To understand why privateering is so concerning, one must first comprehend the uneasy state of patent détente that, at the best of times, characterizes the software industry and much of computing in general. Software firms unavoidably operate in a state of perpetual patent infringement. Over 100,000 new software patents are granted every year, many for ideas that are not very complicated and likely to be independently invented. It is simply impossible for a software firm to know for certain that it is not infringing upon any patents. As a result, most software firms have given up on trying to know. Software engineers are actively discouraged from conducting patent searches because willful infringement would subject the company to treble damages if proved in court.
With every firm potentially infringing on every other firm, software companies live, at the best of times, in a state of mutually assured destruction. If one company sues, then it is likely to face a counter-suit. If nothing destabilizes the equilibrium, then the software industry is free to innovate without the burden of worrying about infringing upon patent rights.
The MAD equilibrium is reinforced to a considerable extent by the rules of various standard-setting organizations (SSOs). Computing standards are so complex that they frequently implicate hundreds or thousands of patents. In order to prevent ex post opportunism, SSOs typically require that their members license standard-essential patents on “fair, reasonable, and non-discriminatory” (FRAND) terms.
Finally, software companies are frequently sensitive to the public relations implications of suing their rivals, who often make beloved products. The reputational costs of aggressive patent litigation are often high. Consequently, the amount of litigation in the software industry is not proportional to the amount of infringement, which is absolutely ubiquitous.
This unpleasant but moderately workable equilibrium has been seriously eroded by the practice of patent privateering. While the term “privateering” mixes pirate and troll metaphors, it denotes the transfer of patents—and outsourcing of patent assertion—to “non-practicing entities,” companies that exist solely for the purpose of collecting patent license revenue.
Because patent trolls are non-practicing, they are not subject to threats of counter-suit and mutually assured destruction. Because they are not members of any SSOs, they do not have any obligation to license on a FRAND basis; standard-essential patents can be transferred to privateers and then asserted against all users of the standard. And because the transfer of patents to patent trolls is often done through various shell companies or other shadowy means, the defendant and the public often cannot know on which practicing software company’s behalf the privateer is working. This means the defendant cannot retaliate through countersuits or a public relations offensive.
One recent example of patent privateering is recounted by the lawyers who represented Google in the matter. In 2011, Microsoft and Nokia transferred around 2000 mobile phone patents, including 1200 standard-essential patents, to a troll named MOSAID for a nominal sum of $19,975. As part of the agreement, Microsoft and Nokia receive two-thirds of the revenue stream associated with the portfolio. MOSAID has the obligation to maximize the value of the royalty stream; if targets are not met, Microsoft and Nokia can reclaim the portfolio for a payment of $10,000. Microsoft also retained a license for itself and its customers to use the patents. The patents appear to be generating a substantial new revenue stream, much of which would not have been collectible under the old equilibrium of mutually assured destruction, FRAND licensing terms, and public relations considerations.
Legislation currently under consideration in Congress, such as the discussion draft being circulated by Rep. Goodlatte, would address the problem of obscuring the real party in interest in any patent litigation. But that remedy only goes so far. Patent privateers would still, for instance, destabilize the non-litigious equilibrium through litigation threats and by evading FRAND rules.
Given the secretive nature of patent transfers to privateers, an FTC investigation—if that’s all it is—is welcome. Better data on the business models of patent trolls can be used to inform the policy debate in Congress. But an FTC investigation of this sort is frequently a precursor to agency action. And that is what is so strange. The FTC is the competition police. Its job is to prevent anti-competitive business practices. But the whole point of patents, rightly or wrongly, is to restrain competition, to give one company a monopoly over the use of an idea. It’s not clear that the FTC has the authority to weaken patent enforcement rights, overriding Congress’s explicit decision to make the economy less competitive.
And as beneficial as more information about privateering would be, simply based on what we already know, the right policy prescription for Congress is apparent. Given that the best status quo outcome, the fragile equilibrium we would have without patent trolls, is that software patents are widely ignored, we should simply do away with software patents. Software patents lead to an outcome that is no better, and frequently worse, than no software patents. Mutually assured destruction is better than nuclear skirmishes, but at some point it becomes time for Congress to take away everybody’s nukes.