by James Bessen
Over the last two years, much has been written about patent trolls, firms that make their money asserting patents against other companies, but do not make a useful product of their own. Both the White House and Congressional leaders have called for patent reform to fix the underlying problems that give rise to patent troll lawsuits. Not so fast, say Stephen Haber and Ross Levine in a Wall Street Journal Op-Ed (“The Myth of the Wicked Patent Troll”). We shouldn’t reform the patent system, they say, because there is no evidence that trolls are hindering innovation; these calls are being driven just by a few large companies who don’t want to pay inventors.
But there is evidence of significant harm. The White House and the Congressional Research Service both cited many research studies suggesting that patent litigation harms innovation. And three new empirical studies provide strong confirmation that patent litigation is reducing venture capital investment in startups and is reducing R&D spending, especially in small firms.
Haber and Levine admit that patent litigation is surging. There were six times as many patent lawsuits last year than in the 1980s. The number of firms sued by patent trolls grew nine-fold over the last decade; now a majority of patent lawsuits are filed by trolls. Haber and Levine argue that this is not a problem: “it might instead reflect a healthy, dynamic economy.” They cite papers finding that patent trolls tend to file suits in innovative industries and that during the nineteenth century, new technologies such as the telegraph were sometimes followed by lawsuits. But this does not mean that the explosion in patent litigation is somehow “normal.” It’s true that plaintiffs, including patent trolls, tend to file lawsuits in dynamic, innovative industries. But that’s just because they “follow the money.” Patent trolls tend to sue cash rich companies, and innovative new technologies generate cash.
The economic burden of today’s patent lawsuits is, in fact, historically unprecedented. Researchshows that patent trolls cost defendant firms $29 billion per year in direct out-of-pocket costs; in aggregate, patent litigation destroys over $60 billion in firm wealth each year. While mean damages in a patent lawsuit ran around $50,000 (in today’s dollars) at the time the telegraph, mean damages today run about $21 million. Even taking into account the much larger size of the economy today, the economic impact of patent litigation today is an order of magnitude larger than it was in the age of the telegraph.
Moreover, these costs fall disproportionately on innovative firms: the more R&D a firm performs, themore likely it is to be sued for patent infringement, all else equal. And, although this fact alone does not prove that this litigation reduces firms’ innovation, other evidence suggests that this is exactly what happens. A researcher at MIT found, for example, that medical imaging businesses sued by a patent troll reduced revenues and innovations relative to comparable companies that were not sued. But the biggest impact is on small startup firms — contrary to Haber and Levine, most patent trolls target firms selling less than $100 million a year. One survey of software startups found that 41% reported “significant operational impacts” from patent troll lawsuits, causing them to exit business lines or change strategy. Another survey of venture capitalists found that 74% had companies that experienced “significant impacts” from patent demands.
Three recent econometric studies confirm these negative effects. Catherine Tucker of MIT analyzedventure capital investing relative to patent lawsuits in different industries and different regions of the country. Controlling for the influence of other factors, she estimates that lawsuits from frequent litigators (largely patent trolls) were responsible for a decline of $22 billion in venture investing over a five-year period. That represents a 14% decline.
Roger Smeets of Rutgers looked at R&D spending by small firms, comparing firms that were hit by extensive lawsuits to a carefully chosen comparable sample. The comparison sample allowed him to isolate the effect of patent lawsuits from other factors that might also influence R&D spending. Prior to the lawsuit, firms devoted 20% of their operating expenditures to R&D; during the years after the lawsuit, after controlling for other factors, they reduced that spending by 3% to 5% of operating expenditures, representing about a 19% reduction in relative R&D spending.
And researchers from Harvard and the University of Texas recently examined R&D spending of publicly listed firms that had been sued by patent trolls. They compared firms where the suit was dismissed, representing a clear win for the defendant, to those where the suit was settled or went to final adjudication (typically much more costly). As in the previous paper, this comparison helped them isolate the effect of lawsuits from other factors. They found that when lawsuits were not dismissed, firms reduced their R&D spending by $211 million and reduced their patenting significantly in subsequent years. The reduction in R&D spending represents a 48% decline.
Importantly, these studies are initial releases of works in progress; the researchers will refine their estimates of harm over the coming months. Perhaps some of the estimates may shrink a bit. Nevertheless, across a significant number of studies using different methodologies and performed by different researchers, a consistent picture is emerging about the effects of patent litigation: it costs innovators money; many innovators and venture capitalists report that it significantly impacts their businesses; innovators respond by investing less in R&D and venture capitalists respond by investing less in startups. Haber and Levine might not like the results of this research. But the weight of the evidence from these many studies cannot be ignored; patent trolls do, indeed, cause harm. It’s time for Congress to do something about it.
