پژوهش های حمایتی و توسعه و محیط زیست محلی

Multinational enterprises (MNEs) increasingly seek to optimise their global innovative capabilities by incorporating subsidiary-specific advantages in different countries. But how important are the different location advantages for subsidiary investments in research and development? This paper adds to our understanding of the effects of commonly cited location advantages by analysing the degree to which they actually influence the incidence and level of subsidiary R&D. We developed hypotheses from the existing literature on location advantages, multinational R&D, and innovation incentives, and tested them in relation to an empirical data set containing over 2000 responses from subsidiary managers in seven countries in Europe. Four aspects of the local business environment were investigated: competitive conditions, supply conditions, scientific institutions, and government support. We found that only the presence of scientific institutions has a consistent, positive effect on the incidence and level of subsidiary R&D. Government support has a positive effect on the incidence of subsidiary R&D, but not its level. However, highly competitive environments have a negative effect, at least in small countries.

Multinational enterprises (MNEs) are no longer content, as in earlier decades, to develop new products and processes at home and transfer these innovations to foreign subsidiaries to adapt to local market needs. They increasingly seek to optimise their global innovative capabilities by incorporating subsidiary-specific advantages in different countries, sometimes engaging in major research at the subsidiary level. The literature on multinational R&D discusses a range of putatively valuable location advantages (e.g. Boutellier, Gassmann and Von Zedtwitz, 1999, Conference Board Europe, 1995, Florida and Kenney, 1994, Frost, 2001, Håkanson, 1992, Kuemmerle, 1999, OECD, 1998, Patel and Vega, 1999, Pearce, 1997 and Von Zedwitz and Gassmann, 2002). But most of this work is based on either patent data or sources from MNE headquarters, not the responses of subsidiary managers. There is also an extensive general literature on the effects of different aspects of the local business environment on R&D (e.g. Porter, 1990 and Von Hippel, 1988)—but none, to our knowledge, analyse these effects from the perspective of the subsidiary manager. This paper adds to the literature on international R&D by analysing which of the commonly cited location advantages affect the incidence and level of subsidiary R&D, based on the questionnaire responses of subsidiary managers. To this end, we generated hypotheses from the literature on location advantages, multinational R&D, and innovation incentives, and tested them with subsidiary level data. The results have implications both for understanding multinational R&D, and the effects of industrial policy. In its reliance on subsidiary assessments, this paper contributes to the emerging literature that takes a “subsidiary view” of the MNE (e.g. Andersson, Forsgren and Pedersen, 2001, Birkinshaw, 2000 and Birkinshaw and Hood, 1998a). The study employs data of the “Centres of Excellence” database,1 which was constructed specifically to investigate the nature of the MNE as a “multi-centre” firm (e.g. Holm & Pedersen, 2000 and Moore, 2001). The survey on which this paper is based contains responses from subsidiary managers in seven European countries: the UK, Germany, Austria, Sweden, Norway, Finland, and Denmark. Our approach differs from parallel research with this database in its focus of analysis. While prior studies have sought to illuminate the affiliate’s relationship with the parent MNE and with its local and global business partners from the affiliate’s point of view, they paid little attention to the effects of the local environment. An important exception is the recent contribution by Frost, Birkinshaw, and Ensign (2002), who investigate the factors (including the local environment) governing the emergence of centers of excellence in MNEs in Canada. The paper explores two interrelated research questions. Which of four commonly cited location advantages—market competitiveness, supply conditions, scientific institutions, and government support—are inducing R&D by subsidiaries? Moreover, which of these factors have a significant impact on the scale of the R&D investment? Three interesting findings emerge from the empirical analysis. First, only the presence of scientific institutions has a consistent and positive effect on both the incidence and level of subsidiary R&D. Second, government support has in some cases a positive effect on the incidence of affiliate R&D—but not its scale. Third, highly competitive market environments have a negative effect on subsidiary R&D, at least in small countries. Taken together, these findings demonstrate that of greatest importance to stimulating subsidiary R&D are institutional conditions, especially proximity to top quality scientific institutions. Thus we suggest that more attention be paid to the assessments of subsidiary managers of the strengths of the local environment in planning R&D investment decisions and strategies at the subsidiary level. Understanding how activation of local-environment potentials can generate dispersed individualised capacities in MNE development is a central role in the value of R&D subsidiaries. The arguments in this paper are structured as follows. We start with a brief review of the relevant literature. The next section develops the hypotheses used to test the effects of different location advantages on subsidiary R&D. We then present the empirical data, and the design of the empirical study. The results of the empirical study are set forth, followed by a discussion of the impact of mode of subsidiary establishment and the size of the host country market on subsidiary R&D. In conclusion, we outline some of the implications of the analysis.

