patent trolls and innovation
Other than financial measures (like ROA) I can’t think of another firm-level variable that is more commonly used in organizational studies than patent activity. Patents are used to track everything from innovation to technological niches to social networks among scientists. Patents are an all-purpose measure because we think they are tightly linked to creativity and knowledge production, the engine that drives both science and capitalist enterprise. But what if this is increasingly not true? What if patent use is becoming decoupled from creativity?
This is one of the questions posed made by last week’s This American Life, my favorite NPR show and one of the most consistently interesting programs of journalism out there. The show talked about patent trolls – companies or individuals who acquire patents for the primary purpose of suing other actors who might use technology that potentially infringes on that patent. The show focused on the firm, Intellectual Ventures, and its founder Nathan Myhrvoid. Through a couple of interesting vignettes and sly investigations, they showed how the company uses lawsuits, brought by a number of shell companies, to get large settlements out of technology companies, some of which are struggling enterpreneurial groups. The show demonstrates how, rather than protect and promote innovation, increasingly patents are being used to stifle innovation by wiping out or financially weakening companies that are actually trying to bring innovation to the marketplace. Meanwhile, patent trolls sit on those patents and do nothing to advance the innovations.
This must have some implications for our current understanding of patents as indicators of creativity and innovation. One of the startling revelations in the program was just how much redundancy there is in the patent system. The number of patents issued that cover the same basic function is often in the thousands, especially in the software industry. Patents may be more indicative of turf wars than they are of real innovation.
Even if you’re not a technology scholar, I highly recommend that you listen to the podcast of the show.
orgtheory.net 
Agreed, it was a great episode and well worth a listen. I considered blogging it but you beat me to it.
The thing that most infuriated me was at the 45:35 mark when Intellectual Ventures could only come up with two extremely vague examples of cases where they actually created value by bringing a patent into production as compared to thousands of cases where they created a deadweight loss for society by demanding royalties from someone who had independently put into production an idea similarly to an overly broad and overly obvious patent in their arsenal. At that point I felt like screaming at my ipod “what part of ‘to promote the progress of science and useful arts’ does Congress and the courts not understand?”
Finally, worth noting that it’s free right now but will cost a buck once the next episode comes out.
Also, there are are also “sample trolls” in music and this has totally changed hip hop. See Creative License by McLeod and DiCola.
LikeLike
gabrielrossman
July 26, 2011 at 9:49 pm
Great episode. Really excellent investigative reporting. Interesting factoid — Neal Stephenson is employed by Intellectual Ventures. I know he’s a divisive figure, but very interesting given Stephenson’s public writings on hacker culture.
NB: This American Life is a production of PRI, Public Radio International, and not NPR (though the Planet Money team is a joint venture).
LikeLike
Trey
July 26, 2011 at 10:48 pm
Nathan Myhrvoid
Indeed.
Neal Stephenson is employed by Intellectual Ventures.
As if I needed any more reasons to dislike the guy.
LikeLike
Kieran
July 27, 2011 at 1:05 am
Like Gabriel, I was left disgusted and furious in equal measure by the parts of the episode where the IP lawyers, facing the accusation that they were nothing but bottom-feeding parasites with zero interest in the fostering of innovation or the defense of inventors, had nothing to offer in their defense but craven evasions and sheer cant.
LikeLike
Kieran
July 27, 2011 at 1:10 am
Kieran, I was thinking of that post when I said “divisive figure.” I really enjoyed Cryptonomicon and Anathem, though I did not care for the Baroque Cycle.
Best part of the episode, though, is where the lawyer states that he will not be able to answer any questions about his clients, including about their corporate ownership, but then finishes with the statement that he loves NPR and ‘what they do.’
LikeLike
Trey
July 27, 2011 at 2:59 am
http://www.dklevine.com/general/intellectual/against.htm
LikeLike
RandomDude
July 27, 2011 at 3:38 am
@Kieran: You are my soul brother – I thought I was the only one who hated Stephenson!
LikeLike
fabiorojas
July 27, 2011 at 4:05 am
Having began my career in patent-related stuff, I am quite sceptical about the measure in any case. One thing that occurred to me is that maybe the causal relationship between patenting and alliances is really down to alliances incentivizing the legal protection of ideas. Patents are relatively costly to acquire, and it is trivial for an electronics company to increase its patenting rate by simply investing more money into the activity.
