This study is the first to investigate the applicability and usefulness of a recently introduced patent-level measure of innovation, which relies on stock market reactions to patent grants, in a large-scale cross-country setting. Based on patent data for firms across 27 countries from 1973 to 2013, our results indicate that the patent-level innovation measure is only moderately applicable in a cross-country study: we detect patent-related stock price movements only in 10 countries. Furthermore, although patent value estimates for US, German, and Belgian sam- ples are positively correlated with traditional citation-based measures of quality, we cannot confirm the measure’s general usefulness across the 10 countries. On a methodological side, we find that inaccurate isolation of stock market reactions can greatly distort between-country differences in estimated value of innovation. Therefore, we stress careful selection of the stock return window used for construction of this type of measure. Finally, we conclude that bet- ter control for heterogeneity in the type of patent grant can improve both applicability and usefulness of the measure.

, , , , , ,
Tham, W.W.
hdl.handle.net/2105/33720
Econometrie
Erasmus School of Economics

Griep, R.A. (2016, May 19). A Patent-Level Measure of Innovation Based on Stock Market Reactions. Econometrie. Retrieved from http://hdl.handle.net/2105/33720