After exploring the dangers of monopolies, the Capitalisn’t podcast investigates a situation in which the government actually works to create monopolies on purpose: the patent system. In the first of a two-episode series, hosts Luigi Zingales and Kate Waldock examine whether the US patent system is helping or hurting capitalism.
Luigi: We’ve done a series of episodes about how bad monopolies are and how the government should fight against the formation of monopolies. Now, I want to dedicate a bit of attention to a situation in which the government on purpose creates a monopoly, and that’s called the patent system.
Speaker 2 :In recent years, a broad fight over the enforcement of patents and what should qualify as a true invention has drawn players from every corner.
Kate: Figuring out how our patent system works, or maybe doesn’t work, is an important part of understanding capitalism today.
Speaker 4: At some point, patents are important for innovation, right?
Speaker 5: No.
Speaker 4: I mean, what, you’d blow up the whole patent system?
Speaker 5: Yeah.
Speaker 6: We open-sourced our patents, so anyone who wants to use our fans can use them for free.
Speaker 7: Your patents are open-sourced?
Speaker 6: Yeah.
Luigi: From the University of Chicago, I’m Luigi Zingales.
Kate: And from Georgetown University, I’m Kate Waldock. This is Capitalisn’t, a podcast about what’s working in capitalism today.
Luigi: And, most importantly, what isn’t.
Kate: On this episode, we’re going to talk about patents and our patent system. Is it a capital-is, or a capital-isn’t?
Do you know anybody with a patent, Luigi?
Luigi: Actually, I discovered that a colleague of mine has a patent on a trading system.
Kate: On a trading, like?
Luigi: A trading mechanism.
Kate: Wow, that’s pretty cool.
Luigi: Yeah, zero citations and zero use, but that’s—you never know.
Kate: Yeah, I mean, I remember when I was a kid, I used to think that getting a patent was like winning the Nobel prize or something. It was this honor that was reserved for a few people. And now they have college classes where the whole point is to try to get a patent at the end of the day, or at least submit a patent application.
Luigi: It’s a bit like having a published paper. It doesn’t really matter to have one. It depends on how influential it is.
Kate: Let’s back up a second and pretend that we’re designing a system from scratch. Let’s pretend that we’re like the founding fathers, and we want to encourage innovation, but we’re not really sure how to do it. Why the patent system? And what are the important things that you need to address, in order to create a functioning patent system?
Luigi: Let me start with a story. I don’t know if you’re familiar, Kate, with a type of violin called a Stradivarius.
Kate: Yes. Yeah.
Luigi: The Stradivarius was invented by an Italian guy called Stradivari, who apparently had a unique varnish that he was using on the violin that makes the violin sound so much better. In order to protect his innovation, he kept his secret. So secret that today, I think three centuries later, we cannot reproduce a Stradivarius violin.
Kate: That’s insane.
Luigi: Nobody has been able to match the quality of the Stradivarius violin. Now, compare that with the saxophone. The saxophone is an instrument created by Mr. Sax. He actually patented the saxophone. He used the patent for a while until the patent expired, and now everybody can produce a saxophone as good as the one that Mr. Sax invented. So, the patent is a way to ensure protection for a while, but it also makes sure that eventually this information is diffused, because the benefits of diffusion of information are enormous, as we know in economics. And we want to make sure that this takes place.
Kate: Yeah. I mean, also, not everyone can keep a secret as well as Mr. Stradivari, apparently. I think if everyone could just hide their secret sauce really well, that’d be one thing. But it’s not as practical these days.
Luigi: Yeah. But historically, the way they were maintaining secrets was actually much more violent. Very often, they were killing people so that the thing would not be reproduced.
Luigi: When rich kings or queens were building these complicated clocks that were unique in their features, the way they were ensuring that there was no other clock like that was to either blind or kill the person that designed it.
Kate: OK. Yeah. So that, that’s not ideal in modern-day societies.
Luigi: That’s not an ideal patent system, yes.
