4.1 Innovation Measure

(C.-H. Chang and Wu 2021)

  1. Obtain information on patenting activities from the National Bureau of Economic Research (NBER) patent database, which provides detailed information for patents granted by the U.S. Patent and Trademark Office (USPTO) from 1976 to 2006.
  2. Use the patent data provided by Kogan et al. (2017) to update patent information through 2010.
  3. Use patent-based metrics to proxy for firms’ innovative activity, as they measure innovation output that incorporates the effectiveness of firms’ use of innovation input following Chemmanur et al. (2014).
  4. Measure a firm’s patenting activity using two measures:

    1. Number of patents applied for and ultimately granted by the USPTO in a given year, using the application year rather than the grant year to measure a firm’s innovation in a given year.

    2. Number of subsequent citations received by all patents filed in a given year that captures the quality and influence of innovation.

  5. Correct for two types of truncation problem as follows:

    1. End the sample period in 2008 to mitigate the potential truncation issue on the lag between patent application and grant date.

    2. Correct for truncation bias in patent citations using the technology class-year fixed effect approach in Hall et al. (2001, 2005).

  6. Use the natural logarithm form of each measure to mitigate bias introduced by outliers.
  7. In extended analysis, obtain results based on alternative measures of innovation activities, such as R&D expenses, innovation efficiency originality, and generality to provide a complete picture.
  8. In untabulated tests, use raw citations, citations excluding self-citations, adjusted citations based on a weighting index, and the economic value of patents following Kogan et al. (2017).
  9. Reach the same conclusion using these alternative measures of patent quality.

References

Chang, Ching-Hung, and Qingqing Wu. 2021. “Board Networks and Corporate Innovation.” Management Science 67 (6): 3618–54.