Andrew Torrance: Patents and the Regress of Useful Arts

As noted previously, I was on the IP panel of the Open Science Summit last weekend. I was please to meet co-panelist Andrew Torrance, a law professor at the University of Kansas School of Law. My presentation was libertarian and theoretical, while his was empirical. He presented the results of a complicated and detailed simulation he did to test the effect of patent law on innovation.

The results of his study are fascinating. They are presented in his 2009 paper Patents and the Regress of Useful Arts. Abstract:

Patent systems are often justified by an assumption that innovation will be spurred by the prospect of patent protection, leading to the accrual of greater societal benefits than would be possible under non-patent systems. However, little empirical evidence exists to support this assumption. One way to test the hypothesis that a patent system promotes innovation is experimentally to simulate the behavior of inventors and competitors under conditions approximating patent and non-patent systems. Employing a multi-user interactive simulation of patent and non-patent (commons and open source) systems (“The Patent Game”), this study compares rates of innovation, productivity, and societal utility. The Patent Game uses an abstracted and cumulative model of potential innovations, a database of potential innovations, an interactive interface that allows users to invent, make, and sell these innovations, and a network over which users may interact with one another to license, assign, infringe, and enforce patents. Initial data generated using The Patent Game suggest that a system combining patent and open source protection for inventions (that is, similar to modern patent systems) generates significantly lower rates of innovation (p<0.05), productivity (p<0.001), and societal utility (p<0.002) than does a commons system. These data also indicate that there is no statistical difference in innovation, productivity, or societal utility between a pure patent system and a system combining patent and open source protection.

He describe this also in his Google TechTalk, “The Patent Game: Experiments in the Cathedral of Law”:

Torrance’s simulation indicates that a society with no patent law would have more innovation than one with patent law. What is interesting is Torrance said that when he began this project he fully expected the simulation to show the opposite: to vindicate the legitimacy of patent law. In this, he is similar to Boldrin and Levine, authors of Against Intellectual Monopoly, who began their empirical study of patent law expecting it justify it, but ended up completely opposing patent and copyright law (Boldrin discusses their change of heart in the podcast mentioned in EconTalk Podcast on IP with Michele Boldrin).

Yet another nail in the IP coffin. (See also my post Yet Another Study Finds Patents Do Not Encourage Innovation.)

Addendum: I explain in There’s No Such Thing as a Free Patent, to justify the patent system on such wealth maximization grounds, first you must show that it (a) stimulates an overall increase in innovation: that is, that the value of the extra innovation that it gives rise to (innovation we would not otherwise have, absent the patent system, or innovation that comes earlier than it otherwise would have) exceeds the value of innovation that is lost because of the patent system. It seems to me that it is impossible to know any of these things, much less value them. How can you know when at a given innovation that you posit came earlier than otherwise, “would have come” absent the patent system? How can you know what innovations were lost due to the patent system?

In any event, even if you somehow posit or establish that the patent system does generate net innovation, you still need to show (b) that the value of this net innovation is greater than the cost of the patent system to the economy or to society.

For example, let us assume that the patent system costs the economy $500B a year. And let us assume that once in place, $200B of extra innovation exists or comes early, and $50B of innovation is lost. This means that a $500B cost is imposed on the economy, to yield $150B of net innovation. This is an overall loss of $350B.

From studies I am aware of, no one has succeeded in showing even (a) yet, much less (b). That is, it appears that overall innovation is diminished due to the patent system. So taking the example above, that would mean something like this: A $500B cost is imposed on the economy by the patent system, and the result of this is that $50B of extra innovation is stimulated, and $200B innovation is lost, meaning a net innovation loss of $150B, for a cumulative loss of $650B. I.e. we impose costs on the economy only to decrease overall innovation. In Costs of the Patent System Revisited, use a similar structural calculation but with more realistic assumptions to estimate that the patent system imposes an overall cost of at least $42B a year (on the US economy alone). And as noted in Patent Trolls Cost The Economy Half A Trillion Dollars since 1990, even more recent patent troll cost studies have led me to revise my estimate of annual patent losses from $42B to $100B a year–and that’s still conservative.

It appears to me that Torrance’s conclusion about a patent system stimulating less innovation than a “commons” system would indicate that in point (a) above, there is not overall net innovation in a patent system. He also concludes there is less “societal utility” in a patent system, compared to a non-patent system, which, if I understand his methodology, more or less corresponds with point (b) above, i.e. it indicates that the social cost of patents, minus the value of the net innovation it spurs, is overall negative–which is what one would expect if the patent system actually decreases overall innovation.

Further, as I argue in Against Intellectual Property (pp. 19-23), there are inescapable methodological and ethical problems with any empirical defense of a patent system, and with empirical studies such as Torrance’s.

Ethical Problems with Utilitarianism

First, the ethical problem:

First, let us suppose that wealth or utility could be maximized by adopting certain legal rules; the “size of the pie” is increased. Even then, this does not show that these rules are justified. For example, one could argue that net utility is enhanced by redistributing half of the wealth of society’s richest one percent to its poorest ten percent. But even if stealing some of A’s property and giving it to B increases B’s welfare “more” than it diminishes A’s (if such a comparison could, somehow, be made), this does not establish that the theft of A’s property is justified. Wealth maximization is not the goal of law; rather, the goal is justice—giving each man his due. Even if overall wealth is increased due to IP laws, it does not follow that this allegedly desirable result justifies the unethical violation of some individuals’ rights to use their own property as they see fit.

