I mentioned Boldrin and Levine’s argument against patents. Their paper (it is posted but labeled a draft) is very uneven, moving between dubious assertions and insightful analysis. Lurking over their discussion, though they do not acknowledge it, is Teece’s paper on profiting from technological innovation, which identifies innovators, imitators, and infrastructure providers as being in a competitive state amongst themselves for a share of the value created by new technology becoming used. Intellectual property is a social innovation of sorts to address the relative strengths in this competition, providing at least initially some advantage to inventors and authors with regard to the subsequent exploitation of their work. The odd thing in Boldrin And Levine’s analysis is that they appear to assume that IP is entirely the domain of companies battling other companies, and that the only purpose of a patent is to protect the failed competitive position of big companies. That comes off as a cynical view couched as academic certainty:
The dead hand of dying institutions . . . gets hold of the industry as they attempt to tax consumers, new entrants and any potential competitor. Far from being encouraged, when an industry matures innovation is blocked by the ever increasing appeal to intellectual property protection on part of the insiders. (p. 5)
The example Boldrin and Levine use is that of the Android operating system. What they don’t consider is why Apple and Microsoft, coming at things from different directions, both apparently despise Android and yet have tolerated other operating systems in mobile phones created by competitors. Surely the smartphone industry is not “dying”–it is in the throes of growth and innovation. It is just the opposite of dying. For dying stuff, maybe we might look at typewriters or film-based photography. I’m not seeing big patent litigation there. Is it possible that far from being “innovative” Android is rather “imitative” without adding much at all to the overall mix? Unlike the other phone operating systems, Android grabs what it likes and recombines these into an “open” (and now fragmented) platform, competing for market share but operating with a different model–one might call it “dumping” or “loss leader” depending on one’s point of view.
Android, like Wikipedia and Linux and Napster/BitTorrent/Pirate Bay, appears to be a massive endeavor to imitate, copy, infringe, and loot the work of others. One could argue Android is a huge rip-off of existing operating systems engineered by Google in its own profit-seeking. Or one could argue that Google aimed to break up the established technology order and create more competition among hardware manufacturers while at the same time opening up operating systems to Google applications and search services. Depending on your point of view, patents are a defense to such rip-offs, or are part of the understructure that prevents progress in innovation once a platform has been established and dominated by a few corporate players or standards. In any case, I don’t see how the Android situation supports Boldrin and Levine’s argument.
It might be here, however, that we start to look at the effect of patents, not so much for the local role in “protecting” inventions, but for the persistence of industry positions long after the original patents have expired and have been replaced by a tangle of others. At some point, patents serve as a barrier to entry to the platform–either one has patents to cross license, or one has to pay to gain access, or one has to go rogue. The public policy question, then might be, what are the conditions under which a persistent technology infrastructure should become public domain? For that, we might consider not only the IP regimes, but also the governmental regulatory role, and the private development of standards. Each of these administrative tool sets may be used to preserve a closed infrastructure or platform, whether it is a phone operating system or an electricity distribution grid.
One might argue that new platforms are reefs, full of potential innovation, but that old platforms are middens, controlled and malodorous. It is in the life of platforms that much of the social impact of technology plays out. Where a patented invention is equivalent to a platform, then the discussion reduces to the role of the patent. But the general case is the platform–whether it is a product made of multiple inventions, or a standard, or a combination of technologies that work together to provide a function (as a television, a cable system, and a remote).
Eliminating patents does not appear to be a robust means to break up old platforms. Old platforms are notorious at blocking threats (through regulation, through purchase, through competitive practices) to preserve their franchises. If patents are not available for the purpose, then other strategies will be employed. The real problem for public discussion is how to deal with an innovative infrastructure that has become a utility. On the one hand, such a utility can provide the foundation for new enterprise. On the other hand, it can limit the directions any new enterprise might take and extract such value from new enterprises that they wither and fail.
For all that, perhaps old platforms are not entirely bad. They become something of the conventions of society, like the horse for which the auto garage now serves as a proxy for the old stable. The point is not to break up platforms for the sake of doing so, but to ask whether there are reasons that at a certain point old platforms, like IP rights, ought to be less well defended, whether by IP, regulation, or cross-contracting. One might throw into the mix building codes, radio spectrum regulations, and company incorporation law along with copyrights in motion pictures and patents on electrical circuits.
If, as Boldrin and Levine argue, over time parties that benefit from governmental regulation lobby for an expansion of the rights that support their interests, then one needs something that offers a reset, or a way of re-weakening these rights to allow for other directions and leaders to emerge. One is reminded of the “Year of Jubilee” in ancient Jewish culture, in which every fifty years there is a reset of property and relationships. While that’s not generally practiced for personal possessions, perhaps such a thing would make sense for technology platforms. At some point a government could declare a platform stable and therefore a public utility, and at a point twenty years thereafter, various government-enabled proprietary positions (IP positions, contract positions, and government regulations) are weakened or eliminated. Copyrights and patents revert to their inventors, may not be enforced against the public utility, but authors and inventors may still have a right of attribution and there could even be some form of mechanical license (as there is, for instance, in sound recordings). Copyright law even has a “termination right” for assignments and exclusive licenses. Perhaps other forms of government regulation would benefit from a similar structure.
Boldrin and Levine argue that since political reform of the patent system is impossible, it should be “progressively dismantled”. One might observe in this a certain cynicism–that because so many players are interested in a strong patent system, governments should dismantle the system to serve “consumers” who will somehow benefit better from no system than from one that cannot be reformed. I don’t see how that follows. However, there is a difference between “reform” and perhaps adjustments over a period of time. Boldrin and Levine argue that for countries with weak IP systems, there is some gain in attracting R&D funds (at least in the documented economy) by strengthening the IPR regime, but that this effect is lost in further strengthenings of IPR in more mature IPR economies. One would think, then, that there might be advocates for a progressive kind of freedom–that we do not abandon the grammer of IP entirely, but that we cycle it back to more primitive forms in particular areas of activity. Such was the case, in essence, in the early aerospace industry, as Sylvia Kraemer has documented.
