Felder, Johannes Licht, Georg Nerlinger, Eric A. Stahl, Harald
Year of Publication:
ZEW Discussion Papers 95-21
R&D expenditures of firms varies vastly between and within industries. In recent years a lot of theoretical and empirical studies attempted to explain the distribution of R&D expenditures. Four main factors repeatedly appeared in this literature: Firm size, market power, appropriability and technological opportunity. The present paper tests the empirical content of the hypotheses raised in this literature. It employs the data of the first wave of the Mannheim Innovation Panel conducted in 1993. Our study divers from the literature in several aspects: (1) The data covers firms from all size classes. The questionaire was designed to minimize the undercounting problem of R&D in small firms. (2) Besides a traditional definition of R&D we also used the concept of total innovation expenditures to capture the importance of non R& D innovation activities which are especially important for small firms. (3) We give reliable estimates on the occurrence of R&D and innovation activities in small and medium sized enterprises in Germany. (4) Our data base includes R&D performers as well as non-R&D performers. (5) We distinguish between the decision to perform R&D or not and the decision on the amount invested in R&D. We show that there are several differences with respect to both stages. As our survey collects information on several R&D enhancing characteristics of firm and market we are able to employ a more extended set of variables explaining R&D performance than most previous studies. The main results can be summarized as follows: (1) Once small firms have decided to invest in innovation activities, the amount they invest as a percentage of sales is larger than the innovation intensity of big firms. On the other hand, the probability that a firm is engaged in R&D increases strongly with firm size. The large and small firms differential in intensity can even be more pronounced if we use total innovation expenditures instead of the narrowly defined R&D expenditures only. (2) Evidence of a positive relationship between seller concentration and innovation input is rather weak. (3) Stronger appropriability conditions and higher technological opportunities enhance firms spending on investment in innovation activities and/or in R&D. (4) Certain other firm characteristics (exporting firm, financial constraints) play a role in determining innovation expenditures and R&D expenditures.