Site Meter

Tagged: innovation


Human Capital, Tacit Knowledge, and the Limits of Intellectual Property

Orly’s ambitious and thought-provoking book covers a significant amount of intellectual ground. She deftly navigates covenants not to compete, nondisclosure agreements, trade secrets, and intellectual property assignments to provide a compelling argument for the free flow of talent in the modern economy. Orly’s work raises a host of questions that space constraints no doubt prevented her from more fully exploring, and I would encourage her to extend her analyses in subsequent work.

One aspect of Orly’s work that I found particularly intriguing is that it reveals a central irony about information. The title of her book is a play on Stewart Brand’s famous phrase “information wants to be free.” While this statement has a contemporary ring, the observation that information is “slippery” and readily appropriable has a long pedigree and has had significant legal and policy ramifications. As Orly notes, Thomas Jefferson invoked the freely appropriable nature of technical information to help justify exclusive rights on inventions. More formally, economists have long characterized technical knowledge as a public good that is nonrival, nonexcludable, and capable of nearly costless transmission. The “slipperiness” of technical information is now largely taken for granted and provides significant theoretical justification for exclusive rights on knowledge assets. Indeed, IP scholars such as Polk Wagner have argued that information’s natural tendency to slip through cracks and build upon itself should alleviate concerns that strict exclusive rights can bottle up knowledge. Information, after all, wants to be free.

Orly’s account of the talent wars, however, reveals that much information does not naturally want to be free. As Orly recognizes, much technical information is tacit and personal to a particular creator or inventor. Such tacit knowledge takes the form of intangible know-how that is difficult and sometimes impossible to codify. Importantly, even when an invention is disclosed in a patent, much valuable technical knowledge related to that invention often remains tacit and is not formally shared. The inadequacy of patent disclosure and the difficulty of transmitting tacit knowledge create a need for companies licensing patents to somehow obtain this information. This is evident, for example, in university patenting and technology transfer, a field that Orly addresses. Empirical accounts of academic technology transfer show that private companies, in parallel to licensing university patents, often seek direct interactions with faculty inventors precisely to obtain their patent-related tacit knowledge.

The tacit, sticky nature of technical information relates to another theme that permeates Orly’s work: agglomeration economies and the importance of place. In theory, patents adequately disclose the inventions they cover, which has the effect of reducing transaction costs in licensing negotiations. Among other implications, such ex ante disclosure should make licensing negotiations less sensitive to geographic proximity; at least with respect to appropriating technical knowledge, a potential licensee should not have a great need to interact directly with an inventor, for the patent itself discloses the technology. However, empirical studies of academic licensing show that licenses tend to cluster around licensor universities. To be sure, a host of factors helps explain such clustering, from universities’ commitment to local economic development to the spatially concentrated nature of professional networks (a theme that Orly also highlights). But the need for faculty inventors to literally sit down with licensee firms to convey patent-related tacit knowledge also contributes to such agglomeration. While some information can be transmitted by reading a patent a thousand miles away, sometimes transferring patent-related technical knowledge requires side-by-side demonstrations of a new technology or that ever-valuable personal conversation over a cup of coffee.

In subtle ways, Orly’s work thus offers a cogent exposition of the limits of patent law and formal technology transfer. In theory, the patent system provides a public repository of technical knowledge from which all can draw in their innovative pursuits. At the very least, licensees themselves should be able to rely on the disclosure of patents to adapt licensed inventions for commercial use. However, much information is not freely appropriable. Even when an invention is disclosed, much information remains tacit and personal to the inventor. Thus, patents are inherently limited as a vehicle for disclosing and transferring technologies, thereby creating a need for much costlier, geographically constrained, tacit knowledge transfer between individuals.

In a broader sense, Orly’s observations highlight an interesting paradox about the “freedom” of information. In the classic economic account, the ease of appropriating technical information represents a problem. This problem is resolved by subjecting technical information to exclusive rights, thus shoring up incentive to invent. However, Orly’s study reveals that much information is subject to a different problem: it is too difficult to appropriate, as it resists formal codification and disclosure. This creates a need for a different type of policy intervention, one that focuses on enhancing the mobility of the underlying sources of information—people—rather than information itself. Paradoxically, the fact that much information is not truly free provides all the more reason that the talent generating that information should be.