Kate: Right. And yet we still want to encourage innovation but also disseminate ideas at the end of the day. And so, one of the tenets of the US patent system, and actually our fee structure is somewhat designed around this, is that everyone needs to be able to participate, right? We want to encourage inventions from big companies and small companies. So, we want everyone to be able to apply.
We don’t want people who have made great inventions to suffer from not being able to get a patent. But on the other hand, if everyone can apply, then anyone can just submit anything, right? And so, you also want to discourage people from submitting bad patents or getting patents on things that are pretty obvious or have already been invented. So, you need some sort of review system. You need some experts who are looking over these patents and making a decision.
Luigi: And actually, do you know, Kate, that Einstein’s first job was to work at the Swiss patent office?
Luigi: Oh, yes. So, in the old days, they had pretty talented people in the patent office.
Kate: Part of the reason I thought this episode was cool was because my Uber driver the other day had been working in the patent office, which is not to say, by the way, that he’s not super smart or Einstein, but he was just going over how terrible it was to work there and how much pressure there was on him and how little time he had to review the patents and the little training that he got. So, it seems like if that was the standard that they used to set in Switzerland a long time ago, at least in the US, it’s changed.
Luigi: You bring up an excellent point that there is a tradeoff between doing a very careful job upfront on the quality of the patents, so having an Einstein review whether the patent is really an innovation and whether it’s worth patenting, or being very liberal in granting patents, but then letting litigation ex post fix the problem.
Kate: Luigi, let’s pretend that we’re submitting a patent.
Luigi: OK. Let’s say that I submit my silly patent, and you are the patent office asking me the questions that need to be answered in my patent. OK?
Kate: OK. First of all, what’s your patent?
Luigi: My patent is a way to underline audiobooks.
Kate: Huh. That’s actually not a bad idea. That’s a good one.
Luigi: Actually, to be completely honest, this is my son’s idea.
Kate: Oh, really?
Kate: That’s pretty cool. Yeah, I have felt that need before in the past. All right, so what’s your son’s name?
Kate: OK, so you and Giuseppe are submitting this patent. What are the first things you need to do?
Luigi: First of all, I need to get a provisional patent application, and this is a sort of bridge patent that lasts for a year. This application gives the patent office the time to have a glimpse at the patent and also helps reduce the chance that somebody steals this great idea.
Kate: So now, you’re working on your actual application, and when you file it, since you guys are just individuals, I as the patent office, I’m just going to charge you $785.
Luigi: Wow. That’s a lot.
Kate: I mean, that’s a decent amount of money, but it’s only roughly double for large corporations. And so, for a big corporation to be paying 1,700 bucks for a patent application is not that much.
Luigi: So, why is this fee not a function, number one, of how complex the patent is, and, two, also a function of the success of the patent?
Kate: Well, first, it’s because we want the system to be open to everybody. We want to encourage small innovators, and so that’s why the fee is relatively low, right? The actual fee of filing an application is less than the cost of reviewing the application for the patent office. But if your patent gets approved, then you have to pay maintenance fees and stuff like that that are higher, even though they’re not that high, they’re $1,000 a year or something.
Luigi: So, there is a bit of a payment for success, but very limited, because the patent office does not want to penalize success. In fact, if broadening access was not an issue, the optimal system would be a system in which you put in a big deposit, and if your patent is applied, you get most of your money back. And if it is not accepted, then you forego the deposit, because that is a way to discourage silly patents.
Kate: But at the same time, I think that would only hurt little guys who have liquidity constraints or borrowing constraints.
Luigi: No, no. That’s exactly the reason why the system is designed this way. But if the only concern was to minimize the silly patents and minimize the time wasted by the patent office in reviewing silly patents, that would be, from an economic point of view, the optimal system. We don’t want to use that system, because we want to grant easier access to the little guys, but the cost of that is that potentially we have a lot of silly applications, and we waste a lot of time in silly applications.
Kate: So, going back to your underlinable audio book, what do you need in your patent application?
