Does academic research affect local growth? Empirical evidence based on Swedish data

ABSTRACT

Does academic research affect local growth? Empirical evidence based on Swedish data. Regional Studies. The hypothesis that research activities do not merely affect the growth rate within the municipality hosting the university or university college but spill over to neighbouring municipalities is tested using data covering the Swedish municipalities between 1990 and 2010. Research activities are measured as the number of dissertations and the number of papers published in academic journals with peer-review system. The results suggest academic institutions by their presence and ‘production’ of human capital and new knowledge through research affect the regional growth pattern in both the municipality where the academic institution is located and the growth pattern of neighbouring municipalities.

摘要

学术研究是否会影响地方成长？根据瑞典数据的经验证据. 区域研究。本研究运用涵盖瑞典各行政区自 1990 年至 2010 的数据，检验以下假说：研究活动不只会影响大学或学院所在之行政区的成长率，更会外溢到邻近的行政区。研究活动，以在具有同行审查系统的学术期刊发表的学位论文及文章数目进行测量。研究结果显示，学术机构的存在，以及透过研究“生产”的人力资本与新知识，同时影响了学术机构所在行政区及邻近行政区的区域成长模式。

RÉSUMÉ

Est-ce que la recherche académique influe sur la croissance locale? Des résultats empiriques fondés sur des données auprès de la Suède. Regional Studies. À partir des données auprès des communes suédoises entre 1990 et 2010, on cherche à vérifier l'hypothèse selon laquelle les activités de recherche influent non seulement sur le taux de croissance au sein de la commune d'accueil de l'université ou du collège mais débordent sur les communes avoisinantes. On détermine les activités de recherche en termes du nombre de thèses et du nombre d'articles publiés dans les revues académiques à partir du système d’évaluation par les pairs. Les résultats laissent supposer que les établissements universitaires influent à la fois sur le profil de croissance régionale dans la commune où se situe l'université et sur le profil de croissance des communes avoisinantes grâce à leur présence et vu leur ‘production’ de capital humain et de nouvelles connaissances fondées sur la recherche.

ZUSAMMENFASSUNG

Wirkt sich wissenschaftliche Forschung auf das lokale Wachstum aus? Empirische Belege auf der Grundlage schwedischer Daten. Regional Studies. Mithilfe von Daten über schwedische Gemeinden im Zeitraum von 1990 bis 2010 wird die Hypothese überprüft, dass Forschungstätigkeit nicht nur die Wachstumsrate innerhalb der Gastgemeinde der Universität oder Hochschule beeinflusst, sondern sich auch auf Nachbargemeinden überträgt. Die Forschungstätigkeit wird anhand der Anzahl der Dissertationen und in wissenschaftlichen Fachzeitschriften mit Peer-Review-System veröffentlichten Beiträge gemessen. Aus den Ergebnissen geht hervor, dass sich wissenschaftliche Institutionen aufgrund ihrer Präsenz und der Erzeugung von Humankapital und neuem Wissen durch Forschung auf die regionalen Wachstumsabläufe auswirken, und zwar sowohl in der Gemeinde, in der sich die wissenschaftliche Institution befindet, als auch in den Nachbargemeinden.

RESUMEN

¿Influye la investigación académica en el crecimiento local? Evidencia empírica basada en datos suecos. Regional Studies. A partir de datos de municipios suecos entre 1990 y 2010, comprobamos la hipótesis de que las actividades de investigación no afectan solamente a la tasa de crecimiento del municipio donde se ubica la universidad o el colegio universitario sino también a los municipios vecinos. Las actividades de investigación se miden según el número de tesis doctorales y artículos publicados en revistas académicas con sistema de evaluación paritaria. Los resultados indican que las instituciones académicas, debido a su presencia y ‘producción’ de capital humano y nuevo conocimiento mediante la investigación, influyen en el modelo de crecimiento regional tanto en el municipio donde está ubicada la institución académica como en los municipios vecinos.

KEYWORDS:

• net migration
• economic growth
• convergence
• academic research
• human capital
• spillovers

关键词:

• 淨移民
• 经济成长
• 收敛
• 学术研究
• 人力资本
• 外溢

MOTS-CLÉS:

• solde migratoire
• croissance économique
• convergence
• recherche académique
• capital humain
• retombées

SCHLÜSSELWÖRTER:

• nettomigration
• wirtschaftswachstum
• konvergenz
• wissenschaftliche forschung
• humankapital
• übertragungseffekte

PALABRAS CLAVES:

• migración neta
• crecimiento económico
• convergencia
• investigación académica
• capital humano
• efectos indirectos

JEL:

• I23
• O15
• R11
• R12

ACKNOWLEDGMENTS

The author would like to thank the late John M. Quigley, the editors of this journal and two anonymous referees for their invaluable comments made on an earlier version of this paper. Comments from the seminar participants at Umeå University, Örebro University and North American Regional Science Council's annual meeting in Miami, Florida, USA, in November 2011, are also acknowledged.

DISCLOSURE STATEMENT

No potential conflict of interest was reported by the author.

Notes

1. For an overview of empirical studies in this area, see Abreu, De Groot, and Florax (2005).

2. There are, however, some differences between the two measures, dissertations and papers published in index journals, which needs to be clarified. Academic papers published in index journals could be ‘produced’ not only at universities and university colleges but also at other institutions such as (local) hospitals, the central bank, government offices, private firms etc. On the other hand, only universities and some university colleges have the right to carry out assessments at a postgraduate level, and theses have to be defended at a university or university college. This implies that the total number of academic papers produced in a particular municipality is by definition not the same as the number of academic papers produced at universities and university colleges in that particular municipality. The number of papers published in index journals in a particular municipality is defined as papers in which at least one of the authors is affiliated with a university or university college located in the municipality.

3. This variable could reflect higher travel costs to the capital and (national) decision-makers as well as a ‘social’ distance. That is, being physically far from the capital could make it easier to make contact and communicate with (national) decision-makers in one's own town via other media (e.g., mail or telephone).

4. This method is frequently used in the spatial econometrics literature. The so-called spatial lag model has been used in many previous studies on regional growth, including the works of Ramajo et al. (2008), Dall'Erba and Le Gallo (2008), and Mohl and Hagen (2010). For an introduction to spatial econometric models, see the seminal work of Anselin (1988).

5. Because is of dimension (), it is not possible to estimate the elements in together with the other parameters in the model. Therefore, the elements in the weights matrix have to be based on some reasonable prior beliefs, the underlying theory or some hypothesis to be tested.

6. By not row-standardizing the weights matrix, which is the standard procedure in spatial econometrics, the variables , , and are treated as public goods.

7. Chi-square tests of model 1: chi-square = 4.69, p = 0.030; model 2: chi-square = 4.74, p = 0.030; model 3: chi-square = 0.23, p = 0.635.

8. Chi-square tests of model 4: chi-square = 1.34, p = 0.247; model 5: chi-square = 0.58, p = 0.446; model 6: chi-square = 3.10, p = 0.078.