This paper applies Sutton's [1998]‘bounds approach’ to the chemical industry. The approach predicts that (i) the lower bound of the ratio of the 1-firm concentration index to product concentration is bounded away from zero at high levels of product concentration only in high R&D markets and (ii) the lower bound on market concentration is higher and increasing in product concentration in markets with higher R&D intensities. We test these two hypotheses and find that the data strongly support them. A novel feature of our analysis is the use of plant level data in defining products and the classification of products into markets based on end-use.