Luigi: In order to file my application, I need, first of all, to have a picture. Second, I must prove that this stuff is new and useful. Useful in the sense that it provides utility in using it, and new in a sense that it should be nonobvious and not disclosed to the public before.
Kate: Right. So, if you would have written an article about your awesome new invention, then that counts as a disclosure, and that would hurt your chances of getting the patent.
Luigi: So, I should actually clear it with my son, because in this episode I’m disclosing his idea. And then it becomes unpatentable.
Kate: I didn’t even think about that. Oh, no. OK. Well, if Giuseppe is really angry, we might need to pay him. He might sue us, we might have to pay him damages for early disclosure.
All right, so now the patent’s in my hands. I’m the US PTO, Patent and Trademark Officer reviewer. I have to read this application, I have to search for prior patents, which for some reason are called prior art. I guess you can consider innovation art, but that’s probably one of the hardest parts, figuring out exactly what other patents in the past relate to Luigi and Giuseppe’s patent, because it’s not necessary for them to have done a thorough review of this, right? I mean, they have to prove that it’s new, but they don’t necessarily need to know every single patent that’s related. That’s my job.
Then, I have to compare all of this prior art with their patent. Possibly, if you’ve hired a patent attorney, which a lot of people do, I have to interview that person. Then, finally, I make a recommendation. So, in my mind, it seems like all of this should take at least a week. You know, maybe a couple of weeks if drawn out, but on average, each patent examiner only spends 18 hours on this entire process, including the final recommendation.
Luigi: So, does it mean that when I file the patent, within 18 hours, I have my patent back?
Kate: No, so that’s another issue. There’s a pretty huge backlog of patents. I think one of the most famous delayed examples is TiVo. For them to get a patent on their TV recording system, it took them 10 years to get word back from the US PTO.
Luigi: Ten years?
Luigi: By that time nobody was watching TV anymore.
Luigi: But help me out here. So, if I start producing the product while the patent is pending, does this make the product not novel and not patentable? Should I wait for the patent to produce?
Kate: No, you can still produce the product. But the question of whether other people have also produced a similar product while the patent is pending and whether you can recover money from sales from them, that I think is still pretty gray.
Luigi: So, if I am TiVo and I’m waiting for my patent, I can still produce TiVo, sell TiVo, and only what I’m not sure of is if somebody else can produce a similar thing at the same time. And, if they do, I don’t have grounds to sue them.
Kate: Yeah, it’s unclear how much you can get back from them. The backlog has gotten better. There were 750,000 patents in backlog, just in backlog, in 2007, and it was recognized that this was a huge issue. And so, the PTO tried to hire a bunch of people, tried to—
Luigi: Including your Uber driver.
Kate: Yeah, actually, that probably would have been around the time that he was hired. So, he might have been swept up in that. And so, they reduced the backlog to 570,000 by 2017, but that’s still a lot of patents in backlog. So, yeah, it usually takes a couple of years for it to get approved.
Luigi: But what are the requirements to work in the patent office? I understand you don’t need to be Einstein, but what do you need to . . . Do you have to have a degree? Do you have to have a . . . in what?
Kate: Yeah, so typically you’re assigned to a specific area, right? The people reviewing things like hammers are not the same people as the people reviewing AI. And, in fact, the amount of time that you get to review a hammer is a lot less than the amount of time you get to review AI. And so, you’re supposed to have a degree in that area. But an undergraduate degree counts.
Kate: I’m assuming that for people reviewing CS AI, they have more than an undergraduate degree, but still. So, let’s say we grant your patent. That means you get 20 years to be the sole producer or licenser of this new technology. And then, let’s say you find someone who is using your idea, right? Slate, who is definitely not one of our competitors, has implemented your, or a very similar, technology so that people can underline their podcasts. If you want to go after them, you can sue them in federal district court. That’s usually where a lot of patent litigation plays out. And, as of 2012, there is a new law that introduced other, more specific patent courts within the US PTO where you could also sue.
