Reader X. Trapnel appears to have a newly created blog, and he has a provocative and well-argued post arguing that copyright and patent are examples of Bastiat’s broken window fallacy. I made a much more limited version of this argument back in May, where I pointed out that it was a mistake to measure the worth of peer-produced projects like Wikipedia by the revenues they generate. But X. Trapnel goes much further and argues that the entire argument for patent and copyright law are examples of the fallacy:
Just as in the Bastiat story, you have the helpful onlooker who says “But everyone must live, and what would become of innovation if every innovator could have his insight copied by the first free-rider who came along?” Just as in the Bastiat story, this is wrong. What is seen is the way in which the protected firm uses his IPR to generate monopoly profits, some of which are then plowed back into R&D, generating a pleasant stream of innovation. What is not seen is what would happen in the absence of this protection: the innovator would have to keep innovating in order to maintain his market, leveraging his expertise into further productive developments, while newcomers would be able to experiment on their own with the knowledge produced by the first. Money that once went to monopoly rents would go instead to other, more productive things–including further innovation. The neo-Schumpeterian retort is that this is hopelessly naive: innovation requires large capital investment and the reasonable hope of monopoly rents to recoup it. But this is mere question-begging, and its plausibility lies, again, with the distinction between What Is Seen and What Is Not Seen: when we give innovators monopoly privileges of this sort, we thereby tilt the playing field dramatically towards heavily capitalized firms by jacking up the costs of the inputs (eg., prior innovations, a skilled legal team, insurance against lawsuits) to production. As a result, What Is Seen is capital-intensive innovation; What Is Not Seen is the less capital-intensive innovation that the legal regime has stamped out.
As I’ll explain below the fold, I think this argument has a certain plausibility (especially for patent law), but ultimately I don’t find it persuasive.
To evaluate the argument, it’s important to identify the “what is not seen.” With the broadest and most intrusive parts of the patent system, this is easy. With software patents, for example, what is seen is the billions of dollars that companies holding such patents earn. What is not seen is what the companies and consumers who had to pay those royalties would have done with the money had it been left in their pockets. Also not seen are the new innovations that were stifled because companies couldn’t navigate the relevant patent thicket, and the millions of dollars that went to keep the patent bar in Mercedes. The NTP-RIM case was one hell of a broken window.
It’s not as obvious, however, where the “what is not seen” is to be found with copyright law. Trapnel mentions the derivative works right in copyright law, but it doesn’t seem to me that being unable to use another author’s characters is a particularly serious burden. (although I’m sympathetic to the argument that the derivative works right is currently too sweeping)
However, it’s not obvious to me that the core of copyright–the exclusive right to distribute copies of your creative work–is an example of the broken window. What makes software patents a broken window is that it imposes costs on unwitting third parties, in this case other programmers who happen to stumble upon the same “invention.” But the proverbial thousand monkeys aside, no one accidentally stumbles across another person’s novel. The only actions that are restricted by the core of copyright are actions that would not have been possible at all had the author not first produced the work being copied.
Trapnel’s response seems to be that because a significant number of authors would continue producing works even without the monopoly of copyright, the need to pay for works that they would otherwise have obtained for free is a broken window. I have two problems with this response. First, it hinges on an empirical claim that I find dubious: that approximately the same quantity and quality of copyrighted works would be produced in the absence of copyright. I find this claim plausible for a few categories of content, notably popular music. But I find it implausible for a number of other categories, such as accounting textbooks, movies, and video games.
But that’s an empirical question which I think we’re unlikely to ever be able to answer conclusively. The more fundamental objection is that it’s not obvious to me that there’s a broken window at all here. Keep in mind that the harm of the broken window lies in the fact that a window was broken, not merely that the shopkeeper had to give money to the glazier. If we could magically transfer six francs from the shopkeeper’s pocket to the glazier’s pocket without breaking anything in the process, that would not be an example of the broken window fallacy. Likewise, to the extent that copyright simply transfers money from the pockets of consumers to the pockets of authors (or their publishers), this isn’t a broken window at all; it’s just a wealth transfer. And one that, frankly, most people including myself feel is morally justified.
Now of course, the copyright monopoly isn’t simply a wealth transfer. It imposes deadweight losses along several dimensions. First, there’s the fact that in setting a nonzero price for copyrighted works, some consumers are inefficiently excluded from having access to the copyrighted work. Secondly, there are the transaction costs involved in paying for access to content. Perhaps most importantly, there are the costs associated with rent-seeking on the part of the publishing industry. We have life-plus-seventy copyright terms and excessive laws like the DMCA because the entertainment industry spends millions of dollars on lobbyists every year.
But it’s not obvious that these “broken window” costs are particularly large compared to the size of the industries we’re considering here. The movie, music, and book industries collectively generate tens of billions of dollars in revenue every year. If even a small part of that output is content that would otherwise not have been produced, it could easily outweigh the deadweight costs I outlined above.
Finally, it’s worth mentioning that this is a highly contextual question. The extent to which copyright law is like broken windows is highly dependent on the details of technology and the law. For example, fair use is designed to minimize deadweight costs by eliminating both transaction costs (excerpts and parodies) and to allow relaxed access to those who likely wouldn’t be able to pay full price (copying for academic purposes).
It seems likely that the deadweight costs have increased as the distribution of content has become more efficient and decentralized. Copyright law worked “with the grain” of 20th century publication technologies that involved shipping books, records, VHS tapes, etc around the country. People were opening their wallets for those products anyway, so copyright didn’t really add much in the way of added costs. But as content distribution technologies have matured, the costs of the copyright system have clearly started to rise.
Finally, unlike the patent system, people can opt out of the copyright system. The free software movement is a community of programmers who have decided that they’re better served by mutually pledging not to charge one another for content. Increasingly, we’re seeing musicians doing the same thing: they make their music available on MySpace or MP3 blogs in the hopes that they will become famous, even if they can’t become rich.
So although there clearly are some deadweight costs to the copyright system, it’s not at all obvious to me that they’re significant. Markets have developed a number of ways to minimize these deadweight costs in cases where the copyright system is not helpful. And because it seems to me that there are still important industries in which copyright is helpful, I doubt I’ll be in favor of abandoning the system any time soon.