In response to both Herb Hovenkamp’s February 27 article in ProMarket and, perhaps more importantly, also to Hovenkamp’s highly regarded treatise, Lawrence B. Landman, first, shows that the Future Markets Model explains the court’s decision in Meta/Within. Since Meta was not even trying to make a future product, the court correctly found that Meta would not enter the Future Market. Second, the Future Markets Model is the analytical tool which Hovenkamp says the enforcers lack when they try to protect competition to innovate.


Meta wants to “own the Metaverse,” and to stop it, the Federal Trade Commission said it would not allow Meta to buy Within. But Within does not own the Metaverse, it only makes “dedicated” exercise virtual-reality software, software specifically designed to give its users exercise. “Incidental” virtual-reality software, by contrast, gives users exercise as a side-benefit of the fun it offers by, for example, allowing users to swing sabers. 

Indeed, Meta’s software does this, and thus gives its users exercise incidentally as they swing sabers to music. But Meta makes no dedicated exercise software. The FTC thus sought to stop Meta from buying a company which did not make a competing product—and, most importantly, was also not trying to make a competing product. The FTC, however, claimed that if Meta did not buy Within then Meta would itself make dedicated exercise software. 

In 10 law review articles, which I summarize in “Protecting Competition to Innovate is Protecting Competition in Future Markets: Ten Law Review Articles Leave No Doubt,” I lay out the Future Markets Model. This is the model all competition authorities always use when they analyze competition in a Future Market, a market in which at least some of the relevant products do not exist yet. 

I derived the Future Markets Model after extensively examining the many cases in which not only the American, but also the European, competition authorities have, over the past three decades, claimed they could protect competition to innovate, but in reality protected competition in markets in which at least some of the relevant products did not exist. These are Future Markets. In the United States, the cases in which the enforcers protected competition in Future Markets have tended to be overlooked because, until Illumina/Grail, all the relevant firms acquiesced to the enforcers’ requirement that, to approve the transaction, the relevant firms license one of the R&D programs which was trying to make a future product.

Applying the Future Markets Model to Meta/Within shows very clearly why the court did not agree with either the FTC, or, as I discuss below, with Herb Hovenkamp. Unlike all the other Future Market cases, in this case the relevant firm was not even trying to make the relevant future product. I apply the Future Markets Model to Meta/Within, in detail, in Competition to Innovate and Future Potential Competition. To summarize: 

A. Does a current product exist?

Yes, Within’s software

B. How many firms are trying to develop a future product?

None—Meta was not trying to make a future product.

C. For each possible future product, is it sufficiently developed that the authority will consider it a possible future product?

Neither Meta, nor any other company, was trying to make a possible future product.

D. How broad will the authority define the Future Market? Will the authority consider future products which are similar, but not identical, as future competing products?

Since neither Meta nor any other company was trying to make a future product, the features of the products that no companies were trying to make are impossible to determine. 

Hovenkamp, by contrast, in “Reclaiming the Antitrust Law of Potential Competition Mergers,” uses standard antitrust tools to analyze Meta/Within. Courts developed these tools to analyze markets for existing products. But, as I explain in “Competition to Innovate and Future Potential Competition,” in the cases courts used to develop the two strands of this doctrine which Hovenkamp discusses, actual and perceived potential competition, the relevant products existed. The key question in the relevant cases was whether the relevant firm would expand geographically.

By contrast, in a Future Markets case the key question is not whether any firm would expand geographically. Instead, the relevant question is whether the relevant firm would enter the Future Market. And that depends on whether the relevant firm will succeed in making the product it is trying to make.

To help courts determine when they should protect competition in a Future Market, I developed a third strand of the potential competition doctrine, which I call Future Potential Competition. This doctrine asks courts to consider all appropriate variables when deciding if they should act to protect competition in a Future Market. They should consider, for example, the number of firms that have made, or are trying to make, the relevant product; if only two firms are trying to make a future product, and thus the transaction may give the merged firm a monopoly in the Future Market, the court should be quicker to act than if six firms are trying to make the same future product. For products still in development, they should consider how well-developed these possible future products are, and thus whether they will consider them “products.”

Further regarding Future Markets, the incentives of the relevant firms will almost always be irrelevant. Any firm investing time and money to make a future product will, it is safe to assume, sell that product—if it develops it. And that of course is the key question.

Returning to Meta/Within, Meta was not even trying to make the future product. The court therefore, quite reasonably, found that Meta would not make the product it was not even trying to make. The court thus found, implicitly, that Meta would not enter the Future Market.

Meta/Within is unique among Future Market cases because it is the only one in which the relevant firm was not even trying to make the relevant future product. In fact, one could therefore argue that this is not a Future Market case at all. As shown above, Question D of the Future Markets Model does not seem appropriate for this case.

But the Future Markets Model explains the court’s actions. First, the issue in this case was not whether Meta would expand geographically, it was whether Meta would make a future product. The issue in this case was thus whether Meta would enter the Future Market. And the court said that since it was not even trying to make the future product, it would not. 

Yet, Hovenkamp says it does not matter that Meta was not even trying to make a future product. “The market will find a way,” he says. But that logic applies to a great number of companies, which may make a great number of products, and which may enter a great number of markets. And that includes markets in which the relevant firm is not even trying to compete. It proves too much.

The Future Markets Model also fills the analytical void which, as Hovenkamp recognizes, currently exists in antitrust law’s understanding of competition to innovate. I explain this, in particular, in Private Plaintiffs Can Easily Protect Innovation. In that article, I analyze, among other things, Lucasys’ claim that PowerPlan improperly excluded Lucasys from the Future Market for better software which would help regulated utilities manage their finances. The Lucasys court, while denying PowerPlan’s motion to dismiss, and thus finding that Lucasys made a potentially valid claim that PowerPlan harmed competition in the Future Market, quoted at length Phillip Areeda and Herb Hovenkamp’s Antitrust Law: An Analysis of Antitrust Principles and Their Application. The section the court quotes says among many other important things:

Restraints on innovation are very likely even more harmful than traditional price cartels, which we usually consider to be the most harmful anticompetitive practice.…Deviations from competitive pricing are much easier for tribunals to measure than impact on innovation.

But, as I point out in that article, Areeda and Hovenkamp simply say that the enforcers cannot easily protect competition to innovate. Areeda and Hovenkamp accordingly fail to explain what analytical tool antitrust authorities can or should use to protect competition to innovate. But that analytical tool, as I also explain, is the Future Markets Model. The Future Markets Model is not the analytical tool I say the American enforcers should use. It is the analytical they do use. In fact, they have been using it for decades. And since the rules of logic apply equally as well in Europe, so too have their European counterparts. 

Finally, it is interesting to note that even Robert Bork, who, as Hovenkamp says, fiercely opposed restrictions on mergers of potential competitors, recognized that, in the right case, a court must protect competition in a Future Market. In FTC v. PPG Indus, Bork said that “because of rapid and continuing technological changes…an antitrust court must, of necessity, attempt to predict the future market and the merging firm’s share of that market” [emphasis mine]. While the relevant products in that case did exist, Bork did imply that in fast moving markets, courts may have to analyze markets for products at least some of which did not exist yet. And indeed, since 1986, the ever-growing number of fast-moving markets have forced competition authorities, not just in the U.S. but in Europe as well, to analyze an ever greater number of markets in which at least some of the relevant products did not exist yet. In fact, they have been doing this for so long that one can now look at all the cases and derive the analytical tool they use to do so. 

Articles represent the opinions of their writers, not necessarily those of the University of Chicago, the Booth School of Business, or its faculty.