Luigi: Now, tell me a bit about this special court. Because you know, as an Italian, I don’t like special courts, because special courts were introduced during the fascist—
Luigi: —to do crazy stuff. So, in general, we don’t like special courts, because we want judges to be impartial. And one way to maintain impartiality is to not be too specific in one particular sector. Here we’re deviating from the general rule, because judges in all the other circumstances are general judges. They can decide from an FBI investigation of drug dealing to the patent. So why, for the patent, do we have this special court?
Kate: Well, I think it makes sense. I mean, you’re right in one sense that it’s weird if you’ve all of a sudden designed a special tribunal that can be subject to the political pressures of that particular institution. But at the same time, patents are so specific. If you’re challenging an AI patent, why not refer that to a judge who’s already familiar with AI patents rather than a federal judge who oversees all sorts of different cases, right? Are they really the right people to be reviewing such complicated matters?
Luigi: This is a very important question, because it is the tradeoff between expertise and capture. Certainly your expert in patents is much more knowledgeable than a general judge. But he also probably has a different agenda, because what kind of job did he do before being a judge, and especially what kind of job is he going to do after being a judge? And who appoints him or her?
Kate: Yeah, so a lot of this concern is legitimate, and some of the proof that it’s legitimate is that these special patent offices, or these special patent judges, do seem to be highly influenced by political pressures, or at least the current political mood. Part of the reason that they were strengthened was because of patent trolls, which we can get into in a second.
But around the ‘90s and the early 2000s, people hated this idea that there were too many patents and there were all these patent trolls that were trying to extract fees from people just using simple technology. And so, even though these patent-specific tribunals or these courts already existed, historically, they always upheld a patent. But then, after the passage of this law, which created more of these types of courts, the pendulum swung in the other direction, and it made it much easier to strike down these patents. So, it’s not really the existence of the court that matters. It’s just the general political mood that matters. And I think that that’s borne out in the evidence of the rulings of these courts.
Kate: Let’s take a step back and talk a little bit about the history of patents. There’s a ton of changes in law, but they’ve been around for a long time, right? The first patent was granted in 1790, so to give you a sense of how the system has changed, in 1980, the US issued about 66,000 patents, and by 2017, they were issuing 347,000 patents. And so, not only the number of patents granted, but also the number of applications, has just exploded in the past 30, 40 years.
A big explosion took place around the IT revolution, right? When we were learning about computers, learning about the internet. And a bunch of people started filing patents then. Then, this concept was born of the patent troll. It’s usually a company, even though it could be an individual, who specializes in getting a bunch of patents and not necessarily using them for any practical purpose, but using them to sue companies that do similar things. And so, these companies, their purpose is either to buy up a bunch of patents or to file patent applications themselves. Not necessarily make the good, but just try to sue whoever has any sort of product that’s similar to something that’s in one of their patents.
Luigi: And actually, Google bought Motorola, not for the quality of the phones they were producing, but for this stock of patents they had, as a way to protect themselves against patent trolls. And also as a way to play some games with other, older patents in a kind of mutual exchange.
Kate: Yeah. The people who hated these patent trolls are understandably large tech companies who are getting attacked constantly by these patent trolls. And to give you a sense of how powerful these patent trolls were, between 2011 and 2012, almost $30 billion of litigation was involved in this patent trolling business. Understandably, high-tech companies hated all of this. And so, they started lobbying pretty hard in the late ‘90s, early 2000s, to try to get Congress to pass legislation that would attack or limit the ability of these patent trolls to sue whatever company had similar patents.
Luigi: Yeah, but to some extent, what is a patent troll? It is a bit in the eye of the beholder. Some patents that, for example, the pharmaceutical companies receive are completely and exclusively designed not to reward innovation, but to keep people out and to force Medicare, Medicaid, to pay higher prices for the drugs, or to force the insurance company to pay a high price for the drugs. So, we can claim that many of those patents are basically patent trolls. But you don’t say the large pharmaceutical companies are patent trolls. You do say that Intellectual Ventures is a patent troll.
