A policy insight into the R&D–patent relationship

Gaétan de Rassenfosse, Bruno van Pottelsberghe de la Potterie

Research Policy (2009), 38(5): 779-792

Research Question

How do variations in the propensity to patent and research productivity impact the R&D–patent relationship across countries?

Why Do We Care

Understanding the relationship between R&D and patents is crucial for policy-making, as it influences how countries strategize their innovation and intellectual property strategies. Traditional views often overemphasize the propensity to patent, overlooking how research productivity can equally influence patent counts. Accurately dissecting these factors is vital to designing effective education, IP, and S&T policies that foster innovation.

What Did We Do

We employed an empirical model to analyze data from 34 countries in 2003, using national priority filings as the dependent variable to gauge the relationship between R&D and patents. The model differentiated between the productivity of researchers and the propensity to patent, testing various policy impacts, including education and IP policy design.

What Did We Find

We find that patent counts per researcher reflect both productivity and propensity factors, with the former contributing significantly to the number of international filings (triadic patents show up to a 1.75 increase in filings per 10% increase in researchers). We also established that national priority filings are more influenced by propensity factors, particularly administrative fees, which negatively correlate with patent counts (a 10% fee increase leads to approximately a 4% decrease in filings). Lastly, strong IP protection correlates positively with filing numbers, emphasizing the interplay between patent systems and innovation.

Primary Data

  • 34 countries observed for 2003 using PatStat and OECD MSTI databases
  • Data include various patent filing types such as national priority filings and triadic patents
  • Human Capital Index and patent policy metrics obtained from multiple databases

Assumptions

  • Variations in patent counts are due to differences in productivity and propensity to patent across countries
  • The model assumes a constant propensity to patent and constant research productivity across countries
  • The design of IP and S&T policies significantly impacts the propensity to patent and research productivity