Erik Hovenkamp has authored a very interesting working paper comparing bottom feeder trolls (as we’ve seen here and here) to the practice of predatory pricing. This paper will likely be very important to the patent debate for three reasons: (1) Legal rulings concerning the intersection of patent law and antitrust law are scarce, therefore it is important to be able to analyze problems in an established area of antitrust law; (2) Hovenkamp mathematically proves that a patent troll can make money by setting licensing fees around the cost of litigation even if the underlying patent is valueless; and (3) Hovenkamp shows how the bottom feeder troll strategy can generate greater social harms.
Predatory pricing is the strategy of pricing goods or services below cost in the hopes of driving competitors out of the market and then raising prices to supra-competitive levels. Hovenkamp notes that like predatory pricing, bottom feeder trolls will litigate patent infringement cases even when they are likely to lose in order to generate a reputation for aggressive litigation. In fact, patent trolls have been found to use the same patents over and over again (8 or more times) despite only winning less than 10% of these cases. Once the reputation is established, defendants can no longer rely on patent trolls rationally deciding to drop loser lawsuits and must calculate the costs of litigating a full trial in making the decision to settle. Hovenkamp calls this strategy of short run losses to generate supra-competitive licensing fees predatory patent litigation.
The predatory pricing and predatory patent litigation strategies share a key common trait – they do not make economic sense unless you factor in the future value of supra-competitive pricing. However, these strategies also have an important difference. For predatory pricing to work, and consequently be illegal under antitrust laws, there must be a dangerous probability of recouping the investment in below-cost pricing. Otherwise, there is a net social gain in the strategy’s failure. This prong of the test is often difficult to meet because a company must not only drive out current competitors, but also prevent future competitors from entering the market. Patent trolls do not face this problem because the patents they hold are unique and do not suffer from competition by any other patent holders. Indeed, the more patent trolls employing the strategy the more severe the threat of litigation becomes.
Hovenkamp, an economist, proves the profitability of predatory patent litigation based on these circumstances. The result of Hovenkamp’s proof is that the licensing value of any patent is the cost of litigation + (probability of plaintiff victory X court awarded license cost) (see section 2 and appendix of Hovenkamp’s paper for the proof). Please note that even a patent with a 0% probability of success has substantial value based on the high costs of litigation.
The fact that a defendant’s litigation costs can be a substitute for patent quality has social consequences beyond the licensing costs to defendant victims. Predatory patent litigation creates a market for otherwise valueless patents. This motivates individuals and corporations to patent anything which has a litigation value greater than the cost to patent. A market for low quality patents also exacerbates the problem of patent thickets, which are a “dense web of overlapping intellectual property rights that a company must hack its way through in order to actually commercialize new technology.” Predatory patent litigation ultimately deters innovation because the only defense to an onslaught of valueless patent litigation is to not practice in a marketplace.
Hovenkamp ultimately proposes the solution of a litigation cost-sharing agreement (LCSA). LCSAs would lower the cost of each defendant’s litigation and increase the motivation to go to trial. This would increase the losses a bottom feeder troll would have to incur to the point where it would no longer be likely to recoup the losses later through supra-competitive pricing. It should be noted that such an agreement has already been tried by defensive patent aggregator RPX. RPX established a joint defense agreement as a service to clients who wished to protect themselves from patent trolls. Ironically, patent troll Cascades Computer Innovation LLC brought an antitrust suit against RPX for this agreement as a “concerted refusal to deal.” If successful, Cascades’ suit would be an unusual case of the letter of antitrust law being used to protect a violation of the spirit of antitrust law. The Supreme Court has formerly rejected such misapplication of antitrust law when it established the doctrine of antitrust injury.
The fact that litigation costs can be a driver of patent value should not sit well with anyone, especially when there are so many tactics to increase those costs. Hovencamp’s paper also belies the claim that patent trolls are helping inventors get compensation; companies aren’t paying for the use of inventions, they’re paying in order to avoid litigation costs. Moreover, litigation costs not only distort the patent market, they can actually create a market where none previously existed.