from the no-one-to-blame-but-yourself dept
There is a lot of sturm and drang in the halls of government these days about corporate mergers – or, at least, tech company mergers (oddly, this ire doesn't seem to necessarily extend to all mergers). But despite all the gnashing and wailing there's not a lot of understanding of why they happen. Which is strange, because if you think there's a problem, it would help to understand WHY there is a problem, because that understanding will give clues on how to fix it.
So let's think about why a company "merges" with another. I put "merges" in quotes, because usually it boils down to one company buying another – how much of a "merger" it is depends on how similarly positioned the respective companies are and the details of the deal, but regulators today seem most upset about the A part of M&A (acquisition) so let's focus on that aspect. Why would a company want to acquire another?
One big reason relates to patent law. Let's say you're a company with a product, and you want to make that product do something more, or better, or have some new feature that it doesn't already have. You could develop it on your own but (a) that will take time you may not have (ex: the market opening may close before you can get it out the door), (b) money you may not have (ex: you may not have the liquidity needed), or other resources you may not have (ex: you may not have the expertise needed or be able to easily hire it), and (c) even if you had what was needed you may still not be able to develop it on your own because it turns out that someone has already developed the best method, gotten a patent on it, and now they can block anyone else from implementing it with at least the threat of litigation if not also actual litigation.
So the shortest distance between two points for many companies, especially larger tech companies, is often to simply buy the other company that has the missing piece of the technology puzzle they want. This acquisition then does a few things. For one, it gives the purchasing company access to that technology, which means it could potentially produce a better product. Of course, it also gives the company exclusive access to the technology and lets them block anyone else from using it, including their competitors.
In other words, through patent law (and also copyright law, but we'll set that aside for now) we intentionally give companies the power to act like monopolies, even when it's not actually in our interest. So of course we're upset that companies use that power, but the problem is that it's a power we gave them. Splitting them up is not the cure for the problem we created; the only solution is to stop giving them so much monopoly power in the first place.
There are at least two things we should do differently. First, we need to stop giving out so many patents full stop (and we need to stop condemning the people calling for fewer to be issued). Too often these patents are not for actually significant innovations (or innovations at all), and all too often they are on subject matter that should be unpatentable. Every few decades or so the US Supreme Court wakes from its slumber to remind the world that, at least under US law, software is not patentable subject matter. But these decisions haven't stopped people from pursuing, and getting, these sorts of patents. So it's very strange that people wonder why there are so many tech companies with software-driven products that have so much market clout, when it's a power our own USPTO has been purposefully giving them.
But even where patents are issued appropriately, there are still things we can do to mitigate their anti-competitive effects. One of the problems with patents today is how they give patent holders the ability to shut out other users of the technology. That's why patents can have this harmful effect on the marketplace, and it's also what has put modern patent law out-of-step with the authority granted Congress by the Constitution to pass a patent law at all. Congress gets to legislate in this area for the purpose of "promot[ing] the progress" of science and the useful arts, but the reality of today's patent law is that instead of promoting progress it ends up creating huge obstacles to it.
This injunctive power that comes with a patent is also unnecessary to achieve anything that patent law was intended to vindicate. Even accepting as true the idea that innovators need some sort of reward for being the first to innovate something the world would benefit from – beyond, of course, the inherent market advantage that comes from being first – all that means is that if there's some profit to be had from the innovation that the patentholder should get to realize at least some of it. But you don't need the power to shut out all other uses to glean that profit; all you need to do is license it.
Of course, backed with the power to enjoin other uses, license fees today are less about reasonable market rates that provide benefit to everyone: the innovator, the implementer, and the public, which now gets to have more innovation in the marketplace at prices the market can bear. Instead, patent revenue today is more about extortive windfalls. The policy change we need is to switch up that balance. And one way to do that is by replacing the current (and often disproportionate) ability of patentholders to enjoin any uses of their technology with some sort of compulsory license system. A compulsory license system means that patent holders cannot say no to competitors and other innovators who want to use or build on their technologies, either directly, by refusing permission, or indirectly, through excessive license fees. Instead the reward for their patent is the reasonable income returned by the license they must offer.
There are several upsides to changing patent law this way. For one, even if it somehow diminishes the perceived luster of having a patent that would not be a bad thing: as explained before, the landgrab that has been trying to turn every technological improvement, no matter how small, into a powerfully enforceable monopoly has been at the root of much of the anticompetitive behavior regulators now lament, and discouraging it would, on its own, help mitigate those problems. (Constraining the Patent Office so that it also grants fewer patents, especially specious and/or software ones would help as well.)
Secondly, it also means that more people can use the technology, or even build on it. And that's good for society in general. The point of patents is to stimulate that innovation, and this change would do so by clearing the way to it. Furthermore, it would also have the effect of diminishing the monopolistic effects we don't like. Not only would patents now provide less monopoly power, but they would also lessen the incentive companies currently have to acquire other companies in order to horde more of it.
Which would also lead to less market consolidation. For instance, smaller companies with a sought after-innovation, instead of being bought out by one company that could now exclusively benefit from it, could stay going concerns and continue to put products in the market. If the innovations were legitimately patentable they could also use those licensing profits to subsidize their own further innovation and product development, and to the extent that fewer innovations may be patentable, the good news is that this reduction in patentability would mean that there would be more technologies available for them to help themselves to in order to compete, even against the companies we currently worry are too big.
Of course, there is a catch: compulsory license systems are great in theory but often cumbersome in practice, and, as we see in the copyright space, they can introduce new, unwelcome, and debilitating costs and regulatory impediments. (We'd also want to keep an eye on where non-practicing entities owning patents should be in this ecosystem, if anywhere.) So this isn't a case of "just add water" where tacking any old compulsory license system onto patent law will automatically make everything sunshine and roses. It will take some extremely careful thinking in how to implement.
In the meantime, however, we are seeing some other industry adaptations, like patent pools, emerge to help mitigate the extortive power of patents. And, in general, the idea of minimizing the exclusionary control of a patent, including through compulsory licenses, is a good one we would be better served to be thinking seriously about, rather than the zealous appetite to break up companies that has currently seized all of our attention. Especially when these proposed break-ups are so arbitrary, unprincipled, and ultimately costly in ways regulators do not seem to be contemplating.
In any case it just doesn't make any sense for the government to on one hand tell companies to go be monopolies and then immediately complain they are being monopolies. The solution to the problem of companies acting monopolistic is to not deliberately give them so much power to be.
Filed Under: big tech, consolidation, monopolies, patents