Posner’s Pontifications on Patents

by Stephan Kinsella on April 16, 2023

I’ve long been disappointed with the Federalist Society’s approach to intellectual property—patent and copyright. They say they are pro-free market and pro-Constitution but their bibliography is missing any principled, radical, informed case against IP.¹ Typical commentators include the pro-IP Richard Epstein and other law professors, federal judges, pro-IP Objectivist Adam Mossoff, and the like. On occasion you’ll here someone say maybe we should consider re-thinking parts of IP law, but almost never do you hear a principled, free market case against IP (surprising since Cato’s Tom Palmer was on the cutting edge of anti-IP theory in the late 1980s). In fact Cato has had articles in support of the TPP even though it would foist US-style IP rights on the rest of the world.² A recent Federalist Society symposium, Is the Patent System Working or Broken? A Discussion with Judges Posner and Michel, exemplifies this. Let’s leave aside the remarks of the host Adam Mossoff, and the confused comments of federal judge Michel (who has argued that increasing the amount of patents issued by the USPTO would increase employment!),³ and focus on the opening remarks of federal Judge Richard Posner. Renowned as a big brain, a law and economics guru, law professor, author of numerous books — I was astounded as I listened to this podcast. The Austrian-aware libertarian surely has to squelch the desire to mutter or curse every 90 seconds or so during this panel discussion, and Posner’s comments are no exception. I had Posner’s remarks transcribed and below offer some comments—not exhaustive—on them. I am beginning to think that only patent attorneys understand patent law: legislators, federal judges, juries, economists, engineers, non-IP lawyers seem to have a very difficult time understanding the nuances of patent law or the differences between patent, trademark, copyright, and trade secret law. This is understandable, as IP is such a highly specialized and arcane field. But, if you are going to defend the statist quo, or defend the patent system, you would think you should know what you are talking about, no? With no further ado, here are Posner’s remarks, with a few of my comments interspersed:

Okay. Thank you very much. I want to correct something. You said I had extensive adjudicative experience with patent law. Well, Judge Michel’s court has the monopoly on appellate adjudication of patent cases. I have, however, well, I actually sat with Judge Michel for a couple of days several years ago as a visiting judge in his court. But also, for several years now, I’ve been doing patent litigation as a volunteer in the district court, as a trial judge. And I’ve also done academic writing on patent law. So I think patent law, our patent laws, have failed our system. I think too many patents are issued.

Kinsella: Too many? What’s the right number?

And I think, when they are issued, too many of them are upheld by the courts. I think our system, for the administration of patent law, is not a happy one. The patent and trademark office, which issues the patents, is not adequately staffed or managed. I don’t think the patent examiners have proper incentives. Of course, the validity of a patent can be challenged in court. And, of course, infringement suits are brought in court.

Kinsella: The problem is not the quality of the PTO or that there are “too many” patents issued, or “too many” uphold. It’s not that we have not tweaked the incentives properly. If the most powerful state of modern times cannot get this optimized, who can?

But I don’t think our court administration of patent disputes is good at all. Part of the problem is expert witnesses paid by the parties. Part of the problem is the use of lay juries in patent cases. And part of the problem is the specialized appellate court, Judge Michel’s court, the federal circuit, because I think it’s had a tendency, which you actually expect in a specialized agency of any sort, of taking a kind of promotional view of the activity that it regulates. Especially since the federal circuit was created at a time, 1982, when there was a lot of concern that the United States was losing its technical edge to Germany and Japan. That, of course, you know, from today’s perspective, seems absurd, but that was the thinking. And so I think the federal circuit has been too welcoming to patents, has relaxed standards for patentability. The result is we have too many patents.

Now when you have too many patents, it means that you are providing inventors with more insulation from competition than they need in order to have adequate incentives to create inventions.

But the stated purpose of patent law, in the federal legislation, is to incentivize disclosure of information, not to incentivize innovation.⁴ Nevermind that it does not even do this.⁵

You have market prices above efficient levels. You have distortions in the allocation of resources. You have wasteful patent races involving a lot of duplication of effort. You have filing of defensive patents. People who don’t really want the patent but they worry that, if they don’t have a patent, they’re going to be sued for infringement by someone else. And there, of course, the patent trolls who buy up large numbers of patents for the sole purpose of extracting license fees by threat of suit and, if necessary, suing. And I’ve dealt with trolls in cases and I’ve dealt with other plaintiffs as well. So not a happy picture. Now, it’s a mixed picture. I don’t want to overstate my position. There’s the pharmaceutical industry which is the poster child for the patenting.

Nevermind chapter 9 of Against Intellectual Monopoly.

I’ll come to that in a moment. Patent races, although they are wasteful, they do accelerate invention.

Do they? And waitasec—if it does incentivize innovation, why is it “wasteful”?

And the requirement that patents be published does provide a lot of valuable technical information.

Really? Most inventors need to disclose key facts as part of bringing a product to market; for these innovations, the disclosure would have happened anyway. For others, the inventor can still keep it as a trade secret.⁶

And a related point, when patent protections aren’t available, inventors will turn, to the extent they can, to trade secrecy as a way of protecting their invention from competition. And by definition, of course, trade secrecy doesn’t involve publication and so limits the dissemination of technological information.

