Chan, Lakonishok and Sougiannis (2001) suggest an R&D intensive investment strategy, which results in excess return with high R&D intensive portfolios. We show how this investment strategy can be improved by taking into account the different industries and the competitive strategy. For this purpose we develop a simple proxy for Porter’s product differentiation and cost leadership strategies from DuPont identity components. Our strategy proxy seems to provide a good screen for stocks with negative excess returns. However, the improved excess returns are partly increased by loading unexplained risk.

We partly replicate Chan et al. (2001) study for the period 1975–2011 with comparable results. We find the highest R&D intensity portfolios have excess returns after controlling for common risk factors. These stocks seem to be past losing stocks. In contrast to the reference study, we found slightly higher excess returns. The difference is explained by the different time period and lack of CRSP delisting return data in our study. We find support for our first (H1) hypothesis.

H1: R&D intensity (measured by R&D expenses to market value of equity) has explanatory power over stock returns.

To develop the investment strategy we explore the excess returns separately in ten different industries classified by SIC codes. After controlling for common risk factors, we show that R&D intensive portfolios’ excess returns are the phenomenon only in certain industries. These industries are hitec, healthcare and manufacturing. However, results for manufacturing excess returns were only slightly positive and lasted only for a year. We find support for our second (H2) hypothesis.

H2: R&D intensity (measured by R&D expenses to market value of equity) has explanatory power over stock returns only among the high R&D intensity industries.