In conclusion, three findings in particular stand out: 1. Porterian effects of market and supply conditions on capability development via R&D could not be confirmed in this study. In fact, we found a negative effect of a strong local environment in small countries: competitive market environments had a significant, negative effect on the incidence of subsidiary R&D, and superior supply conditions had a negative effect on the level of subsidiary R&D. In consequence, small countries cannot like large countries rely on the strength of the endowment of market and supply conditions, but have to rely on other means to attract R&D affiliates. Prior research having focussed on large countries may have overstated these effects. 2. Local scientific institutions have a strong effect on both the incidence and level of subsidiary R&D. This we found highly significant in all variations of our analysis. This result underlines the importance of local institutions in formulating strategies in foreign countries (Oliver, 1997 and Meyer, 2001). 3. Public support for R&D appears effective only for the incidence of subsidiary R&D, in particular with respect to Greenfield projects, and inducing subsidiary R&D in small countries. But it does not encourage affiliates to engage in extensive R&D. The subsidiaries appear to do just as much as necessary to qualify for support, but with limited amounts of resources committed to R&D. Support offered to other types of affiliates appears ineffective. The underlying cause may be twofold. Governments may provide support with the specific aim of attracting Greenfield investors; or MNEs may opportunistically take advantage of available subsidies and set up Greenfield projects with low levels of R&D activities. The policy implication is that governments should consider coupling the amount of support available for R&D to some quantitative measure of R&D activity. As all research, this study has its limitations. For one thing, we did not pose the same set of questions to sources at headquarters and the subsidiary. Thus while our findings reflect the views of subsidiary managers, we cannot make specific inferences as regards differences between the headquarters perspective, and the subsidiary perspective, for these particular respondents. This issue could profitably be addressed in future research. Moreover, many of our empirical measures are based on single items, and the dependent and independent variables are based on responses from the same person, which may lead to common method biases. Future studies may take these issues into account in the questionnaire design stage. In terms of generalization, the analysis is limited to the effects of location advantages on affiliates in Europe. Our results may, to a certain extent, reflect aspects of the local environment which are specifically European, and thus of limited applicability to non-European countries. It should also be emphasised that since the same EU research and development policies apply to all affiliates, it is not possible, in this paper, to make more general inferences about the effectiveness of EU policies. Further research may address the two surprising findings of the study. The failure to establish a link between local market and supply conditions and R&D requires new theoretical as well as empirical studies. The latter may employ more differentiated measures of the business environment based on both survey and archival data. The negative effect of market competitiveness for small countries in particular, suggesting a dichotomy of subsidiary roles in these countries, merits special attention. Additional research is also called for to analyse the effectiveness of public sector research support programs with respect to their uptake by foreign affiliates. Are these affiliates free riding, or are they really contributing substantive research (and possible spillovers) to the local business community? More generally, conclusions (2) and (3) urge empirical researchers of any aspect of entry or affiliate strategy to incorporate institutional variables in their analysis. This research, moreover, calls for a reassessment of the suitability of the Porter (1990) framework for the analysis of industries with strong MNE involvement. Along similar lines, scholars have criticised Porter (1990) for not adequately taking into the account the realities of the modern multinational enterprise. Both Dunning (1993) and Rugman and Verbeke (1993) have argued that it is not enough to analyse the competitive advantage of the MNE solely in terms of its home country diamond. Our findings indicate that more research is needed to determine which is the relevant market for R&D subsidiaries: the immediate host country market, or the larger (regional or even global) market. The earlier-mentioned studies of multinational R&D (e.g. De Meyer, 1992, Florida and Kenney, 1994 and Gerybadze and Reger, 1999), which emphasise the importance of trying out new products with sophisticated lead customers in the local environment (see Section 3.1), also note the wider importance of these lead customers, to the degree that the innovation will be subject to similar customer demands in other markets around the world. Our result of important differences between small and large host countries point to interesting avenues to deepen this line of research. Here, and more generally, much depends on the subsidiary’s role. Many affiliate conducts R&D locally but aim to exploit the fruits of this R&D internationally, such that the relevant competitive market may have to be defined more broadly. Similarly, a “stand-alone” pre-competitive laboratory unit will clearly be influenced differently by the local environment than the fully integrated R&D–production–marketing unit. Thus an important extension of this research, when trying to assess the effects of particular location advantages, would involve categorising R&D subsidiaries according to the roles of the affiliate within the MNE. Managerial implications arise with respect to the role of subsidiary managers. As has been emphasised, this study, in contrast to most other work on the MNE, has been based on the views of subsidiary managers as regards what matters (and does not matter) to subsidiary R&D in the local environment. To the degree that our results differ from the expectations outlined in the hypotheses, it does seem that subsidiary managers possess key insights into the characteristics of the local environment that have not sufficiently been picked up by existing studies. To encourage higher levels of subsidiary R&D, greater efforts might be taken at other levels in the MNE to understand the assessments of local subsidiary managers as regards their placement in, and interactions with, the local environment. Two specific implications for MNE management might be advanced. First, to the degree that multinational enterprises seek to optimise subsidiary R&D, more attention should be paid to the quality of local scientific institutions (and less to market and supply conditions) in making the initial location decision, and in allocation decisions regarding resources for R&D. Second, subsidiary managers, particularly in small countries, should be encouraged to establish and nurture as close contacts as possible with local scientific institutions.