Policy —

Former top patent judge: system works “quite well,” needs to be “stronger”

Prominent judges with radically different perspectives square off in debate.

The US Court of Appeals for the Federal Circuit at its creation in 1982. Chief Justice Warren Burger is third from the left in the front row.
The US Court of Appeals for the Federal Circuit at its creation in 1982. Chief Justice Warren Burger is third from the left in the front row.

People have radically different views of the patent system. Critics see it as an unmitigated disaster, while many in the patent bar don't understand what all the fuss is about. But it's rare for prominent advocates of these contrasting perspectives to engage each other directly.

But that happened on Wednesday in a teleconference debate hosted by the Federalist Society. Defending the status quo was Judge Paul Michel. Until 2010, Michel was the chief judge of the Federal Circuit Court of Appeals, a post that made him the nation's top patent judge. He debated Judge Richard Posner, a Chicago-area appeals court judge who threw out a patent lawsuit between Apple and Motorola earlier this year, complaining that the patent system had descended into "chaos."

"A failed system"

Because Congress gave Judge Michel's Federal Circuit jurisdiction over all patent appeals in 1982, Judge Posner's own Seventh Circuit Court of Appeals doesn't hear any patent cases. But in recent years Posner has sought to learn about patent law by volunteering to serve temporarily on other courts that do hear patent cases. That's how he came to be the trial judge in the battle between Apple and Motorola. And it also allowed him to spend time on the Federal Circuit bench.

Posner's first-hand experiences with the patent system helped to turn him into a harsh critic. "Our patent law is a failed system," he declared on Wednesday. He cited a laundry list of problems, including an under-staffed and poorly managed patent office and expert witnesses whose credibility is undermined by fees paid by litigants.

Striking closer to home, Posner accused the Federal Circuit—the court Michel headed until 2010 and a sister court to Posner's own Seventh Circuit—of being too favorable to patent holders. A specialized court like the Federal Circuit, he said, tends to take a "promotional view of the activity it regulates." That "promotional view" has caused the court to become "too welcoming to patents," watering down the standards for patent validity.

Michel disputed this point, insisting the court he served on for more than two decades has been "quite balanced." Readers can judge for themselves by reading our recent 30th anniversary retrospective on the Federal Circuit.

Michel acknowledged the current patent system has "friction points" and "shortcomings." But he insisted that "to say that it's broken or failed hugely overstates the real situation."

"The best patent system on the planet"

Michel argued the focus on the costs of litigation is misguided because most patent disputes lead to a license agreement being signed before they reach the courthouse. This system of licensing, he argued, is "working quite well on the whole." Michel contends the United States has "the best patent system on the planet," and that it is "widely imitated" around the world.

In Michel's view one of the big problems with the patent system is actually that people criticize it too much. No, really: "The patent system has been so excessively criticized with overstated claims by so many commentators that it has been substantially weakened in the last five to seven years," Michel said. (We at Ars had no idea our words were so powerful!)

Michel downplayed concerns that 20 years of patent protection is excessive for fast-moving industries. "I don't see the harm of patents being 20 years even in technologies that may have a very fast situation as in the software industry. Patents that have no commercial value aren't enforced. If they become passe, they're still there, but they don't cause much problems that I can see."

"We need a stronger patent system, not a weaker patent system," Michel argued. The incentives currently provided by the patent system "are not adequate now and we need sharply increased incentives."

These calls for even more patent protection "frightened me a little bit," Posner countered. "For most inventive activity, the incentives are quite adequate without patent protection. So when you throw in protections, all you do is increase the costs of the competitive process by entangling firms in litigation."

Judge Posner acknowledged that patents are needed in some industries, notably pharmaceuticals, to provide incentives for investment. But he argued that "in most industries there's very little need for patent protection." He argued this is particularly true in software, an industry where product development costs are low and first-mover advantages can be significant. "I don't want to say that [patents are] never appropriate for software," Posner said, but "I think generally it's just a clog on competition," enriching trolls, trial lawyers, and expert witnesses, but not the general public.

Posner pointed to several other reasons that patents were a poor fit for software. He argued that the 20-year term of patents was way too long for the fast-paced software industry. He noted that software innovation tends to be piecemeal. "A software program or tablet might have thousands of components, resulting in huge patent thickets," he said. This will create "rich opportunities to try to hamstring competitors by suing them."

Judge Michel didn't directly address Posner's point about the dangers of patent thickets. And indeed, if our May interview with Michel was any indication, he may not even understand how patent thickets create problems for IT firms. "If software is less dependent on patents, fine then. Let software use patents less as they choose," he told us in May. Of course, the whole problem is that software firms can't opt out of the patent system; choosing not to seek patents doesn't give a software firm any protection against lawsuits from others.

Channel Ars Technica