Do stronger intellectual property rights incentivize female participation in innovation? We provide new evidence on this question using a unique database of artificial intelligence patents publicly shared by the USPTO. Our identification strategy leverages China’s WTO TRIPs accession, which led to stronger intellectual property rights in 2002. We find a significant rise in the number of female inventors and an increase in the number of patents with females on inventor teams vis-a-vis a control group of countries. We also find that after stronger intellectual property rights, the quality of Chinese artificial intelligence patents with female inventors on the team improved. Results are robust controlling for unobserved heterogeneity at the country, technology class, and over time. Additional robustness tests with synthetic controls, coarsened exact matching, randomized inference and alternative control groups support our benchmark findings. Our results highlight that stronger intellectual property rights can be helpful in improving gender division of labor thereby benefiting society and innovation.
Authors
Shubhangi Agrawal, Sawan Rathi, Chirantan Chatterjee, Matthew J. Higgins
- Acknowledgements & Disclosure
- Funding from Hoover Institution Visiting Fellowship and University of Sussex Business School is acknowledged. Higgins acknowledges support from the Sorenson Center for Discovery and Innovation Studies, David Eccles School of Business, University of Utah. The views expressed herein are those of the authors and do not necessarily reflect the views of the National Bureau of Economic Research.
- DOI
- https://doi.org/10.3386/w32547
- Published in
- United States of America
Table of Contents
- Introduction 3
- Impact of China Joining the WTO 6
- Institutional Changes 7
- Innovation Environment of China 8
- Data 9
- Empirical Specification 11
- Findings 13
- Increasing Female Led Innovation in China After WTO Accession 13
- Baseline Findings 13
- Alternate Control Group 14
- Coarsened Exact Matching (CEM) 15
- Alternate Sample - Extended Period 16
- Non Existent Pre-Trends 16
- Synthetic Control 17
- Randomized Inference 18
- Heterogenities Across AI Technology Sub-Classes 19
- Mechanisms 19
- Share of Domestic Female Inventors 20
- Percentage of Women in STEM PhDs 21
- Firm Characteristics 22
- Impact on Quality of Innovation 23
- Discussion 24
- Appendix 52
