Stratégie des firmes

Location criteria of activities related to innovation

ImageAuthor : Fabrice Hatem

Editor : OECD, october 2009

Location criteria of activities related to innovation: an econometric study on OECD AFA and FATS data bases

As the rest of the economy, innovation-related activities (IRA) [1] have gone through a steady internationalization process in the recent years. This trend has consequently led to a growing awareness of host territories regarding the need to offer a favourable environment to companies willing to develop their innovation-related activities in the most attractive location. This concern is of special importance for OECD countries, which have to rely heavily on these innovation-related activities to make up for the decline of some of their traditional manufacturing industries.

At the request of the OECD secretariat, an econometric study has thus been implemented on the international location criteria of innovation-related activities (IRA). These activities include, on the one hand, high and medium high tech industries, as defined by OECD (OECD, 2007), and, on the other hand, all business R&D activities. OECD’s AFA and FATS data base on activities of MNEs abroad have been used as the major sources of data for the explained variables.

The results of this econometric study show the importance of market size, agglomeration effects, and, to a minor extent, the quality of public governance for the location of international activities in innovation-related activities. The overall degree of the country’s openness to FDI also appears to be a significant location determinant. As expected, the location of foreign-controlled R&D expenditures seems very sensitive to the overall size and/or efficiency of the domestic innovation system, as measured by R&D expenditures as well as by patents.

Another important finding of the study is that there are significant differences in location criteria depending on the nature of the explained variable. The level of foreign- controlled value added is mostly sensitive to access to market and industrial agglomeration effects. Employment (expressed in terms of headcounts) is the only variable for which labour costs play a significant role as a location criteria. Foreign controlled R&D expenditures are very sensitive, as could be expected, to innovation related criteria (such as total domestic R&D expenditures and patents granted to residents).

These results give interesting insights on the possible existence of global location strategies carried out by MNEs in order to internalize the competitive advantages of each of the potential host countries by locating each segment of their product value chain in the places most fitted for this kind of activities. While some general requisites (such as the proximity to market, a favourable business environment, an openness to FDI) will be influential for all kind of activities, the most labour intensive segments will be located, as expected, where labour is cheap, the most R&D intensive where the national innovation system is efficient, while overall value added in monetary terms will be produced where there is already a significant industrial base and where skilled labour is available.

While this global model seems to be fully instrumental in many manufacturing industries, such as electronic or automotive, it seems less relevant in some services industries, especially in telecommunication service, where access to market definitely seems to play a much more prominent role than other criteria such as resource or efficiency seeking strategies.

The relative importance of the various criteria may also vary depending on the industry. For instance, labour cost play a less important role in pharmaceuticals and chemicals (both very capital intensive industries where per capita productivity is very high on average) than in automotive (a quite labour intensive industry employing large staffs of people in manufacturing plants).

These results bear a significant contribution to the existing knowledge on location criteria by industries and business functions. They also confirm the value of the AFA and FATS data bases as a major source of statistical information for the implementation of in-depth analytical studies on issues related to MNEs’ internationalization, location decisions and countries’ attractiveness to foreign direct investment.

In addition to these finding for OECD countries, the analysis of other sources with a larger geographical scope show that, while emerging economies still only play a limited role in most of the innovation-related industries, they are presently attract a large share of new projects, due among others, to the quick growth rate of their market and the availability of large pools of labour force).

Faced to this growing competition from emerging economies, OECD countries should more than ever, consider the enhancement of their countries’ attractiveness for international projects in IRA as a major policy priority. The stimulation of local markets for innovation-related products and services, the increased efficiency of the national innovation system, the improvement of the regulatory and administrative environment, the control of production costs, and the implementation of targeted promotion policies for IRA can be considered as the five major components of this agenda, as they correspond to the major location determinants of international projects.

 [1] The acronym IRA will be used in the rest of this report to refer to « innovation related activities ».

To download this document, please click on the following link : OECD

Laisser un commentaire

Votre adresse e-mail ne sera pas publiée.