Kate: Yeah. You raise a good distinction. Basically, you’re describing two different types of patent manipulation, right? The pharmaceutical company that’s manipulating their drug a little bit and trying to get a new patent on it, that’s not what we’d consider a typical patent troll. But the reason that they do that is because the patent lifetime is limited to 20 years.
Luigi: But it is a manipulation of the patent system.
Kate: Yeah, it’s absolutely manipulation. But it’s specifically because they want to extend the life of their patent, not because they don’t have any legitimate use for the patent within their own company, right? They’re basically just trying to get monopoly rents for longer. And you raise a good point. I mean, we have this 20-year limit on patents. It’s really hard to get an extension. So, that’s one feature of the system that’s difficult to manipulate, which is that patent extensions are almost never granted.
In some cases, if it took a really long time for you to get your patent granted, maybe you’ll get a one- or two-year extension. But they’re pretty rare. In all other circumstances, they require an act of Congress, and Congress is hard to corral.
Luigi: Good luck.
Kate: Yeah. So, it’s pretty tough to get a patent extension, which is why these companies try to manipulate the system by changing their drugs slightly and getting a new patent.
Luigi: But then, actually, of all people, a hedge fund manager, Kyle Bass, saw an opportunity to challenge some of the pharmaceutical patents that are pretty similar to patent trolls. So, you know that when the patent for Prilosec, which is an antireflux drug, expired, the producer, Pfizer, introduced Nexium, which is basically the same chemical formula with a slight variation, but they gave it a patent.
And so, they pushed this Nexium over the Prilosec. The idea of Kyle Bass was to use this special patent system to challenge this kind of pretend patent or useless patent. And you’d say, how would it make money? He was making money by shorting the stock of the company that he was challenging.
Kate: Yeah, I mean, I think that is the perfect example. That’s the case in point, which is that there’s all these issues of the patent system. It’s so hard to get the rules right, and the rules swung in a massive direction from one side to another around 2010, 2011. Prior to that, it was really easy to get patents, and it was really easy to enforce them. And so, we saw all the manipulation by these patent trolls, but then, after these laws were passed that made it really easy to sue and invalidate a patent, then there started being these specialty occupations, or they’re called prior-art searchers, who would just search through these archives of patents to sue specific types of patents for the exact reason that you just described. Either you could get hired by a competitor and undo a competitor’s patent, or if it’s a public company, you could short their stock. Either way, it’s being manipulated.
Luigi: But actually, the interesting story is, I thought he was providing a public service, because he was destroying these useless monopolies of drugs that have no innovation in them. And it’s done to exploit the patent system. And he was trying to get some reward for it, but he was doing something that was socially good. Unfortunately, his enterprise in this direction did not succeed, because the pharmaceutical industry is too powerful, even in the special courts, or I will say, especially in the special courts.
Kate: Good one, Luigi.
Luigi: Who decides? The US patent office decides by itself what are the levels of the fees? And who appoints the patent office?
Kate: It’s actually the director of the US PTO, which doesn’t look great. But there’s actually an interesting study that was done around this. It was conducted by two law professors, Michael Frakes from Duke Law and Melissa Wasserman from UT Law. And they were trying to figure out whether or not the financial constraints of the US PTO actually affected what types of patents were being granted. So, they looked at periods in which the US PTO had some resource constraints, maybe they were a little underfunded. And they compared patent grant rates between the types of applicants that were likely to be repeat patent filers, so good sources of revenue, versus patent filers that were more likely to be one-time patent filers. And they found that when the US PTO was resource-constrained, they were more likely to grant patent applications to these repeat filers, potentially because they were a good source of business.
Luigi: Incentives do work, and even perverse incentives do work perversely.
Think about this: I have a patent, so I am suing somebody else. Imagine that after our episode, Amazon is going to develop for Kindle a product like the one my son Giuseppe wanted to invent. OK?
Kate: Oh, no.
Luigi: And of course, Amazon being Amazon, they’re going to put it in every Kindle, and it’s going to be a big success. And my son decides to sue. Is he a patent troll, or is he a legitimate protector of his right?
