From Patents to Profit: The Impact of Genuine and Counterfeit Innovations on Firm Performance in China
Description
This is the source data and code of the paper : "From Patents to Profit: The Impact of Genuine and Counterfeit Innovations on Firm Performance in China". This paper examines the relationship between patent acquisition and corporate profitability in China’s evolving innovation ecosystem, using data from 2008 to 2022 and applying Difference-in-Differences methodology. As firms increasingly acquire counterfeit patents to exploit government tax incentives, a misleading surge in patent filings often hides underlying inefficiencies, ultimately hindering productivity. Leveraging a quasi-natural experiment triggered by a 2016 government policy to promote genuine innovation, we categorize firms based on their patent application behaviors: those with high genuine filings versus those relying on external acquisitions. Our findings show that firms with substantial genuine patent portfolios experience significant profitability improvements, while those dependent on counterfeit patents suffer negative financial outcomes. This paper highlights the importance of distinguishing between authentic and counterfeit innovations when designing incentive structures. Promoting genuine technological advancement is essential for sustainable economic growth. Our findings advocate for strategic government interventions that encourage authentic innovation while addressing the risks of opportunistic patent acquisition. df.csv is a financial dataset of listed companies, which contains annual financial reports over the years, as well as stock market returns. patentstype.csv is a patent dataset, which contains the type and number of patent applications filed by each listed company over the years. totforreg.csv contains all the enterprise financial data and R&D data, the original data, after df and patentstype two data sets and, the original data has not been data cleaning jg.csv dataset is the dataset after data cleaning, mainly used for mapping, and robust analysis. pytempforpaper.py is the python source code for this study.
Files
Steps to reproduce
see readme file