NBER Working Paper #8807
Full text (subscription required)

The economic function of the patent system depends on the “quality” of the patents that are granted. Poor-quality patents may ignore prior art, they may be obvious to the industry involved, their claims may be excessively broad, and so forth. The economic effect of the patent system generally depends on the qualitative standards set by the law and the courts, and on how effectively the patent office implements those standards.

Patent examiners inevitably make mistakes; for instance, they may be unaware of prior art, especially in fields such as software and business processes, where patent protection is relatively recent. “Quality control” is therefore a critical aspect of the patent system. The system will operate better if the institutions involved can bring errors to light and fix them quickly and cheaply. Concerns about poor-quality patents in software and business processes in the U.S. have given rise to reform proposals that call for an administrative opposition system—a way for third parties to challenge patent claims—to do just this (Merges, 1999). Reform proponents argue that an opposition procedure like the one used in Europe would reduce costly litigation and eliminate economic uncertainty about patent validity.

This paper conducts a baseline empirical comparison between the European Patent Office (EPO) opposition system and the US Patent and Trademark Office (USPTO) re-examination procedure. The EPO grants patents that are enforceable in the national courts of multiple European states. For nine months after the EPO issues a patent, the validity and claims of that patent can be challenged in an administrative proceeding. The opposition procedure is an adversarial proceeding, one that involves, among others, the patentee, the original patent examiner, and, most important, the challenger.

In the U.S., patents can also be challenged, but the role of the challenger is very limited, and the proceedings involve primarily the patent examiner and the patentee. These are not adversarial proceedings at all; the paper reports that close to 50 percent of patent re-examinations are initiated by the patent holders themselves—to reaffirm the patent’s validity in light of newly reported prior art, for instance.

These very different institutions generate very different outcomes. The study finds that 8.3 percent of EPO patents are challenged in an opposition proceeding, while only 0.2 percent of USPTO patents are re-examined. Patenting standards are similar under both systems, which means that forty times as many EPO patents are subject to quality control proceedings.

The proceedings also lead to very different results under each system. Over 35 percent of challenged EPO patents are revoked in EPO opposition proceedings; many of the remainder have claims reduced. Only 9 percent of U.S. re-examinations result in cancelled patents, though in about half of these cases, the claims were amended. The authors also analyze the factors that affect the likelihood of opposition/re-examination; this includes some analysis using a matched set of patents issued by both the USPTO and the EPO.

The EPO system does seem to provide a more rigorous means of quality control, short of litigation. However, the authors also find that while opposition proceedings are less costly than litigation, they may last just as long. The authors promise future research that will look at the relationship between opposition and litigation more thoroughly, as well as a discussion of what these different systems imply about the ultimate quality of patents.

Merges, Robert P. 1999. “As Many as Six Impossible Patents Before Breakfast: Property Rights for Business Concepts and Patent System Reform,” Berkeley High Technology Law Journal, 14:577-615.