Except that it’s difficult to keep an innovation secret when you sell your product (opening it up to reverse engineering) and tout its features and benefits in market slogans.

To provide a little bit of an analytical framework for discussion of the optimal amount of patent protection,

optimal?

it’s useful to evaluate patent protection as follows. You consider the cost of inventing and the cost of copying the invention. And the higher the ratio of the cost of inventing to the cost of copying, the stronger the case for a patent protection. And the ratio is very high for pharmaceutical drugs.

Except, … it’s not. See Boldrin and Levine, ch. 9. Sigh.

The cost of inventing a new drug, in particular including the cost of the extensive testing that is required for a new drug, can be in the hundreds of millions of dollars. Yet, for most drugs, the cost of copying, of producing an identical substitute, is very low. Most, not all of course, but most drugs consist of sort of commodity chemicals. So production cost is very low. And what this means is that the inventor who spent these hundreds of millions of dollars is going to find it very hard to recoup his expenses without patent protection because copiers can duplicate the invention, produce the invention, at a very low cost and underprice him and he’ll never get his expenses recovered. So that’s the strongest case.

No it’s not. The purpose of law is not to permit entrepreneurs to “recoup their expenses.” This Chicago mentality is so confused and disastrous. See Hoppe’s criticism of similar “Coasean” reasoning in the Chicago Diversions section of “The Ethics and Economics of Private Property,” in The Great Fiction.

And there are other industries that have these characteristics, but actually very few. And in most industries there really is no need for patent protection in order to provide an incentive for a new invention.

Many new inventions are inexpensive. They are ground out by teams of engineers who are employed by a manufacturer or what have you. They’re pretty well paid but it’s not a big deal. They’re salaried employees whose business is invention. Even without any sort of patent protection, the first firm to invent a product usually has significant protection from competition in the near term which is often the only relevant term.

And that’s what the legal system of a free market is for, right—to make sure people are protected from competition!

So he gets a head start. He’s first to the market. He gets a head start on moving down his cost curve, right? Experience as he makes the product demonstrates ways of cutting costs and improving the product and the first in a market. Unless he is followed a day later, he is going to have this head start. That is very valuable.

How valuable, exactly, who can say? But federal judges and law professors know, they know, that it’s somewhat valuable!

Also, the public is quite likely to identify his brand with the entire product and keep buying the brand even after there is competition because he has gotten it … people have associated his brand name with the actual product. Also, many new products have a very short shelf life. Not that they’re going to wear out, but, you know, they’re fashionable. They’re fad products. They’re going to be replaced. And the inventors and producers of such products don’t need twenty years of protection.

They don’t need 20! They only need … well, something less. Just give ’em what they need, for Chrissake!

So where you see the problem of patent protection where it’s not needed, it’s the software industry. I don’t want to say that it’s never appropriate for software,

I mean we would not want to take a principled stance, or go to extremes!

but I think generally it isn’t necessary and it’s just a clog on competition and a source of wealth for trolls and trial lawyers and expert witnesses and the like.

On the other hand, you know, sometimes it’s good to clog competition and feed trolls and trial lawyers and expert witnesses and so on. You know, gotta balance things out.

So this is a very progressive industry. There’s very rapid invention. Most of the invention is incremental.

Hey, I’d a federal judge and author of books on law and economics—we just know these things.

It’s created by teams of software engineers, usually at modest cost.

Hey, we’re just talking “usually” here; and “modest” covers me.

These inventions are usually very quickly superseded so that, when you give twenty years of patent protection, what you’re really doing is creating an enormous mass of stuff to search for in the patent office and traps for the unwary.

We only want to have traps for the wary, in a free market, you see.

This software innovation tends to be piecemeal. It’s not an entire device; it’s components. So that a cell phone, a tablet, what have you, may have tens of thousands, even more, separate components, each one of which may be patentable,

So…. software patents cover … hardware components…? Hunh.

resulting in huge patent thickets, creating rich opportunities to try to hamstring competitors by suing them. And also for incentives to infringe, then challenge the validity of the patent when the patentee sues you. And that sets the stage for some kind of settlement, some reasonable license fee.

Look, there’s nothing wrong with a “reasonable” license fee. Who could object to that? The guys just want to wet their beak. That’s all.

Also, as I understand it, the patent office has a shortage of patent examiners with the requisite technical skills. The judges and jurors are often at sea trying to understand technology. The difficulty of assessing damages for infringement of a component, rather than a complete product, is really acute because you have to somehow figure out what value do the consumers attach, not to the product, but to one little component of it, one of the many functionalities of a software device. And because the software industry, the internet related industry, the computer related industry generally, has historically been a very unstable, you know, companies rise and they’re immensely successful and some new device is invented; now they’re bankrupt maybe. A lot of, you know, major firms have disappeared. So that makes these companies very anxious, both to get patents and fight patents. So pharmaceuticals and software, those are the extremes in terms of how much patent protection we want and where we want it. But my sense, and there is extensive academic literature on this, is that, in general, we have too much patent protection. I’d go farther, although it is not the subject of our discussion, and say, in general, we have too extensive intellectual property law. We’re too hung up on the evils of copying. So that completes my opening comments.