Research and Development, Spillovers, Innovation Systems, and the Genesis of Regional Growth in Europe

Abstract

Rodríguez-Pose A. and Crescenzi R. Research and development, spillovers, innovation systems, and the genesis of regional growth in Europe. Regional Studies. Research on the impact of innovation on regional economic performance in Europe has fundamentally followed three approaches: (1) the analysis of the link between investment in research and development (R&D), patents, and economic growth; (2) the study of the existence and efficiency of regional innovation systems; and (3) the examination of the geographical diffusion of regional knowledge spillovers. These complementary approaches have, however, rarely been combined. Important operational and methodological barriers have thwarted any potential cross-fertilization. This paper tries to fill this gap in the literature by combining in one model R&D, spillovers, and innovation systems approaches. A multiple regression analysis is conducted for all regions of the group of 25 European Union countries (EU-25), including measures of R&D investment, proxies for regional innovation systems, and knowledge and socio-economic spillovers. This approach allows the discrimination between the influence of internal factors and external knowledge and institutional flows on regional economic growth. The empirical results highlight how the complex interaction between local and external research, on the one hand, with local and external socio-economic and institutional conditions, on the other, shapes the innovation capacity of every region. They also indicate the importance of proximity for the transmission of economically productive knowledge, as spillovers are affected by strong distance decay effects.

Rodríguez-Pose A. et Crescenzi R. R&D, retombées, systèmes d'innovation, et la genèse de la croissance régionale en Europe. Regional Studies. Les recherches sur l'impact de l'innovation sur les performances économiques régionales en Europe ont essentiellement suivi trois approches: (1) l'analyse du lien entre l'investissement dans la R&D, les brevets et la croissance économique; (2) l'étude de l'existence et de l'efficacité de systèmes régionaux d'innovation; et (3) l'examen de la diffusion géographique des retombées régionales du savoir. Ces approches complémentaires ont cependant été rarement combinées. D'importants obstacles opérationnels et méthodologiques ont contrecarré toute fertilisation croisée potentielle. Dans cet article, nous essayons de combler cette lacune dans la littérature en combinant dans un modèle unique les approches de type R&D, retombées et système d'innovation. Une analyse de régression multiple est menée pour toutes les régions de l'UE à 25, incluant les mesures pour l'investissement en R&D, les mandataires des systèmes d'innovation régionaux, et les retombées socio-économiques et du savoir. Cette approche nous permet de faire la distinction entre l'influence des facteurs internes et des flux externes de savoir et institutionnels sur la croissance économique régionale. Les résultats empiriques soulignent combien les interactions complexes entre la recherche locale et la recherche extérieure, d'un côté, et les conditions socio-économiques et institutionnelles locales et extérieures, de l'autre, façonnent la capacité d'innovation de chaque région. Ces résultats indiquent également l'importance de la proximité dans la transmission du savoir économiquement productif, les retombées étant affectées par les forts effets de dégradation qu'entraîne la distance.

Croissance économique Innovation R&D Savoir Retombées Systèmes d'innovation Régions Union Européenne

Rodríguez-Pose A. und Crescenzi R. F&E, Wissensübertragung, Innovationssysteme und die Entstehung von Regionalwachstum in Europa. Regional Studies. Bei der Erforschung der Auswirkung der regionalen Wirtschaftsleistung in Europa auf die Innovation wurden bisher in erster Linie drei Ansätze verfolgt: (1) die Analyse der Verknüpfung zwischen Investitionen in F&E, Patente und Wirtschaftswachstum, (2) die Erforschung der Existenz und Effizienz von regionalen Innovationssystemen und (3) die Untersuchung der geografischen Diffusion von regionaler Wissensübertragung. Diese Ansätze ergänzen einander, wurden jedoch bisher nur selten miteinander kombiniert. Eine potenzielle gegenseitige Befruchtung wurde durch wichtige operative und methodologische Hindernisse konterkariert. In diesem Beitrag versuchen wir diese Lücke in der Literatur zu schließen, indem wir die Ansätze für F&E, Wissensübertragung und Innovationssysteme in einem Modell kombinieren. Wir führen für alle EU-25-Regionen eine multiple Regressionsanalyse durch, die Messungen der F&E-Investitionen, Vertreter für regionale Innovationssysteme sowie Wissens- und sozioöonomische Übertragungen umfasst. Mit diesem Ansatz lässt sich zwischen dem Einfluss von internen Faktoren und externen Wissens- und Institutionsströmen auf das regionale Wirtschaftswachstum unterscheiden. Die empirischen Ergebnisse verdeutlichen, wie die komplexen Wechselwirkungen zwischen lokaler und externer Forschung einerseits und lokalen sowie externen sozioökonomischen und institutionellen Bedingungen andererseits die Innovationskapazität jeder Region prägen. Ebenso weisen sie auf die Bedeutung der Nähe für die Übertragung von wirtschaftlich produktivem Wissen hin, da die Wissensübertragung mit zunehmender Entfernung stark beeinträchtigt wird.

