly, showing that mandatory disintegration is almost uniformly bad for consumers.
Do you approve of the shift from competition towards regulation?
MANNE:
for reasons I just suggested, I don’t think there’s a sufficient basis to justify a shift toward regulation, or even for that matter, more presumptive legislation. I think in order to justify such a shift, you really ought to have a pretty strong evidence. And as I said, those who point to the claimed evidence of harms, I think are standing on very weak ground. I think what we tend to see, in particular with the calls for common carriage type regulation, are people pointing to some very specific narrow-alleged problems that are leading to advocating for a much broader change to the law, because almost by definition, they say, and they’re probably not wrong, the law doesn’t address those specific problems. But that may be actually how antitrust works best. Antitrust works best when it restrains enforcement against novel conduct, new business models, evolving technology, and the like, allowing some problems to go unaddressed in order to ensure that the non-problems are not deterred or prohibited.
A lot of people are very critical of this error-cost framework today, but as I said, I think those arguments are extremely weak. So the common carrier notion in particular, I think is really inept here. I think of it like this, common carriage was created for situations where carriage was fairly homogenous, like railcars or telephone service. Consider a classic case like a city’s only harbor, maybe ensuring access for all ships on regulated terms might make sense there. Common carriage for online platforms, to me, doesn’t seem to be the same thing as mandating access to the harbor. It’s more like saying, “Anyone who wants to operate a tug boat in the Harbor should be allowed to do so and should be able to do so on their preferred terms.” We’re not talking about customer access, or even for that matter at a very broad level complementor or input access, as we traditionally would with common carriage like making phone calls or a logging company buying transport on a train, we’re talking about opening up ancillary services to businesses, not end users.
And what it bumps up against is the reality that platforms are managed facilities, managed entities. They are simultaneously both open in part and closed in part. They are open in part in order to attract users and complementers, they’re closed in part to ensure that the platform earns revenue and to optimize the platform operation. There’s really no basis in the literature or in evidence to suggest that mandating more openness, under something like a common carriage regime, would improve outcomes. It might improve outcomes for some specific complementors who might benefit from a common carriage-like regime, but that’s hardly the same thing, and I think a really problematic basis for pushing this kind of notion.
How would you ensure antitrust is enforced vigorously if no changes are made to the current antitrust system?
MANNE:
I think it’s really important to note that antitrust is enforced most vigorously and certainly in most substantial part by deterrence, not by specific enforcement actions, by which I mean, counting the number or vigor of enforcement actions tells you very little about the optimality of the enforcement regime. So today, for example, literally today pharma and biotech companies are putting mergers on hold essentially. The current environment in which the whole idea of the killer acquisition originated in the pharma industry is simply too hostile. Now this isn’t the result of multiple enforcement actions that have made it very clear, that have forwarded these potential mergers. This is because of a series of policies, statements by regulators, one enforcement action, the Illumina-GRAIL case that’s going on again, literally today, and a sense of the general zeitgeist.
If you’re just looking at the number of enforcement actions against alleged killer acquisitions, there may be very few, if any. And yet, I think we can safely say that at least in the pharma and biotech industries, at least today, no killer acquisitions are taking place because no acquisitions are taking place. And I just point out that shows how easy it is to find ourselves in a regime marked by overenforcement. Even without bringing a raft of cases, the current administration has effectively deterred all mergers, an entire industry. I don’t think it’s plausible that every single merger that would have been attempted otherwise would have been anti-competitive, therefore there’s a serious risk of overdeterrence.
I don’t think ensuring vigorous enforcement is a problem, despite the common refrain suggesting enforcement has been woefully lax for the last several decades. As I said, there’s very little evidence to suggest that’s true. Enforcers are people who go into enforcement agencies, the staff at enforcement agencies are our public servants who are interested in the program of the agency. They want to bring enforcement actions despite claims that the courts view these things too lax and won’t allow a case to go through that might be a valid case. There’d been an enormous number of cases brought, and even one in the proper circumstances, identifying that the precise number is somehow indicative of an overly-lax regime, is extremely difficult. And I don’t think anyone has demonstrated that. And meanwhile, I think all of those assumptions, as I said, really fail to account for the extent of deterrents here.
What are your thoughts regarding start-up acquisitions?
MANNE:
I’m really wary of this approach. I think it’s worth noting that the most recent data from the NVCA, the National Venture Capital Association, on acquisitions and IPOs showed that last year, there was approximately a 10:1 ratio, if I remember right, I think it was 10:1 or 9:1, of acquisitions of VC-backed companies relative to IPOs. Something like 1,000 venture-backed companies were acquired and something like 100 entered the public markets. So restricting startup mergers would curtail one of, if not the most significant means of exit for funders of startups, and would thus curtail funding in the first place, hardly seems desirable. But now we have to acknowledge that there’s probably some percentage of those acquisitions that might be bad for consumers. Of the 1,000 acquisitions last year, some of them were probably bad for consumers. And maybe we would want to stop them if the bad outweighs the good. The real problem is how would we know? We have virtually no ability to identify the few harmful killer acquisitions ex-ante from among the ones that properly fuel the entire startup ecosystem.
So according to the killer acquisitions article, I mentioned in the pharma industry, about 6% of the studied acquisitions were of the killer acquisition sort. And again, that was just the pharma industry. Let’s just assume for the sake of argument that that 6% applies to all industries. It’s hard to imagine a plausible scenario in which stopping 940 mergers of these VC-backed startups would thwart enough anti-competitive harm from the 60 bed mergers that would be stopped to make it worthwhile. And that’s really the choice we have, you can either essentially stop them all or allow them all, because the one thing that we can’t reliably do is pick out the 6% that seem to be problematic and enforce against those, and allow the other 94% to go forward. I don’t think that’s a good trade-off at all.