For example, imagine A has a $10M estate that he earned from a business he created on the free market. The utilitarian could argue that taking $1M from him and distributing it in $10k chunks to 100 needy people increases social utility, since those 100 recipients each “really value” the $10k, and the sum of their utility increases is far greater than the loss of $1M to the millionaire, since he still has $9M left. Yet the taking of $1M is still theft. The fact that the recipients value it “more” than he does, does not change this. Or, as I pointed out in The Trouble with Libertarian Activism, “the utilitarian standard would permit, for example, a very desperate rapist to rape a woman of loose morals, since the damage to her is arguably small (by utilitarian standards) and the benefit to him great, providing a net benefit to society.”

Of course this is the problem with the entire utilitarian, law and economics, wealth-maximization mentality. It just assumes that the purpose of law is to be tweaked to maximize certain values, or to “incentivize” behavior. It is really a statist, central planning type mentality. It even trips up the analysis of anarchist libertarian David Friedman. For example, in his Law’s Order, ch. 11, he writes:

Property rights serve two related functions: They provide both a way of deciding who gets to use what when and an incentive for creating things.

In the case of intellectual property, the first function is not merely unnecessary but perverse. We cannot both drive the same car to different places, which is an argument for property rights in cars. But we can both use the same idea to build different machines, or simultaneously read different copies of the same book, which is an argument against property rights in ideas or writings.

Having one more person choose to read a book does not increase the cost of writing it; that must be paid in full in order for there to be any readers at all. Hence when we include in the price of a book a royalty payment to the author, we are, from the standpoint of efficiency, overcharging. If the price of an eleven dollar book represents a ten dollar production cost plus a dollar in royalties to the author, an additional copy produces a net benefit as long as the purchaser values it at more than ten dollars. But if he values it at more than ten and less than eleven, he will not buy it, which is an inefficient outcome. The same is true for the potential user of an idea who values it at more than zero but less than the license fee set by the patentholder. On the dimension of how many people use an idea or a writing private property gives an inefficiently low result, commons an efficiently high one.

There remains the second function of property. It is hard to read a book if nobody has written it, and authors may choose not to write books if they cannot collect royalties on them. Similarly for inventions. So the protection of intellectual property does provide some benefit. But the case for treating ideas as property is, economically speaking, weaker than the case for propertizing many other things, which may help explain why intellectual property is a relatively recent institution.

Friedman recognizes the need for property rights as a response to scarcity in his first “function,” but assumes that the “incentive” effect of property rights is an independent justification for property rights. For normal property, like scarce resources, assigning property rights in the scarce resources, so that peaceful, productive, conflict-free use may be made of these resources, also happens to give owners “incentives” to use the property efficiently, and to transform the property into more valuable arrangements, i.e. to “create things”–actually, to create wealth, not things (see my post Locke on IP; Mises, Rothbard, and Rand on Creation, Production, and “Rearranging”). But this does not mean that property rights in non-scarce things can be justified simply to try to incentivize “creating things.” But thinking of property as having “two functions” leads to the bizarre conclusion by an anarchist that legal protection in ideas might be justified–that is, the right of one person who did not do anything to produce or homestead someone else’s property, to veto that other person’s uses of his property. As I explain elsewhere, such IP rights may be viewed as negative servitudes, a type of property right in the IP holder, in someone else’s property. But negative servitudes are justified only if granted contractually by the owner of the servient (burdened) estate. Legally giving such a right to the IP holder without the owner’s consent is a form of theft, or redistribution of property. (See also my post David Friedman on Copyright, and Hoppe’s criticism of the similar “Coasean” type reasoning in the Chicago Diversions section of this article.)

Methodological Problems with Utilitarianism

Second,

Advocates of IP often justify it on utilitarian grounds. Utilitarians hold that the “end” of encouraging more innovation and creativity justifies the seemingly immoral “means” of restricting the freedom of individuals to use their physical property as they see fit. But there are three fundamental problems with justifying any right or law on strictly utilitarian grounds.

First, let us suppose that wealth or utility could be maximized by adopting certain legal rules; the “size of the pie” is increased. Even then, this does not show that these rules are justified. For example, one could argue that net utility is enhanced by redistributing half of the wealth of society’s richest one percent to its poorest ten percent. But even if stealing some of A’s property and giving it to B increases B’s welfare “more” than it diminishes A’s (if such a comparison could, somehow, be made), this does not establish that the theft of A’s property is justified. Wealth maximization is not the goal of law; rather, the goal is justice—giving each man his due.[39] Even if overall wealth is increased due to IP laws, it does not follow that this allegedly desirable result justifies the unethical violation of some individuals’ rights to use their own property as they see fit.

In addition to ethical problems, utilitarianism is not coherent. It necessarily involves making illegitimate interpersonal utility comparisons, as when the “costs” of IP laws are subtracted from the “benefits” to determine whether such laws are a net benefit.[40] But not all values have a market price; in fact, none of them do. Mises showed that even for goods that have a market price, the price does not serve as a measure of the good’s value.

[Mises]