The question of how to deal with patent trolls has plagued patent academics and practitioners both. But the term patent troll is an empty one, which is defined differently in each new blog post or article. One of the most common definitions of a patent troll is “an entity that obtains patents not as a basis for producing and selling goods, but instead primarily for obtaining licensing fees.” This is the definition of a non-practicing entity, and many policy makers and academics equate patent trolls and non-practicing entities. This definition is misleading and by focusing on the identity of the plaintiff it misdirects attempts to reduce excess patent litigation. Other definitions focus on the behavior, not the identity, of the plaintiff and define a patent troll as someone who sues or threatens to sue on broad, and often vague and likely invalid, patents. These behavior-based definitions more accurately explain the patent troll problem and are more useful when attempting to decrease excess patent litigation.

Focusing on behavior instead of identity makes sense because the damage of a patent troll comes from exploitative litigation not from owning but not practicing a patent. As Schwartz and Kesan say in Analyzing the Role of Non-Practicing Entities in the Patent System, “[t]he debate should be reframed to focus on the merits of the lawsuits, including patent system changes focusing on reducing transaction costs (e.g., lawyers’ fees) in patent litigation, instead of focusing solely on whether the patent holder is a non-practicing entity.” In The Direct Costs from NPE Disputes, a recent paper by Bessen and Meuer, the cost of NPEs on the American economy was estimated to be $29 Billion. But the paper shows (though it doesn’t recognize) that these costs don’t stem from the fact that it was an NPE who was suing, but from the fact that a party was suing a large number of parties on a likely invalid and overly broad patent.
The crucial difference between behavior-based and identity-based definitions is highlighted by examples of non-NPEs who have engaged in economically damaging and innovation stifling litigation. Working from the assumption that the sin of a patent troll is that it sues on likely invalid and overly broad patents, companies like Apple, Microsoft, and Samsung have all acted as trolls. Apple has been called a troll for many of its suits against Samsung, including suits for infringement of generic smartphone and tablet cases. Additionally, Apple has moved many of its patents to a third party company named Digitude Innovations, an NPE shell company. Rockstar Consortium is another patent-holding company set up by Apple and Microsoft, which engages in troll-like behavior. Samsung, the victim of Rockstar’s attacks, got into the patent trolling game when they created Intellectual Keystone Technology in 2013. The line between NPE and companies that make and use inventions has disappeared. The damage isn’t derived from who the plaintiffs are but what the suits are.
Universities also must be mentioned when discussing the definition of patent trolls. Universities create no products, and so, using an identity-based definition, are patent trolls when they attempt to enforce a patent. Some authors have attempted to distinguish between “good NPEs” and “bad NPEs” but this distinction only highlights the futility of defining a troll based on whether it makes a product. Universities, like any other institution can, and are, trolls when they act in a troll-like manner.
This isn’t to say that there is no relationship between NPEs and trolls. The NPE’s business model is to profit through patent licensing, settlements, and litigation, which promotes litigiousness. As found in paper mentioned above by Bessen and Meuer, it is true that most troll-like litigation is initiated to NPEs. Despite this, it is still important to distinguish between behavior and identity when attempting to solve the patent troll problem. Recent attempts have been made to curb patent trolling through legislation, but these attempts are doomed to fail if they focus on the plaintiff’s identity. For example, last year the Saving High-Tech Innovators from Egregious Legal Disputes (SHIELD) Act was introduced. This law, which was introduced partially in reaction to President Obama’s statements on curbing patent trolls, makes the losing plaintiff patent-owner pay the defendant’s litigation costs if the court finds that the plaintiff is not the inventor, is not exploiting the patent, or is not a university.
This act severely misunderstands the patent troll problem. The Act punishes NPEs for being NPEs, not for their actions. It excludes universities, though they can also be trolls, while punishing firms who are not necessarily trolls. Fortunately, this Act didn’t become law, but it shows the need for a firm behavior-based definition in order to solve the patent troll problem. This issue has becoming increasingly important to foreign jurisdiction as well. Companies have expressed concern about the European Union’s proposed Unified Patent Court and the potential for troll abuse in the new court system. A correct understanding of the problem is necessary for the EU to deal with this developing problem.
In truth, the term patent troll appears to only mean “someone suing me for patent infringement.” Apple, Microsoft, Samsung and others have all been called patent trolls. They have also all complained about patent trolls, claiming that others are the trolls and they are the innocent victims. In order for a realistic solution to be reached the patent community needs to agree on a behavior-based definition of a patent troll. Without that, scholars and academic will continue to talk past each other, using the same words but meaning vastly different things.
Michael Crowley is a J.D. candidate, ’15, at the NYU School of Law.

more »