This month, the PTO issued for the very first time a five-item research agenda to be overseen by the new Office of Chief Economist.  This deserves applause, both for innovating and for taking a major step in ensuring that the patent system promotes innovation and economic growth — not just patents.

For the last five years, Congress has struggled with major patent reform legislation to no avail, despite major studies by the National Academies and the Federal Trade Commission, both of which recommended major changes.

Patent reform is hard. The last round, which resulted in the appealingly entitled American Inventors Protection Act of 1999, took most of a decade to pass what ended up as a much-diluted bill.

Since then, the patent system has grown more complex and contentious. As shown in the 24 days of FTC hearings in 2002, there are growing divisions among industries as well as different value chain perspectives on how the system is (or isn’t) working. There is no shortage of posturing, misinformation, and xenophobia when it comes to patent reform. No wonder Congress finds it so hard to act.

Fortunately, the Supreme Court has stepped in after a couple of decades of bandoning patent law to the patent-specialized Court of Appeals for the Federal Circuit. It answered one big problem that the reform legislation didn’t dare address: the low standard of patentability favored by the Federal Circuit (KSR v. Teleflex, 2007).

The Court also put an end to the Federal Circuit’s automatic injunction rule that gave trolls great leverage over producing companies (eBay v MercExchange, 2006)—another area too hot for Congress to handle. But it famously wimped out when it came to defining the limits of the patents, deciding not to decide one case (Labcorp v. Metabolite, 2006) and then deciding another (Bilski v. Kappos, 2010) on a we-know-when-we-see-it basis.

The Supreme Court cannot accept all of the cases where interpretation of the law is at stake. Many issues must be addressed by statute, and only Congress can do that. And unlike other agencies, the PTO has very little power to interpret what Congress legislates.

Role of the PTO

The PTO is charged with examining patents, a very technical function. However, the PTO is also charged by statute with advising the federal government on intellectual property policy. How does it decide what advice to give? Well, traditionally it has asked its “customers” and their attorneys – the same people that Congress turned to year after year for advice on patent policy. After all, nobody else really knew anything about patents, right?, not even the Supreme Court.

In the mid-1990s, the PTO decided that its mission was “to help customers get patents” and emblazoned it at its Crystal City headquarters. Surely, if PTO was helping customers get patents, the customers could help provide policy advice in return. At least, their lawyers could since they were the experts.

Fortunately, with the economy front and center in everyone’s mind, the PTO has found a more logical source of advice: economists. Logical to some anyway. When the Federal Trade Commission had the temerity to recommendthat patent policy would benefit from economic learning, the recommendation was categorically rejected by American Bar Association and the American Intellectual Property Law Association. Yet the European Patent Office has had a chief economist beginning in 2004, and the World Intellectual Property Organization hired one in 2009.

In a speech last April at the annual Innovation Policy and the Economy conference of the National Bureau for Economic Research, Under Secretary for Intellectual Property David Kappos announced the appointment of a chief economist: Stuart Graham, a professor at Georgia Tech, with an outstanding record of empirical research. Kappos also outlined three areas as exemplary of the PTO’s research agenda. The five-item agenda released this week is somewhat different.

What’s missing? In April, Under Secretary Kappos talked about the problem that diversity of innovation poses for the patent system:

To coin a phrase, there are as many patent systems as there are technologies. This vast differentiation suggests that, while there is an economic reason to preserve a unitary system in conception, a unitary system does not exist in practice. The implications for Patent Office policy and behavior deserve further study.

And this, of course, is why patent reform is so hard and getting harder. Sure, from an administrative or legal standpoint, it is easier to have a one-size-fits-all system, but the people who actually do the innovation would naturally prefer a system adapted to the realities of their technology and their industry.

The system is working especially poorly for IT because it produces far too many low-value and low-quality patents that end up in the hands of trolls. Favoring the pharmaceutical sector and patent lawyers over the IT sector and software developers is a backhanded industrial policy that inevitably leads to less innovation and economic growth. And, it is the principal cause of the present impasse over patent reform.

PATENT QUALITY

Under Secretary Kappos also talked about quality as a critical study area. Almost everybody agrees that this is front and center the biggest problem facing the PTO, along with delay and the tradeoff between time and quality. He noted:

The term “quality” may mean one thing to individual inventors, another to patent applicants, another to firms, another to patent owners, another to lawyers, another to courts, and still another to PTO employees.

If we are looking to patents for economic effects, I would rather hear what economists think. A recent study by former European Patent Office chief economist Bruno van Pottelsberghe concludes that the quality of USPTO-issued patents lags far behind the quality of EPO patents. This kind of discrepancy bodes ill for future workload sharing among patent offices, which is often touted as a solution to the growing backlog of 1.2 million patent applications.

From a lawyer’s perspective, quality is a matter of whether the patent has been upheld by the last tribunal to look at it. Ultimately, patent quality has to be evaluated at a systemic level. It is very costly to assess quality on a patent-by-patent basis, especially with 170,000 patents issuing every year.

WHAT DO YOU THINK?

While federal bureaucracies are easy and popular targets, the real problem may be in the statute itself. Today, the rule is that a patent must be granted unless it is obvious to the “person having ordinary skill in the art.” In a competitive globalized economy, ordinary skill is a pretty low standard. Then again, the statute was drafted by lawyers who wanted to be able to satisfy their clients.

So it is disappointing that quality is off the agenda. But take a look at the now official agenda and see what you think.

Tagged with →  
Email this to someoneShare on FacebookTweet about this on TwitterShare on Google+Pin on PinterestShare on LinkedInShare on Reddit