Please rotate your device to landscape mode to view the charts.

The Use and Misuse of Patent Data: Issues for Finance and Beyond

Journal: Review of Financial Studies

Date: 20220601

Author: Lerner, Josh; Seru, Amit

Abstract:
Patents and citations are powerful tools increasingly used in financial economics (and management research more broadly) to understand innovation. Biases may result, however, from the interactions between the truncation of patents and citations and the changing composition of inventors. When aggregated at the firm level, these patent and citation biases can survive popular adjustment methods and are correlated with firm characteristics. These issues can lead to problematic inferences. We provide an actionable checklist to avoid biased inferences and also suggest machine learning as a potential new way to address these problems.

Link: Google Scholar

Background and Context

Research Focus

This study examines how patent and citation data, increasingly used in financial research, can lead to biased results due to truncation issues and changing composition of inventors.

Scope

Analysis covers patent data from 1976-2006 in the NBER database, comparing it with extended data through 2012 to identify systematic biases in patent and citation measurements.

Methodology

The study analyzes patent and citation biases at firm level using machine learning approaches and traditional adjustment methods to assess measurement errors.

Rising Importance of Patent Citations in Finance Research

  • Shows dramatic increase in use of patent citations in top finance journals over time
  • Share of papers using patent data grew from 0.1% in 1990s to 2.6% in recent years
  • Demonstrates growing importance of patent data in financial research

Growing Patent Application-Grant Gap Across Technology Classes

  • Illustrates uneven growth in patent applications across different technology sectors
  • Computer/Communications sector shows dramatically higher growth compared to other sectors
  • Highlights potential bias in patent data analysis across different industries

Regional Disparities in Patent Activity

  • Shows significant regional differences in patent activity growth
  • California/Massachusetts experienced 250% growth while Delaware saw only 5% growth
  • Demonstrates importance of controlling for regional effects in patent analysis

Performance of Machine Learning vs Traditional Adjustments

  • Compares effectiveness of different adjustment methods for patent data
  • Machine learning approaches show significantly better performance (R² = 0.81) than traditional methods
  • Demonstrates potential of ML in addressing patent data biases

Firm Characteristics Impact on Patent Bias

  • Shows how different firm characteristics correlate with patent bias
  • Leverage and firm size have the strongest relationship with patent bias
  • Highlights importance of controlling for firm characteristics in patent analysis

Contribution and Implications

  • Identifies systematic biases in patent data that can affect financial research conclusions
  • Provides a practical checklist for researchers to avoid biased inferences in patent analysis
  • Demonstrates the potential of machine learning approaches in improving patent data adjustments

Data Sources

  • Citation Trend Chart: Based on Google Scholar compilation discussed in introduction
  • Patent Gap Chart: Based on Figure 2 technology class comparisons
  • Regional Disparities Chart: Based on Figure 3 state-level analysis
  • ML Performance Chart: Based on Table 4 model comparison results
  • Firm Characteristics Chart: Based on Table 1 regression coefficients