Last week, the Federal Trade Commission and the Attorneys General of 47 states as well as the District of Columbia filed antitrust lawsuits against Facebook. The Communications team at the University of Pennsylvania Carey Law School spoke with antitrust expert and James G. Dinan University Professor Herbert Hovenkamp about the filings. The interview has been edited for conciseness and clarity.
Comms: Can you please give us a brief overview of the recent lawsuits filed against Facebook?
Prof. Hovenkamp: Both seek to sue Facebook under the antitrust laws. They overlap quite a bit, but they’re not identical. Both of them look at Facebook acquisitions that they allege eliminate competition and make it harder for competitors to grow. They focus principally on Instagram and WhatsApp, although they mention several others along the way.
So the acquisitions are one bag of the complaints and the other bag is exclusive agreements that Facebook made with various developers and creators of applications that are designed to integrate with Facebook. These agreements forbid these developers from competing with Facebook or writing for others who do so. For example, if an application creator or an app maker wants to have access to Facebook accounts or to have its technology available on Facebook, it has to agree that it won’t provide equal access to a competitor and that it will not compete with Facebook.
Application developers who want to access Facebook need to access Facebook’s APIs or application programing interfaces. Those are connectors that a developer needs in order to achieve functionality with Facebook. So Facebook is basically saying, if you want to do business on Facebook, you have to agree not to provide equal access to competitors.
The emphasis is a little bit different in the two complaints, but I would say they’re about 80 to 90 percent overlapping and fundamentally allege that Facebook is a monopolist in a relevant market of a general social networking services. Facebook is alleged to have about 60 to 70 percent of this market. That figure is on the lower end for a monopolization claim, and it’s a market that excludes other big social networking sites like YouTube. Facebook is an aggregation of services, including messaging or posting, video posting, and so on, compared to the narrower specialized services like YouTube, which basically does only videos.
Comms: What relief is being sought?
Prof. Hovenkamp: Both complaints are asking for, number one, an injunction against these exclusivity provisions that forbid application developers from competing with Facebook. Frankly, I think that will be the easier of the two things for the plaintiffs to obtain.
The plaintiffs also want to unwind some of these acquisitions, particularly focusing on Instagram and WhatsApp. That’s going to be a little tougher row because those acquisitions are now six to eight years old. The technology for them has been at least partially folded into Facebook. It will not be easy for Facebook to get those companies back out and leave them intact with a distinct customer base. This is not like the classical merger case where a firm acquires a production facility. One famous case is when Ford acquired Autolite, which made spark plugs, and had to divest it. That would leave the original plant more or less intact.
But intellectual property rights don’t really work that way. If you’re going to spit out WhatsApp, it doesn’t have any established position in the market anymore. So divesting these things is going to be really hard to pull off. And the complaint doesn’t provide any detail for how that’s supposed to be done. I think that’s going to give the district court pause.
Comms: Do you find the claim of improper acquisitions itself to be strong?
Prof. Hovenkamp: It’s extremely strong. And I think it makes a whole lot of sense when applied to future acquisitions. That is to say, this is not limited to Facebook but also to other big platforms like Google, Amazon, Apple. One of the reasons those big platforms have been able to maintain their dominant positions is because whenever some small company emerges that has a great idea that overlaps with theirs, they buy the company.
Every one of these companies, including Facebook, all started out as tiny firms. Some started out in a garage. Facebook started in a Harvard dorm room. These were tiny companies with brilliant owners or developers and some pretty good intellectual property rights. Facebook knows that. The record and the complaint is pretty strong on that. The emails quoted in the complaint show how paranoid Zuckerberg was about the idea that some small startup could grow into a formidable competitor.
If Facebook decides today to buy XYZ and the government goes into court to request an injunction, that’s a pretty simple thing to do. But going after an eight-year-old merger that has been totally integrated – it’s like taking the sugar out of the iced tea. It’s not a very easy thing to do once it’s in there. You can stop somebody from putting it in the first place, but once it’s in, you really can’t take it back out again. And that’s what the government appears to be asking for. But it’s a one-sentence request towards the end of the complaint, and it really has no detail about exactly how the court or Facebook would go about unloading these acquired firms. That is a hurdle I see down the road.
Comms: Taking those difficulties into consideration, what kind of remedies do you see as possible?
Prof. Hovenkamp: Information sharing. I think that’s possible for both Facebook and for the parallel Google case a couple of weeks ago. One of the reasons these companies are big is because they are built on enormous databases of information.
That’s Facebook’s attraction after all – that it’s got all this information so you can search and have widespread friends because everyone else is on it. And if we are sharing that information, it’s a lot less disruptive in the sense that you don’t have to unwind a business or try to recreate a different business. You could have one big information database that multiple competing companies could share.
It would be a little bit like the phone company today. We have a single network for telephones and a very large number of users. In fact, it’s a global network, but it’s not a network dominated by a single firm. Literally thousands of companies own a piece of that network. They make instruments or lines or wireless signaling devices, cell towers, you name it, and all of them have access to the entire network. So if you want to call your aunt who’s got a landline in Georgia somewhere and you have a cell phone in Pennsylvania, you can make that call, even though that call might go through a technology that’s controlled by a dozen or more different companies. And there’s no reason you can’t have something like that for Facebook.
We do it with e-mail. Email is global. That is to say, if you’ve got an Outlook account, you’re not limited to talking to other Outlook users. You can talk to Gmail users as well, and the connection is so seamless that when you start exchanging emails with people, you really don’t even know what program they’re on. We could do the same thing with a technology like Facebook.
