Are patent fees effective at weeding out low-quality patents?

Gaétan de Rassenfosse, Adam B. Jaffe

Journal of Economics & Management Strategy (2018), 27(1): 134-148

Research Question

Are patent fees an effective mechanism to deter the filing of low-quality patent applications?

Why Do We Care

Concerns have risen regarding the increasing number of low-quality patents that elevate litigation costs and impede innovation incentives, particularly in the United States and China. Such patents create inefficiencies within the patent system, leading to increased backlogs and diminishing returns from research investments. Understanding whether patent fees can act as a deterrent to low-quality applications is essential for improving patent quality and the overall effectiveness of the patent system.

What Did We Do

We utilized a quasi-natural experiment involving the Patent Law Amendment Act of 1982, which increased patenting fees in the United States, to assess its impact on patent quality. We analyzed a sample of 222,886 utility patents filed from January 1, 1981, to June 30, 1984, focusing on first filings by domestic applicants and second filings by foreign applicants. Employing a difference-in-differences econometric model, we sought to capture the change in the distribution of low-quality patents after the fee increase.

What Did We Find

Our findings indicate that the fee increase led to a 10% reduction in the proportion of low-quality patents, specifically those in the lowest quality decile. Furthermore, we find that 15% of patents in the lowest value decile were filtered out. Notably, the effect was particularly significant among firms with larger patent portfolios (over 20 patents), highlighting that fee structures can effectively screen out low-quality patent applications.

Primary Data

  • 223,329 utility patents filed between January 1, 1981, and June 30, 1984, from the PATSTAT database, with a final analytical sample of 222,886 patents after exclusions

Assumptions

  • The parallel-trend assumption holds, meaning that the quality of patents by local applicants follows the same pre-treatment trend as those of foreign applicants
  • The increase in fees primarily affects low-quality patents rather than merely low-value patents
  • The response to the fee change varies by the size of the patent portfolio held by applicants