What Are the Limits to What Hackers Produce?

I’m writing this from an airplane somewhere over the US-Canadian border. I forgot my copy of Coding Freedom at home, and was cursing my ineptitude. But then it occurred to me that, given the subject, I could probably find a copy online. Sure enough, I downloaded a pdf via the airport wifi. (For free! – those Canadians…).

This, in the most mundane of ways, is a simple reenactment of what Gabriella Coleman writes of so compellingly in her new book. Gabriella, inspired no doubt in part by her years of exposure to hacking culture, struck a deal with her publishers. The resulting CC license gives all of us who might want to read the book more freedom to do what we want with it – read it on any device, search it, and even pull it up in an airport so we can file a nearly-too-late contribution to a terrific online discussion. Gabriella didn’t know I’d forget my book at home when she decided to negotiate the license. But she did have the sense – I assume – that she needed something more than copyright law to help her achieve what she wanted from her book. Which was in part to give to the rest of us more freedom than standard copyright law would allow.

But how far does that freedom go? This is surely one of the most important and interesting questions about this new form of making software, and the new legal forms that attend it. So that’s what I want to focus on here. One of the book’s great strengths is the spectacularly detailed and clear-eyed account that it provides of hacker culture, or at least a certain hacker culture. As it points out, this is a culture that is built upon a deep commitment to the pleasures of technology (like Ed Felten, I loved the bit on hacker humor), a ferocious conception of self-help and meritocratic ordering, and also to an overt aversion to things “political.”

As a few others have in the course of this discussion, I wonder too about the limits of a form of practical revolution that starts here. How far can this new mode of production take us, if it is characterized by technoelitism, an aversion to politics, and by a subject position that is decidedly fairly privileged and high-skilled? After all, you can’t be part of this crowd and lack access to a computer and internet connection, or be bereft of free time.


Read More


Taxonomy of Innovation Incentives

SIP folks don’t talk enough with tax-law folks, and vice versa. This has some unfortunate results. IP has become a leading tax-avoidance vehicle, without drawing sufficient notice from IP scholars and practitioners. And R&D tax incentives are rarely evaluated alongside patents, prizes, and research grants as effective ways to foster innovation.

An insightful article forthcoming in the Texas Law Review, by Daniel Hemel and Lisa Larrimore Ouellette, takes a big step in bridging this gap. They observe that all innovation incentives can be broken down along three dimensions: (1) who decides (government vs. the market), (2) when paid (ex ante vs. ex post), and (3) who pays (government vs. users). For example, patents are market-driven, with money delivered ex post, from users of the patented technology. By contrast, R&D tax incentives are market-driven, with money delivered ex ante, from the government.

These three dimensions lead to a 2 x 2 x 2 matrix, suggesting a total of eight types of innovation incentives. But only five are currently used: patents, prizes, research grants, R&D tax incentives, and patent boxes (which provide favorable tax rates on patent income). As a result, Hemel and Ouellette’s taxonomy suggests three new mechanisms to encourage innovation. Their taxonomy also teases out some exciting new insights on the relative merits of existing innovation incentives, including some previously overlooked benefits of R&D tax incentives.



Who Knew? Patents Don’t Really Promote the Useful Arts

Golf Tee Patent2Andrew Torrance and Bill Tomlinson have a paper out that challenges “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.” The paper, Patents and the Regress of Useful Arts, “employ[s] 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.” In other words, the two have taken the idea of a Sim and created PatentSim “to simulate the innovation process in one of three scenarios: a patent system, a “commons” system with no patents, or a system with both patents and open source protection.” In the words of Johnny Carson wild, weird stuff, and quite fascinating too. Under their model and testing system it appears “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.”

This post at Against Monopoly has a nice summary of some of the major articles on the topic of patents and innovation. Which reminds me, folks interested in empirical research should take a read of the paper too as it is trying to fill a gap by testing the innovation assumption in patent theory.

Image: World’s first patent for a golf tee; British patent #12941 of 1889, by Bloxsom & Douglas
Source: WikiCommons
The image size is reduced for our site, but go to the original to see/read the patent.