Customers could still control their privacy by deciding what to exclude from the pool, but we would get more widespread sharing, and this would improve consumer welfare.
Comms: Do you see any drawbacks to this potential remedy?
Prof. Hovenkamp: The thing about it is that it’s new right now. The remedy that the FTC and the Attorneys General are asking for is a very well-established, old-fashioned remedy. But it was designed for situations where acquired firms were made out of bricks and real estate and machinery.
When you’re talking about purely digital rights, like the stuff that WhatsApp owns, they are intellectual property rights that give it a technology for messaging. It was a superior technology to the messaging technology that Facebook had.
Whatever building that WhatsApp owns or whatever piece of land it was sitting on, those were trivial and they were not important parts of the acquisition. It’s just the intellectual property that makes it so valuable.
But once that intellectual property is folded into Facebook, taking it back out again is – I wouldn’t go so far as to say it’s absolutely impossible. I’m not a computer engineer, but it’s not obvious to me how you unwind that transaction insofar as the firms are already integrated.
Comms: You mentioned the Department of Justice case against Google. Why is there this movement now in antitrust cases against the tech giants?
Prof. Hovenkamp: I think it’s a political alignment of the stars. It’s not just Democrats who are after the platforms, and both of these cases are now being filed during a Republican administration. The Democrats, if anything, are even more aggressive about going after them. When President-Elect Biden takes office in a month, I don’t expect him to back off of the complaints. For one, he doesn’t really have any control over the Federal Trade Commission. He does, however, have control over who the Attorney General is. I think his administration will let them go on as they’re going or he may even try to expand them, but for different reasons. In any event, President Biden himself will very likely not be involved, but it could factor into the choice for head of the Antitrust Division.
Both the Democrats and the Republicans have found it politically useful to go after the big platforms. The Democrats, because they’re concerned about political power, and because they have this fear of private business that’s hurting small business. A lot of literature from the Democrat’s left flank raises these arguments.
The Republican rationales are different. The big platforms are highly innovative firms and highly innovative firms traditionally have been Democrats. If you look at, for example, tech or patents, Republicans tend to line up with the Rust Belt, old heavy manufacturing, chemicals, steel, and that sort of stuff. And that’s why they are strong in areas like Ohio, Pennsylvania, and the South.
Democrats, on the other hand, tend to align with tech, even though they have also been clamoring for these lawsuits.
The other thing that Republicans are unhappy about is what they believe to be anti-conservative bias, particularly by Facebook and Google. They believe these firms either bias results or suppress information in ways that disfavor conservative positions. This has been a big issue in just the last few months.
That may be a rationale for not liking them, but their control of political information is not an antitrust issue. It might come up under the federal under communication regulation; it might even come up under state tort law, but normally informational biases like that are not antitrust issues, and I would oppose expanding antitrust into politics.
Nevertheless, for various reasons, both Republicans and Democrats have it in for the platforms. And I think that’s where we are.
Comms: How does this case compare to the DOJ’s case against Google?
Prof. Hovenkamp: The Google case does not emphasize acquisitions. Both the Google and Facebook cases are somewhat reminiscent of the Microsoft case from the beginning of the century. That case was decided by the D.C. Circuit in 2001, which raised many of the same issues about what you might call contractual favoritism.
The issue with Microsoft was eerily similar to the Facebook issue today. Basically, there was this new kid on the block – Netscape, which offered a very popular browser. Netscape was the first big commercial consumer-friendly internet browser, and Microsoft began entering into all kinds of agreements with program developers, software writers, and so on, that basically said to them, if your program uses the internet or invokes the internet in any way, you’ve got to favor Internet Explorer (Microsoft’s own competing browser) and exclude Netscape. There was a whole series of these various types of agreements.
The technology is different, but if you look at the Facebook complaint, you see a whole lot of the same thing about these app writers who want to be compatible with Facebook, and Facebook will permit them to do it only if they cut themselves off from rival companies. Pointing out these similarities is a good legal strategy for the government because courts are governed by precedent, which means they find it a lot easier to do something that they’ve done before than to do something that’s brand new.
And by using that set of analogies, the lawyers can basically say, look back to what the D.C. Circuit Court did. The D.C. Circuit Court is often said to be the second most important federal court in the country, second only to the Supreme Court. So they can look back and say, see what the D.C. Circuit did; we’re asking for the same thing.
Comms: Last question: How long is this all going to take?
Prof. Hovenkamp: Oh, very good question. First of all, theoretically, you can have a settlement next week. Facebook could say, “OK, we give up, we will do this, and this, and this.” And if the government agrees, that’s the end of the case. It could settle early.
I do not expect that to happen. I expect them to litigate. Both of these are court cases, which means they will go through a motion to dismiss and then they will be in a discovery period, which could stretch for a year.
Then they will have to go to trial. These will be non-jury trials. And actually there’s some talk about consolidating the two Facebook cases. They’re not exactly the same, but it would be efficient to consolidate them. A district judge would make some findings of fact and some conclusions of law and issue an opinion either granting or denying relief or more likely, some compromise. That’s what happened in Microsoft’s case – the government won some issues and lost some issues. But whatever the district judge does, that would then be subject to an appeal. Both of these cases are filed in the District Court of D.C., which means that they would both be appealed to the D.C. Circuit Court, which is the same court that decided the Microsoft case 20 years ago. And then, of course, whomever lost could still go to the Supreme Court.
So this could easily run to three years or so before it’s all done. Like I say, if it settles, then it could end much, much earlier than that.