BW covered a speech recognition company that filed patent lawsuits against all companies it wanted to acquire in order to push down their valuations and, in some cases, force the owners to sell their company because it would be impossible to secure venture capital with a (fake) lawsuit on-going.
This is an excellent case for teaching about competitive stratetegy and business ethics:
http://www.businessweek.com/magazine/content/11_22/b4230037736600.htm
LikeLike
Henri
July 27, 2011 at 6:37 am
As a practicing patent litigator, I’m very glad to see orgtheory take an interest in these issues. Some might be interested in how Myhrvold’s perspective (not a response to the TAL, but interesting in its own right):
http://www.bloomberg.com/news/2011-07-20/patents-are-very-valuable-tech-giants-discover-nathan-myhrvold.html
The measurement problem — i.e., using patents as a proxy for innovation — is actually the mirror image of a problem I face everyday: how much is a fair price to pay for a patent or patent license? Sociologists could make a huge contribution to patent law by coming up with a better way to measure innovation than number of patents generated, thereby permitting patents an independent valuation not tied to the cost of litigation. Please help!
Even thinking this through in the most bone-headed way, a few tentative conclusions seem reasonable. Whatever standard for valuation is adopted will have to be drawn (as a practical matter) from public or private financial statements. But even when revenue streams are broken out into individual products or versions of products, quarterly financials can only discriminate semi-annual or slower trends.
In short, there is no way to solve the innovation measurement problem without changing the way financial statements are reported.
http://brokensymmetry.typepad.com/broken_symmetry/2010/03/how-do-you-measure-innovation-without-reference-to-patents.html
LikeLike
Michael F. Martin
July 27, 2011 at 5:36 pm
One other comment — to be fair to Peter Detkin — the PTO website that lists the title history for patents is a total mess. They were inferring too much from his refusal to answer questions about a chain of title he hadn’t looked at before.
LikeLike
Michael F. Martin
July 27, 2011 at 6:22 pm
Michael,
Asking how we can contribute to more effective adjudication of patent disputes reminds me of Reagan’s response when Richard Allen asked him about his Soviet policy: “We win and they lose.”
I say this for a few reasons grounded in my understanding of the sociology of science:
1) Tacit knowledge is more important than explicit knowledge, which by extension means that the kind of knowledge described in patents is less important than the “work the bugs out” kind of stuff that only comes with putting ideas into production
2) Individual authorship is largely a myth — knowledge production is thoroughly socially embedded
3) Incentives matter, but so do transaction costs and uncertainty
4) As an empirical point, the frequent re-invention of concepts covered by patents is pretty good evidence that we are over-estimating incentives and underestimating obviousness
5) Innovation per capita is stagnating despite substantial investments in science. This may reflect an objective technological plateau (Cowen’s argument in TGS) but if we’re lucky this reflects bad institutions, which we can fix
All this suggests much more radical reform of the patent system than just finding new metrics for valuation. At the least it suggests that far more patents be stricken on grounds of obviousness, either on initial application by USPTO or by courts when the issue comes to litigation. More radically I’m thinking of policy implications like anti-trust waivers to solve the collective action problem of challenging nuisance suits or moving software back to copyright (preferably with only a short term) as the “idea vs expression” standard would solve a lot of issues around over-broadness and independent invention.
LikeLike
gabrielrossman
July 27, 2011 at 8:27 pm
Great points:
1) is actually a hot topic among patent law scholars right now. Received wisdom is that patents don’t really enable squat. That’s the obvious conclusion to anybody who has to read these things regularly, right?
Alternative view is that the filing of the patent creates permits more permeability between organizations that would otherwise keep knowledge trade secret. There is much to be said for this view, and for the dream more generally of a patent office as a repository of knowledge. The last point in particular always comes to mind when I see Google attacking the patent system. Perhaps they view it as their most serious competitor in the sense that they share the same mission (i.e., to index and make available the world’s knowledge)?
