Exploring the Patent Explosion
Bronwyn H. Hall (UC Berkeley and NBER)
Link to FULL TEXT of article
Summary
The number of patents granted each year in the United States has exploded, more than doubling since the mid-1980s. In this short paper, Bronwyn Hall reports on three related empirical explorations of this phenomenon. In her investigation, she uses patent data and firm data from Compustat that have been matched by the National Bureau of Economic Research (NBER).
Did a “break” occur and, if so, when?
Hall first attempts to determine whether the average growth rate in successful patent applications changed discretely in a particular year. She finds clear evidence of a discrete change in this growth rate around 1984. This result is consistent with the idea that the creation of a unified court of appeals for patents (the Court of Appeals for the Federal Circuit) in 1982 brought about a “pro-patent” regime.
Hall also looks for a break in trend by the nationality of the inventor, and by the technology class of the patents. The 1984 break occurred almost entirely among U.S. inventors. Asian inventors also showed signs of an increase in their rate of patenting, but a few years earlier, around 1981. Europe and other regions showed no significant evidence of any break. The break (among inventors of all nationalities) also occurred across several technology classes, but it did not prove significant in chemicals and pharmaceuticals, two industries that have consistently reported in surveys that patents are effective.
What accounts for the growth in patenting?
Next, Hall analyzes the proliferation of patents by U.S. inventors to determine the relative contributions of various technology classes and industries. She finds that all major technology classes contributed to growth during the late 1980s. But when the growth is broken out by industry, patenting in electronics, instruments, and computers accounts for the entire increase. Hall concludes that firms in these industries are driving the patent explosion, and that these firms take out patents across the spectrum of technology classes.
Has the change affected the value of patents to firms?
The final analysis concerns the influence of patents on the market value of the firm, above and beyond the contribution of R&D. Here, Hall is particularly interested whether patents increase the market value of entrant firms—firms that recently went public. This result would indicate that patents facilitate financing for such firms. Hall compares the contribution of patents to market value for entrant and incumbent firms, before and after 1984.
As it turns out, a high ratio of patent stock to R&D stock does little to enhance the market value of incumbent firms. In some industries it may even decrease their market value, perhaps because a high rate of patenting may indicate that the firm is prone to litigation.
Entrants in general also seem to gain little from a higher ratio of patents to R&D. But after 1984, for entrants in electrical and “other” industries, there is a significant positive correlation between firm market value and the ratio of patents to R&D.
Based on this finding, Hall suggests that patents may have become a signal to investors of a firm’s viability. In other words, patents make financing easier. As evidence from earlier interviews with firms and venture capitalists shows (Hall and Ziedonis, 2001), patents have indeed become important to obtaining funding in the semiconductor industry.
Still, Hall’s analysis does not indicate the underlying cause of this change. It could be that changes in patent law permitted entrant firms in these industries to protect their innovations from imitation better after 1984. On the other hand, it could be that after 1984, with increased cross-licensing behavior in these industries, investors wanted to make sure that entrant firms had taken out sufficient “defensive” patents to retain significant profits after cross-licensing.
Reference
Hall, Bronwyn H. and Ziedonis, Rosemarie H. “The Determinants of Patenting in the U. S. Semiconductor Industry, 1980-1994.” Rand Journal of Economics, 2001, 32, pp. 101-28.
Notes