Rather than go along with Boldrin and Levine that things need to be dismantled altogether, I suggest that what we have is a workable system overrun by a much greater volume of dubious exploitation. The greater volume of exploitation washes out any understanding available by means of coarse economic metrics of the entire activity set. Imagine a concert at which there are a few hooligans. Something for security to deal with. But if the concert promoters find that hooligans pay more and produce greater profits, then the concert may become largely a matter of hooligans, and one response is to ban all concerts. Another is to roll back the incentives for profiting from hooligans.
My sense is that IP regimes work best when there is a robust public domain, both operating in parallel with IP and anticipated at the expiration of the term of IP rights. Jefferson made such an argument for the term of copyright–that it should be no longer than half an average adult lifespan (i.e., the time being an adult), or 19 years (at the time). As Jefferson argued, “the earth belongs to the living”. While a patent term of 20 years from date of filing may yet be within the adult lifetime, it is a long time to wait to gain access to live-saving drugs or to be able to build one’s own 3d powderbed printer.
There are a number of ways to progressively weaken IP laws. One is to shorter their term. Another is to limit their scope. Copyright has dramatically expanded scope, leading to such things as “thin” copyrights in such things as treating lamination of a workbook page as a “derivative work” (preventing reuse of the page by students). A third is to expand mechanical or compulsory licensing. In sound recordings, if one is published, then others may “cover” that same music with their own recorded performance, subject to paying a royalty per copy as set by a “royalty tribunal”. Perhaps something like that for software would work well, as it is, in its way, mostly expressive “algorithm recordings” with virtually nothing that is non-obvious from the perspective of utility beyond expression. Allowing re-use with a mechanical royalty would end most of the fuss in the mobile communications market, or at least limit that fuss to royalty tribunals.
As for universities, there is a similar problem, where the role of IP being proposed by Bayh-Dole is substantially different from that of the typical patent owner, yet that distinctive role is generally ignored by university licensing offices. It doesn’t mean that university licensing should cease, but it does suggest that some folks, hooligans if you will, have made a better case for profiteering, and in doing so have swamped out the underlying activity that does appear to be productive with a substitute that shows little sign of that productivity while offering proxy metrics in the form of increasing patenting (enclosing the commons, denying the public domain, adding noise to the system) and income (extracting rents from industry).
Universities have the opportunity to implement weaker IP systems–which turn out to be stronger approaches to advance university missions of research, instruction, and public benefit. Weaker IP systems can implement privately suggestions along the lines of those above, with the added advantage that a variety of approaches can be explored, and the more effective of these adopted broadly. Consider, then, if universities went back to a voluntary assignment approach. Faculty inventors could choose between the university, another agent, the public domain, or (in the case of federally funded inventions) the government. The university’s pitch to manage would not be “our licensing professionals are less inept than you are” but rather “we offer a clear public approach to making your work broadly available on terms that are deliberately weaker than those that a primarily rent-seeking agent will use.”
The weakness would be the pitch. Short term. Default non-exclusive access. Nominal mechanical royalties. Faculty inventors choosing this approach would be choosing to support the university’s (and their own) commitment to public scholarship. Others could make other choices, to pursue rent-seeking monopoly deals. There are reasons to do that, just as there are reasons to still have KISS concerts. It’s okay. Everyone has to make a living somehow. Royalties could be set by a public-industry-university tribunal that didn’t include the university managing the rights, with the condition that the costs of patenting would be recovered, and the inventors recognized with a share (in part in recognition for choosing this approach), and anything else would be to support research and education, as per Bayh-Dole’s great hope. Except here, the tribunal might also oversee how that research and education funding was spent, so it doesn’t become administrative slop funds.
Such weaker forms of IP management open up significant opportunities for university work–instruction, since more people will want to learn how to use the new stuff easy to obtain; research, to follow on with expanding the understanding of how the new stuff works and how it may be applied; consulting, to provide custom, personal assistance in getting up to speed with the new stuff; and startups, which can launch without a bunch of painful, slow negotiation over a license. Oh, to be the happy dog again! If a university wants to make a lot of money, build goodwill, and show itself to be a key player in an innovation ecosystem, then it will drop compulsory IP claims, weaken its own management systems, and align those systems with its public mission and scholarly ethos. Rather than seeking to maximize the self-interest of making its licensing program appear “successful” through “big hit” revenues, such a university would seek to maximize its overall goodwill by participating broadly in innovation activity, aiming to recover its expenses, and allowing others to contribute beyond this to reward the activity and encourage additional support for emerging innovation. The weaker IP position may not be the best for patent attorneys or for rent-seeking agents, and from their perspective, anything less is foolish, fluffy, and feckless. For them, it’s a power play all the way. The weaker IP position, however, is by far the stronger innovation and money position for universities. It’s time university leadership wake up to this realization and change their direction.
While the Boldrin and Levine paper seems slapped together and defective in a lot of places, it does open up the discussion of whether we have overshot the sweet spot in IP rights for innovation. That is worth discussing, and it is worth considering how we could gain by weakening rather than expanding and strengthening IPR regimes. I just don’t see the case, yet, for abolishing. For that, Boldrin and Levine will have to make an argument about what companies will do in the absence of patent rights (in the US, but facing those rights everywhere else in the world).