A recently published paper - The Effect of Patent Litigation and Patent Assertion Entities on Entrepreneurial Activity - funded by the Computers and Communication Industry Association, a prominent supporter of US patent litigation reform legislation, has garnered wide publicity with a claim that “VC investment would have been at least $8.1 billion higher over the course of five years but for litigation brought by frequent patent litigators”; those “frequent patent litigators” being patent assertion entities (PAEs). The report’s author Catherine Tucker – who discloses in her CV that she has received up to $155,000 of funding from Google since 2009 - suggests two possible policy initiatives:
We find empirical evidence that “higher quality” patents, such as patents that meet the thresholds of patentability in the U.S., the E.U., and Japan, are associated with reduced levels of litigation. This suggests that one approach to stemming excessive patent litigation would be to ensure that the patent approval process in the U.S. uses similar quality thresholds to the processes in the E.U. and Japan.
We find that litigation by frequent litigators is associated with a direct and negative effect on innovation. Therefore, improvements to the patent litigation system which address the potentially harmful effects of such litigation should be considered. One potential remedy could be changing the current way that litigation costs are allocated; at present, there exist low barriers to bringing a lawsuit and currently defendants bear disproportionate risks and costs of being involved in patent litigation relative to plaintiffs.
Over on Ars Technica Joe Mullin, a fine journalist who has nailed his colours firmly to the pro-reform cause, states: “Turns out there is a very real, and very negative, correlation between patent troll lawsuits and the venture capital funding that startups rely on.” For his part, CCIA attorney Matt Levy claims: “For the first time, we have an economic model proving patent reform is not a zero sum game between protecting intellectual property and reducing abusive patent litigation … Professor Tucker’s research reveals the harms of skewing the patent system too far in favor of protecting low-quality patents.”
Having been through the study, though, I cannot agree with either Mullin or Levy. On my reading there are major flaws in its coverage that mean it actually adds very little to current debates. This is not because of what Tucker has included in the study, but what she has left out. Let me explain.
Clearly, Tucker has worked very hard on producing this paper. I am no statistician or mathematician, but the methodology looks tight enough to me and I am not going to argue with the idea that VCs would have ploughed more money into certain companies if they had not been hit with lawsuits by PAEs. That seems pretty self-evident to me – VCs, like any other kind of investor, dislike uncertainty.
However, what I did not see in the study is any evaluation of the merits of the cases brought by the PAEs Tucker writes about. Instead, I saw a few anecdotes about what seem like egregious cases, but nothing that demonstrated these were typical. It seems inarguable to me that PAEs willing to spend millions of dollars taking their cases to court when they cannot get someone to take a licence believe that their patents are being infringed and that they have a good chance of convincing the court to agree. Thus, it could just be that Tucker has spent her time and the CCIA’s money discovering that VCs are unlikely to sink money into companies whose products infringe patents. I could be wrong, of course; but we don’t know because Tucker does not look into it.
Furthermore, the patents that PAEs seek to license and are sometimes forced to litigate do not just appear out of the ether. Instead, they have either been granted to the PAE itself or they have been acquired. Either way, they are the fruit of someone’s labour and investment. Tucker seems to assume that if PAEs were not looking to get a licensing fee from them, then nobody else would be. That strikes me as quite a big assumption to make – and certainly one that needs some justification. There is none in the study.
On top of all this, if you are going to claim that Tucker’s findings support the case for legislative patent reform, you have to show that such reform will not cause less VC money to be invested. In other words, would a potential $8.1 billion investment shortfall under current circumstances turn into a much bigger one in a reformed system? Again, this is not something that Tucker explores – and neither do Mullin or Levy. However, we do know that in March the National Venture Capital Association – representing entities that hold 90% of venture capital under management in the US – wrote to Congressional leaders warning in the very strongest terms against the reform legislation that was on the table:
NVCA believes it is critical that Congress should take a carefully measured approach to making changes to the patent system because almost any such change is likely to alter the risk/benefit equation in ways that affect individual companies and specific industries in different ways. While some innovative companies have business models in which intellectual property does not play a significant role, for other companies patents and legal protection against well-established incumbent competitors are essential to survival. A “one-size fits all” approach to patent enforcement is likely to have unintended and unforeseeable consequences for some small companies and could have a significant impact on others to the point that incentives to invest are diminished or extinguished.
It could well be that introducing reforms that affect all kinds of patent owners, and which make it less risky to infringe rights while making it harder and more expensive to enforce them, could have a far more chilling impact on VC investment than the status quo. Again, however, we cannot make that judgement, because it is not something that Tucker looks into.
Then there is the idea that if the US had standards that matched Japanese and European levels of patentability there would be less litigation. Perhaps – but, as Tucker does not explain (or know?), that would also mean substantially reducing the scope of patentability in areas, such as software, which enjoy far less protection in Europe and Japan than they do in the US. That’s not an issue of quality, but of policy. And, let us remember, the US has the most innovative and successful software industry in the world – one that attracts very high levels of VC funding. The contrast with Europe is stark.
I could also go into the way in which Tucker cites both Bessen & Meurer and Professor Robin Feldman to back up the points she makes. But we have dealt with the problems posed by the work of both parties previously on this blog (here and here, for example), so there is no need to do it again. Suffice it to say, they should be used as sources only with the utmost care and in the knowledge that there are major flaws in the work that they have done.
All in all, therefore, this study does not come close to making a case for legislative patent reform. At best it shows that some companies are affected by PAE litigation and that this may have an impact on VC investment. But anyone with half a brain would have known that already. The real issue is whether new laws that affect all patent owners will make things better or worse. Tucker does not begin to address that; but we know that a wide coalition of parties have and have concluded that the proposals which were dropped from the Senate’s roster after being passed by the House did not pass muster.
None of this is to say that there are not problems and issues to address with regards to abusive patent litigation in the US, clearly there are. The trick, though, is doing it in the right way and doing the minimum amount of harm. Over recent weeks the Supreme Court has issued decisions on fee shifting and patent validity that may well make a difference, while market-based solutions are also emerging. It might not be ideal, but it is a whole lot better than passing sledgehammer laws based on anecdotes and flawed research. The US and its economy deserve a whole lot better than that.
Licensing, IP politics, IP litigation, Patents, IP business
There is a flaw in the Tucker's methodology, which you highlighted: correlation does not equal cause and effect. You learn that in Statistics 101. There could be many other explanations for the relationship. I like yours -- VCs don't want to invest in companies that infringe patents.Barry Leff, IP Navigation Group LLC on 18 Jun 2014 @ 14:36