2) Agreed. But see 1)
3) No doubt. Hence the received wisdom that software patents suck whereas most startups in biotech and pharma space couldn’t survive without them. Transactions costs overwhelm the benefit of the bargain in software when you have to license thousands of patents to clear patent rights for most products. In pharma, 1 product = 1 patent (or maybe 2 if they get the enantiomer) and you can look up which patent in a book (“the Orange Book”) In fact the costs of gathering and processing information about validity and infringement are the major component of transactions costs for companies selling computer-related products. I think there are many reforms that could be adopted to address those costs (change the way these products are claimed to match the nomenclature of the field in question, for example). None of them are part of the reform bills passed in both house and senate. :-(
4) Agreed. Donald Campbell made the point years ago.
http://brokensymmetry.typepad.com/broken_symmetry/2009/04/donald-t-campbell-on-multiple-inventions.html
Now we see patent law profs making the same observation
http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1856610
5) Disagreed with both the factual premise and the conclusion.
From what I can tell, there has been much less investment by the government in basic science over the past few decades than there was during the cold war. In part, this is because private institutions (including venture capital) have stepped in to do this funding. Problem is that private investors tend to make highly correlated decisions (part of principal agent problem of taking other peoples’ money). Result is less innovation, with a few huge hits.
But one of those huge hits is this communication technology that’s making this converation between us possible. The repercussions of the last major transition in how we communicate (i.e., the printing press) were not fully appreciated for decades, even centuries. Even if we assume that process will be accelerated, there has not been time for institutions to catch up with the new realities of the Internet.
If you buy the story Andy Clark tells about how the technology we use to record and communciate actually extends our cognitive capabilities — and I find it hard not to since I can’t even do simple arithmetic without a paper and pencil — then I think you have to be optimistic about what kinds of innovation we’ll see in our lifetimes despite the relative lack of investment, optimistic because we are literally learning faster than ever before.
LikeLike
Michael F. Martin
July 27, 2011 at 9:09 pm
Michael,
Just want to agree with you that it’s very important to distinguish between areas. The patent system seems to be a much better fit with pharma than software for exactly the reasons you listed. By extension, we can only imagine that the entertainment industry would be an even worse fit with patents than is software. Hollywood would grind to a halt if Warner Brothers were filing patents for “A plot for structuring narrative fiction in which boy meets girl, boy loses girl, and boy gets girl back” or “A method for developing screenplays by adapting comic book properties.”
LikeLike
gabrielrossman
July 27, 2011 at 9:27 pm
If only you knew how deep the rabbit hole goes…
http://www.google.com/patents?id=ROenAAAAEBAJ
LikeLike
Michael F. Martin
July 27, 2011 at 9:49 pm
Tim Lee‘s reaction to the TAL episode is worth reading. Basically, he has lost hope that Congress will make good policy if it injures well organized stakeholders and the best hope for reform is that SCOTUS (read: the liberals plus Scalia) uses some old precedents about “algorithms” to strike software patents entirely. I wish I could share his optimism but I can’t given the wishy washy nature of Bilski and the disregard SCOTUS showed for the Constitutional language of “promote” and “limited” in Eldred. Also as red meat for the Gladwell haters (which in general I’m not), Lee has an earlier post about how badly Gladwell misread IV.
On a side note, it’s fun to think about what’s more likely: that SCOTUS will strike software patents or that ASA will file an AC on an issue that we have expertise about and which structures a sizable portion of the economy but isn’t the kind of thing that people scream about on cable news or put in direct mail.
LikeLike
gabrielrossman
July 28, 2011 at 5:21 pm
Thanks very much for this discussion guys…It is going to be very useful for my actual PhD work.
LikeLike
jfespino
July 29, 2011 at 9:19 am
To offer a counter-point, it is easy to express frustration at the patent system and those entities that seek royalties for patents that are not practiced by the enforcer. What everyone seems to forget, however, is that a patent is meant to promote the advancement of science and innovation. By forcing others to avoid a patent, the system requires innovators to take the next step forward and develop the next big thing. Too often “innovation” is really just a very marginal improvement on an existing technology. Patents require inventors to go beyond what has been done in the past and think forward.
There are many legal mechanisms protecting legitimate manufacturers from overly broad patents. These include a number of invalidity doctrines that both the courts and the USPTO have been strengthening in recent years. While the system is far from perfect and requires many significant modifications, I don’t think patent trolls are at the top of the list of problems. The real problem is the USPTO’s willingness to issue broad patents with very little support in the applications for their claimed breadth. If the USPTO did its job more effectively, there would be fewer trolls to contend with. Proposals are being made in Congress to address some of these issues, including a European-style system that allows people to challenge the validity of patents before they issue (and can be enforced). Hopefully this will stem the tide.
LikeLike
Eric Martin
August 4, 2011 at 3:30 pm