Kate: Well, obviously, in the way that you describe what just happened, he’s not a patent troll, right? A patent troll, their singular purpose is to aggregate a bunch of patents and to go after claims. They didn’t necessarily invent anything. They’re not necessarily trying to improve the system at all. They’re just taking advantage of the legal system. And so, if your son actually had this idea and wanted to produce the idea, then he’s, by definition, not a patent troll, especially if he’s not making a living doing this. But the legal system now has become so antagonistic towards people suing for these sorts of patent claims, or to assert their patents, that it has become much, much harder for someone like your son to win a ruling against a company, like Amazon in this case, because of the lobbying that these high-tech companies did in the early 2000s.
Luigi: Yes, so I think there’s a serious distortion, because I imagine my son has better things to do than to sue for this. So, he sells the patent to Intellectual Ventures. Intellectual Ventures sues for him. They did not invent anything, but in a sense, they pay my son some money under the expectation they can make that patent hold. The existence of these brokers is useful to reward innovation and make sure that the little guys get protected. So, I feel that all this rhetoric about patent trolls is very much pushed by big tech that doesn’t want to have any little guy in the middle.
Kate: I think that that’s totally fair, right? I mean, some patents should be litigated, but there’s also obviously room for abuse, right? If you patent a stick and then sue anyone who’s ever made anything involving a stick, then that’s probably you trolling and not looking out for the little guy. But that’s exactly what they claim. They claim to be protecting the little guy. But obviously, in some cases, they can take it too far.
Luigi: But isn’t this saying that our judicial system doesn’t work? Because—maybe the case is too silly, but if I sue for a stick, how long will it take for a judge to dismiss my case? And probably dismiss my case with some notion that I have to pay some costs, some legal costs, because it was preposterous that I was suing about that. I should be punished by the court system. And that would be the best deterrence against patent trolls. Not to add a special court system in which, basically, the big companies get huge favors, because many of the lawyers involved work constantly for them and then become sort of an arbitrator in this judicial system. Thus, I think that that is to me a very serious distortion.
Kate: Great. So, I think that’s an interesting segue into an issue that often comes up with regards to patents, which is that the little guy is becoming less and less a part of this whole system. To put some numbers on that, the share of patents granted to small entities, either individuals or small businesses, those used to be greater than 30 percent of all patent grants in 1995. And, as of 2014, it was less than 20 percent, and I think that number is continuing to drop. Is this something that we should worry about?
Luigi: It depends on what caused this drop, because I can see an argument saying, look, innovating is more and more expensive, so it is hard for the guy in his basement to create a really innovative product. You need to be part of an organization that invests a lot of money to create those products. And so, you see fewer patents produced by individuals. That would be a natural technological explanation, and I would not be worried about that. The other explanation is that the system is broken, I don’t feel that I am protected anyway as a small guy, so I don’t even bother to file a patent, because the patent for me is useless.
Kate: Yeah. I think that one of the reasons that this is a difficult issue to address is precisely what you said. It’s hard to figure out what’s causing this, and obviously, there is a trend towards consolidation of businesses, big businesses. It’s much, much easier for them to innovate. And so, can we really tell whether this is the fault of these strengthened patent courts? Which, by the way, since they make it easier for companies to undo patents, this has in a sense increased the bargaining power of big companies vis-à-vis little firms if the big companies want to buy the patents. And so, that’s one argument, is that it’s the fault of these strengthened patent courts, but it’s hard to tease that apart from the general trend. And so, I don’t know.
Luigi: Hopefully, our conversation has highlighted some of the technical aspects of patent law. But we’ve not touched yet on the fundamental question, which is, does our patent system spur innovation? Because the only reason we want to have a patent system is because we think that we have more innovation with it rather than without it. Is that the case?
Kate: In our next episode, we’re going to be broadening the question a bit to not just patents but all sorts of intellectual property. That also includes trademarks as well as copyrights. And we’ll ask ourselves, at the end of the day, is this whole system worth it?