One of the critical aspects of research on patents is evaluation of the economic impact of policy changes. A number of studies have attempted this difficult task by analyzing individual policy changes as “natural economic experiments.” In this important paper, Josh Lerner examines not one change but the effects of over 100 patent policy changes in the world’s sixty largest nations over 150 years.

Drawing on a range of historical sources, Lerner identified a large number of policy changes of various sorts. He divided the changes into ones that “strengthened” patent protection—changes that were favorable to patent holders, either by making it easier to obtain patents or by providing better or longer enforcement of the patent right—and ones that had a negative or ambiguous effect on patent holders. He classified the following policy changes as pro-patent: 1) increases in the subject matter covered by patents, including the initiation of patent coverage of any sort; 2) extensions in the length of the patent term; 3) reductions in patent fees; and 4) elimination of limitations on patent grants, including elimination of requirements that patents must be “worked” (put to commercial use) to avoid revocation or compulsory licensing. The result of this research effort was a list of 177 distinct policy changes. Although these changes differ greatly from one another, by grouping them together, Lerner is able to draw general inferences about pro–patent holder policy shifts.

Lerner points out that he would have liked to measure the effects of these changes on innovation rates. Lacking the necessary data, he explored their effect on three rates of patenting: 1) patenting by domestic entities in the country making the policy change; 2) patenting by foreign entities; and 3) patenting by domestic entities in Great Britain. Great Britain serves as a reference point because the identities of foreign applicants can be identified, and because its patent system was widely used since Britain was an important market. Lerner assembled data on these patenting rates for an “event window” from five years before a given policy change to five years after. He then adjusted the rates to correct for overall global trends in patenting.

The resulting picture shows a clear pattern (see Figure). Pro-patent holder changes sharply increased patenting by foreign entities. But patenting by domestic entities decreased, although to a lesser degree, both in country and in Great Britain. All of these differences are statistically significant. These results imply that although pro–patent holder changes may have helped international trade by allowing foreign firms to protect their products, they appear to have had a negative effect on domestic innovation.