New legislation could force trolls to pay millions of dollars in legal fees when they lose in court. That could go a long way toward killing their business model. 

Watch out, troll. (USPTO, Becca Rosen)

It’s hard to think of any business more inherently obnoxious than a patent troll. These are the tech-world parasites that buy up troves of intellectual property, not so that they can make a product, but so that they can turn around and sue successful companies for patent infringement with the aim of nabbing a quick and profitable settlement. They’ve infested the courts over the last decade, and by one count are now responsible for more than half of all U.S. patent cases, potentially costing American businesses some $ 29 billion a year.

So kudos to Oregon congressman Peter DeFazio, who Thursday morning introduced a bill aimed squarely at putting the trolls out of commission. His smart and simple legislation, called the SHIELD Act, would force trolls that lose in court to reimburse the companies they sue for their legal fees, which can amount to millions of dollars. That might not sound particularly bold. But it’s a carefully calibrated step that could go a long way to containing the the troll problem by driving up the cost — and risk — of bringing flimsy patent cases. 

Trolls have flourished over the last few years largely because it’s now easier and cheaper to bring a patent case than it is to defend against one. Much like personal injury lawyers who advertise on TV, the attorneys who represent trolls often work on contingency, meaning they only take a cut of what they win. Defense lawyers, on the other hand, ask for their pay up front, and usually bill by the hour. As a result, a single troll can file a barrage of lawsuits without putting much skin in the game, while the small companies they tend to target — about 55 percent of the businesses sued make less than $ 10 million a year — are forced to mount a costly defense that saps their finances with each passing day. 

The upshot is that it’s usually cheaper for companies to settle with a troll than take them to trial. And settle they do. According to a study by professors at the University of Texas at Austin and Stanford, about 90 percent of cases brought by trolls — which academics politely refer to as “non-practicing entities” or “patent assertion entities” — end in a deal.

That’s crucial to their business model. The fact is that when their cases do go to trial, trolls overwhelmingly lose. The Texas-Stanford study found that when suits don’t settle, trolls win just 9 percent of the time. A PricewaterhouseCooper study found that they prevail about 24 percent of the time — somewhat more impressive, but still not great odds. These companies are essentially betting that they won’t have to justify their junk demands in court. 

DeFazio’s bill makes it more likely that they will. Companies facing particularly weak lawsuits will be able to fight back confident that when the final ruling comes down, their defense costs will be covered. And the more companies that feel capable of mounting a defense, the dicier business becomes for trolls. First, defendants in infringement suits can ask the U.S. Patent and Trademark office to re-examine and possibly invalidate the patent at issue. That process used to take years and years, but was massively sped up as part of the the patent reform bill signed by President Obama. Second, if a case proceeds to court and a judge or jury ultimately finds that a patent is invalid, it can never be used in another lawsuit, thanks to a legal doctrine known as “offensive non-mutual collateral estoppel.” In short, the more often someone calls a troll’s bluff, the more chances there are that a judge, jury, or government bureaucrat will toss out their IP claims for good. 

There’s also a more subtle way DeFazio’s bill throws a kink into the troll business model. At the start of each patent suit, the plaintiffs will have to show that they are either a university, the original inventor of the patent, or a company sincerely trying to turn it into a commercial product. If they can’t, they will be officially deemed a troll, and be required to post a bond to cover the defendant’s costs, should they lose a case. That will tie up their money, which in turn will make it more cost-intensive to bring lots of suits while simultaneously cutting their return on investment. 

The SHIELD Act is not a perfect solution. Some trolls will undoubtedly do the cost-benefit analysis and decide it’s worthwhile to stay in business. It may encourage some of them to target small companies even more exclusively, since the upfront costs of a defense might still be too onerous for them to contemplate going to trial. The penalty it imposes could be much more severe; companies that are found to willfully infringe a patent pay triple damages, so why shouldn’t trolls pay triple defense costs? It also does nothing about the underlying problem that has fueled the growth trolling: the comically broad nature of software patents, which are involved in 62 percent of troll suits. 

But during a short conversation, DeFazio told me his bill wasn’t to eradicate the problem entirely. Rather, it was to come up with a fix capable of drawing broad support from industry while withstand a legal challenge. “I’m certain there will continue to be some abuse or potential abuse out there, but I think this will eliminate a lot of it,” he said. 

He’s right that it should. Hopefully Congress is smart enough to pass it.