Wirtschaftswachstum Innovation F&E Wissen Wissensübertragung Innovationssysteme Regionen Europäische Union

Rodríguez-Pose A. y Crescenzi R. I + D, ‘spillovers’, sistemas de innovación y la génesis del crecimiento regional en Europa. Regional Studies. La investigación sobre el impacto de la innovación sobre el desempeño económico en Europa ha seguido fundamentalmente tres enfoques: (1) el análisis del vínculo entre la inversión en I + D, patentes y crecimiento económico; (2) el estudio de la existencia y eficacia de sistemas de innovación regionales y (3) el examen de la difusión geográfica del conocimiento (spillovers). A pesar de su complementariedad, estos enfoques apenas se han combinado. La presencia de barreras metodológicas y operacionales ha minado cualquier posibilidad de interacción. En este artículo nuestra intención es cubrir este hueco en la literatura, combinando en un modelo los enfoques basados I + D, spillovers y sistemas de innovación. Esto se realiza mediante un análisis de regresión múltiple que incluye variables de inversión en I + D, componentes de los sistemas de innovación regional y spillovers de conocimiento y de carácter socioeconómico. Este enfoque nos permite discriminar entre la influencia de los factores internos y los flujos externos de conocimiento e institucionales sobre el crecimiento económico. Los resultados empíricos subrayan cómo la interacción entre la investigación local y la realizada en otros espacios, por un lado, con las condiciones socioeconómicas e institucionales tanto en el ámbito local como en otras áreas, por otro, influye en la capacidad innovativa de cada región. Los resultados también ponen de manifiesto la importancia de la cercanía geográfica en la transmisión del conocimiento productivo, ya que la eficacia de los spillovers se ve fuertemente afectada por la distancia.

Crecimiento económico Innovación I + D  Conocimiento Spillovers Sistemas de innovación regiones Unión Europea

Keywords:

• Economic growth
• Innovation
• R&D
• Knowledge
• Spillovers
• Innovation systems
• Regions
• European Union

Acknowledgements

The authors are grateful to Roberta Capello, Carlo Pietrobelli, the anonymous referees, and to participants in the seminars held in London, Naples, Rome, Volos, Edinburgh, and Lisbon for their comments to earlier drafts of this paper. The authors are solely responsible for any errors contained in the paper. This paper could not have been written without the financial support of the Royal Society–Wolfson Research Merit Award.

Notes

1. This paper adopts the definition of ‘knowledge’ developed by Döring and Schnellenbach (2006): understanding that ‘knowledge as comprising all cognitions and abilities that individuals use to solve problems, make decisions and understand incoming information … knowledge is a tool that can be consciously used by individuals’ (p. 377).

2. GDP per capita is usually considered as a proxy for the level of productivity: the lower the productivity (GDP per capita) of a region, the farther it is from its technological frontier.

3. Standardized in order to range from 0 to 1.

4. As discussed in the previous section there is no reason that knowledge should stop spilling over just because of the (often arbitrary) boundaries of the NUTS regions on which the analysis is based.

5. Taking into account these caveats, the measurement of spillovers represents not only ‘pure knowledge externalities’ but also, more generally, the broader set of knowledge flows produced by any external source and appropriated by local innovative agents. ‘The pathways by which knowledge spills over in this way are many and various; they include written texts, informal conversations, input–output links, inter-firm mobility of workers, strategic alliances and so on’ (Scott, 2006, p. 9). The analysis of such pathways is outside the scope of this paper which, in this regard, inevitably shares the limitations of other studies based on a similar approach (compare Breschi and Lissoni, 2001).

6. The indicator of accessibility to innovation used in this article is purely geographical. While acknowledging that geographical distance may neither be a sufficient, nor a necessary condition for the assimilation of spillovers, and cognitive, organizational, social, and institutional proximity play an important role in the diffusion of knowledge (Boschma, 2005; Iammarino and McCann, 2006), the quantitative nature of the analysis prevents one from focusing on these other forms of proximity. Hence measurement is made of the geographical distance between different socio-economic structures in regions, but not the social distance between these same structures.

7. As the time–distance matrix is calculated either at the NUTS1 or at the NUTS2 level, in order to make it coherent with the current data which combines different Nuts levels reliance was made on the NUTS distance matrix using the NUTS 2 regions with the highest population density, in order to represent the corresponding NUTS1 level for Belgium, Germany and the UK.

8. In the case of the new Member States, data availability has prevented the calculation of the mean of the explanatory variables over the 5-year period (t − T − 5) forcing the use of a shorter time span. For some EU-15 countries slightly different time spans have been used, as a consequence of differences in data availability for each variable.

9. As far as specific regions are concerned, no data are available for the French Départments d'Outre-Mer (Fr9), Uusimaa (Fi16) and Etela-Suomi (Fi17) were excluded from the analysis owing to the lack of data on socio-economic variables. Trentino-Alto Adige (IT31) was also excluded as it has no correspondent in the NUTS2003 classification. Because of the nature of the analysis, the islands (PT2 Açores, PT3 Madeira, FR9 Departments d'Outre-Mer, ES7 Canarias) and Ceuta y Melilla (ES 63) were not considered, as time-distance information, necessary for the computation of spatially lagged variables, is not available.

10. The value of the Moran's I from the regression residuals is reported in the tables for each regression, alongside the usual diagnostic statistics. The weight matrix for the computation of the Moran's I is based on the same weighting scheme (equations 2 and 3) adopted for the calculation of the spatially lagged variables included in the model (spillovers and social filter conditions of neighbouring regions). In addition to this weighting scheme (based on distance), first order contiguity has been also tested delivering similar results.

11. In this case:

• As a result the variable is equal to the sum of the region's social filter index and the inverse distance-weighted average of other regions' social filter index (accessibility to innovation-prone extraregional areas).