The recent decade has shown a surge of firms globalising their innovation activities. This means that these firms have enlarged the number of countries where they perform R&D projects. Hence, for more and more firms R&D is no longer a purely domestic activity. A major motive underlying the decision to shift corporate R&D activities abroad is that firms seek to increase their chances to participate in international knowledge sharing. Absorbing knowledge from abroad is aimed at enhancing firms’ innovativeness and consequently their competitiveness. This paper aims at analysing whether a firm’s innovation performance is enhanced by globalised R&D. In order to evaluate the benefits of a strategy that disperses R&D across foreign countries, we compare firms which have globalised their R&D activities with firms performing R&D only in their home country and with firms that are engaged in no R&D activities at all. One main novelty of our research is that we furthermore distinguish between different degrees of R&D internationalisation, that is, the number of foreign countries in which a firm has R&D laboratories. This allows us to shed light on the two competing hypotheses by Malecki (1980) and Von Hippel (1988) whether stronger dispersed international R&D activities hamper or stimulate firms’ innovation. Compared to previous research that is based on patents, we use two different well-established innovation output indicators to shed light on the effects of domestic and foreign R&D: first, the introduction of new products and second their market success which is measured as the sales growth rate due to these innovations. We additionally examine whether the effects of international R&D vary with different degrees of product novelty (firm and market novelties). The study is based on a large data set of about 2100 firms surveyed in the Mannheim Innovation Panel (MIP). The MIP represents the German part of the European-wide harmonised Community Innovation Surveys (CIS). Based on econometric analyses, we confirm that firms with both domestic R&D and foreign R&D activities are more likely to launch new products than firms with home-based R&D only or non-R&D performers. This can be observed for both kinds of product innovations: market novelties and firm novelties. Regarding the innovation success, however, the results are not as clear-cut. Given the introduction of a firm novelty, firms with international R&D centres are also more successful than non-R&D performers and tend to be more successful than firms conducting solely domestic R&D. On the contrary, the findings reveal that the location of R&D does not matter for the success of market novelties once the market novelty has been launched to the market. By comparing different degrees of R&D internationalisation, the study points out that firms with medium decentralised foreign R&D activities (in 2-3 countries) have a significantly higher likelihood to develop both market and firm novelties compared to firms with centralised foreign R&D in only one foreign subsidiary. A more decentralised foreign R&D organization does not exert an additional stimulating effect on the propensity to develop market novelties. On the other hand, it further increases the likelihood to develop firm novelties. The results further elucidates that firms with medium decentralised R&D abroad achieve a higher sales growth with new products than firms which conduct R&D at home only or than firms with highly decentralised foreign R&D activities. Again, the higher sales growth rates are mainly driven by firm novelties, not by market novelties. That is, given the introduction of a market novelty, the degree of decentralisation of international R&D activities does not play a role for innovation success. Finally, another salient finding that originates from this analysis is that German firms that have expanded their R&D only to one foreign country do not outperform firms with domestic R&D.
Peters, Bettina and Anja Schmiele (2010), The Influence of International Dispersed vs. Home-based R&D on Innovation Performance, ZEW Discussion Paper No. 10-102, Mannheim.