Is break-up the best solution for the digital economy and for consumers?
MANNE:
I think breakup is definitely not the right way to go for several reasons. Let’s see – the first is economies of scale. If the thing that makes the incumbent platform valuable is its size, the size of its network, breaking it up would only hurt consumers, 1,000 small Facebooks will not be as valuable to consumers as a single Facebook. By the same token, if the efficient-market structure is a single, or maybe only a very few firms, that’s what will eventually happen anyway. So if you break up Facebook, just using the example, and find that in five years, you simply have another dominant company anyway, because the market tends toward a single firm, what did we gain by breaking it up Facebook in the first place? Another reason I think it’s problematic points back to the vertical integration question you asked earlier.
I think a lot of the arguments around breakup, certainly not all of them, but a lot of them are really arguments against vertical integration, against platforms favoring their own products that they offer in competition with other services, excluding certain input providers because they provide the services themselves. A lot of the arguments about breakups suggest that we should separate these kinds of platforms vertically, but for reasons I suggested, I think there’s very little evidence to suggest that that kind of vertical integration is harmful, and in fact, it seems to be quite beneficial. I do get that some certain complementors or competitors don’t like it, and it’s a compelling political story, but giving into it, I think would straightforwardly harm consumers for the benefit of a few relatively large companies. And then finally, the problem with breakup again, on the assumption that breakup is a precursor to allowing all complementors to have equal access to the facility, it’s making sure that Amazon isn’t selling on Amazon’s platform and precluding some other third-party merchants from being able to sell.
If that’s the intention, breaking up doesn’t end with a breakup, it leads to invasive regulation. I mean, there’s virtually no way to ensure equal access and pro-competitive access in a dis-aggregated regime without imposing some sort of regulatory oversight and often price controls. That’s why we have a Federal Communications Commission, for example. So even if you could imagine the scenario in which a breakup might possibly make sense, it has to also make sense in combination with the ensuing regulatory regime that I think may very well have to be imposed. And I think often that won’t really be the case.
How do you see the role of the FTC and the DOJ in ensuring competition works for consumers?
MANNE:
Certainly they should do what they have always done, which is enforce the antitrust laws. But I think where they may have actually fallen down over the years, is in the extent to which they haven’t taken advantage of their circumstances to help shore up the information deficit I talked about. I think instead of presuming they know what should be done, they should use their subpoena authority, use their persuasive authority, use their resources to try to find out what should be done, collect data, anonymize it, make it available for analysis, convene smart people to actually try to solve uncertainties and problems, try to identify the best solutions instead of just latching on to whatever’s politically expedient. I think even if it were true that the status quo were failing, as I said, we certainly don’t know what the right solutions are to the alleged problem. And that seems like something the agencies could really be beneficial in.
And I don’t mean in the half-assed way, like the FTC here at recent FTC hearings, where you had lots of people, including me pontificating and talking past each other in the hope that their preferences will be adopted or given a voice by the agencies. What there wasn’t, was any systematic aggregation of views, ongoing dialogue or effort to empirically validate any of the claims, use economic models or experiments to try to predict outcomes. It’s a weak first step, and I think if the agencies were serious about finding the right solutions to problems, they could put their resources there. And we don’t have to necessarily live with this troubling information deficit, which as I’ve indicated, I think is an important reason why we shouldn’t take a lot of the steps that are being proposed today. But that also means by implication, that if we could solve those information problems, if we knew that proposal X were better than status quo A, then we could in fact adopt it. But I don’t think we know that now, and I’d like to see the agencies doing more to help us find that out.
How would you reconcile competition and competitiveness? Should antitrust reforms take into account the potential impact on proposed changes vis-à-vis China?
MANNE:
that’s a tough problem. I certainly don’t think that antitrust adjudication, per se, in any particular case, should be taking account of this. I don’t think that a court should say, “Well, this merger is anti-competitive, but because we’re afraid of what it might mean for competition in some industry with China, we’re going to go ahead and allow it to take place.” I don’t think it’s the province of antitrust enforcement or adjudication to sort out geopolitical problems. But it is absolutely the province of the legislature and the regulators to take account of that before they pass any new laws or implement any new regulations. And I think there is a real risk here, and it’s something that should very much be on the minds of legislators and regulators as they’re considering these new proposals. You could argue that the current path is creating a US landscape where there is a very limited viable path to innovation.
The current environment is one that is extremely hostile to patents. You saw the language against SEPs in the executive order rescinding the previous administration’s IEEE letter, various moves that suggest this administration is going to be hostile to SEPs, and perhaps patents more generally. Clear hostility to bigness, which in many cases is a precursor to R and D. I mean, we see these large companies, but we also have to recognize that these large companies are the world leaders in research and development spending. You can’t engage in that kind of spending if you don’t have the size and the cash to do it, you can’t be bought. There’s all of this hostility to mergers.
These are collectively serious impediments to further growth and further innovation, and they could easily hamstring the US economy precisely in competition with China, for example. So, yeah, I absolutely think that this is a valid concern that should and must be taken account of by the legislature and by the regulators. But again, I don’t think it should be an issue for judges adjudicating antitrust cases, or even probably for enforcers, this may be the harder question, but for enforcers bringing antitrust cases. But by all means, it’s something that has not been discussed enough in the context of current legislative reform proposals.
