Finding Winners based on Intellectual Property Accumulation

Caroline Leth & Frederikke Sofie Due Olsen

Student thesis: Master thesis


This study examines the link between firms’ ability to innovate and operating performance as well as stock price returns. We propose a new innovation-related return predictor, Innovation Accumulation (IA), that is distinct from existing return predictors, measuring firms’ pace of intellectual property (IP) generation. Specifically, we investigate IA as an anomaly across publicly listed European firms over the years 2010 to 2020. We argue that a firm’s ability to expand its IP portfolio is a close proxy for its ability to innovate, and based on the idea that firms’ innovative activities should translate into operating performance, we propose that the IA measure is positively associated with contemporaneous and subsequent operating performance. Consistent with imperfect rationality theory, we propose that investors are unable to process and price the information contained in IA effectively, owing to the difficulty in evaluating the economic implications of the publication of patents and trademarks, which results in underreaction and short-term mispricing. In addition, we test if the information contained in IA is incremental to that of existing stock market factors. We show that there are large cross-industry variations in IP activity, and we propose that the relation between IA and operating performance as well as stock price returns is stronger in industries that are more focused on generating IP. Furthermore, agreeing with limited attention theory, we believe that small-cap stocks receive less investor attention, and we investigate if delayed stock price reactions to the information contained in IA amongst smallcap stocks result in greater return predictability. We do not find supporting evidence to reasonably establish a positive relation between our Innovation Accumulation measure and operating performance as well as stock price returns, which we attribute to be a consequence of the quality of the dataset. We contend that the lack of a reliable point-in-time database on European firm’s intellectual property is a fundamental problem in the academic field of innovation studies. This considered, we show that financial markets are not efficient in pricing in IP, and stocks issued by firms that are IP-focused, proxied by having a patent portfolio of at least 100 patents and trademarks, are undervalued relative to existing factor model benchmarks.

Educations, (Graduate Programme) Final Thesis
Publication date2021
Number of pages122
SupervisorsThomas Plenborg