FINNOV

Abstract: Given recent attention to ‘gazelle’ firms (small high growth innovative firms), the paper asks whether a more fine-grained analysis of the relationship between R&D and growth can help to understand the factors that drive firm performance. We highlight that whilst SME firms can be associated with high R&D intensity and fast growth, R&D scale is still important and so policy targeting a narrow view of performance related to ‘gazelle type’ firm characteristics may misunderstand the R&D-growth relationship. Quantile regression techniques are used to test whether the behavior of the top growing firms (the 90th percentile) is different from that of firms with lower growth rates. Data on the US pharmaceutical industry from 1950-2007 suggest that what matters is not high growth in itself, but rather a mix of characteristics that includes high growth. These include R&D intensity, R&D scale, persistence of patenting, and venture capital funding. We also contextualize the study of high growth firms within the competitive environment in which they operate and find that the ‘high growth’ characteristic is most important, in influencing the R&D-growth relationship, during periods when competition is fiercest (proxied by low concentration and high market share instability). In such periods, being a growth laggard renders R&D only a cost, and hence can even influence growth negatively. In less competitive periods (high concentration and high stability of market shares), the R&D-growth relationship does not differ much between high growth and low growth firms, making it appear that that ‘variety does not matter’.

Key words: R&D, growth, venture capital, quantile regression, pharmaceutical industry.

JEL Classifications: L1 Market Structure, Firm Strategy, and Market Performance, O3 Technological Change.

To request a copy of this deliverable, please contact the FINNOV team.