Social capital and local cultural milieu for successful migrant entrepreneurship

Abstract

This paper investigates the cultural gravity mechanism behind the bonding and bridging effects on the performance of migrant entrepreneurship in the Netherlands. The culture-based development gravity approach – suggesting a joint interaction of individual cultural capital and local cultural milieu as a critical factor for local development – will be operationally translated into a model on bonding and bridging preferences of migrant entrepreneurs. A logit model is employed to estimate a culturally augmented standard production function, in which the dependent variables are alternatively a high-profit or increased turnover and in which the independent variables are – in addition to the standard input factors – also the interaction terms of, respectively, bonding and bridging with local cultural milieu. Different specifications of the estimated model, controls for clustering effects and two modifications of the data-set are used to achieve the best extraction of information from the available data-set and to cope with its limitations. We find that bonding is positive in the context of a culturally closed local milieu; yet, this approach secures mostly profits, but is less likely to generate economies of scale for the enterprise. The most interesting result is that the choice to be rather ethnically locked in (i.e. a bonding preference), due to heuristically following the ‘role model’ imposed by the choice of the other migrant entrepreneurs in the locality, has a significantly negative association with the success of the migrant entrepreneur.

Cet article examine le mécanisme de la gravité culturelle qui sous-tend l'impact des liens relationnels et des liens affectifs sur la performance de l'entreprenariat des migrants aux Pays-Bas. L'équation de la gravité par le biais du développement basé sur la culture – suggérant que l'interaction du capital culturel individuel avec celui du milieu culturel local est un facteur critique du développement local – sera traduite opérationnellement en un modèle sur les préférences des entrepreneurs migrants en matière de liens relationnels et de liens affectifs. Un modèle logit est employé pour estimer une fonction de production standard enrichie par la culture, dans lequel les variables dépendants sont alternativement de fortes marges de profit ou un chiffre d'affaires en augmentation, et dans lequel les variables indépendants sont également – au-delà des facteurs standards de production – les termes de l'interaction des liens relationnels et des liens affectifs, respectivement, avec le milieu culturel local. Différentes caractéristiques du modèle estimé, des contrôles pour les effets d'agglomération et deux modifications de l'ensemble des données sont utilisées pour améliorer l'extraction des données depuis leur ensemble disponible et composer avec leurs limites. Nous démontrons que les liens relationnels sont positifs dans le contexte d'un milieu local culturellement fermé ; pourtant, bien que dans la plupart des cas, cette approche garantisse des profits, elle est moins susceptible de générer des économies d'échelle pour l'entreprise. Le résultat le plus intéressant est que le choix de rester enfermé dans sa propre appartenance ethnique (c'est-à-dire une préférence pour les liens relationnels) pour pouvoir suivre de manière heuristique le « rôle modèle » imposé par le choix des autres entrepreneurs migrants dans la localité, est associé de façon significativement négative au succès de l'entrepreneur migrant.

Keywords:

• bonding and bridging
• social capital
• cultural capital
• entrepreneurship
• culture-based development

Mots-clés:

• liens relationnels et liens affectifs
• capital social
• capital culturel
• entrepreneurial
• développement basé sur la culture

JEL Classification:

• Z10
• O31
• O43
• R11

Acknowledgements

The authors wish to thank Mediha Sahin for her assistance in making the data available to us.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1. For an overview on social capital as a concept in a broader sense as well as a critique on the theory of Putnam, we refer to Portes (1998, 2000).

2. For a detailed overview of the broader role of networks for entrepreneurship, see Nijkamp (2003).

3. The essential difference between the classical gravity model and cultural gravity model is in the way they express the role of the gravity function. The standard gravity model uses the flows of economic exchange between two countries to capture the gravity effect between them. Instead, cultural gravity defines the gravitational force possessed by a locality as a characteristic – mixed product of the local cultural milieu and cultural diversity – which acts as a pull factor for human capital of other localities based on the cultural distance between the local culture and the culture of the sending country.

4. The notion of cultural capital originates as such in a concept for intergenerational transmission of social status on individual level between parents and their children, tailored by Pierre Bourdieu (1986). The CBD concept lifts the notion of cultural capital from individual to local level and applies this aggregate level of local cultural capital as a starting notion in explaining cultural impact on place-based development. For more details on the link between individual and local cultural capitals, see Tubadji (2012, 2013).

5. Our data-set contains an equal number of people from each big city; thus, if out of one and the same size of a pool, there are different shares of people opting for the same choice, we assume this is showing a different likelihood of people making a given type of choice in this locality. A limitation of the data-set is that only a stopping rule was applied in determining the number of people interviewed. So, our estimation approach here is providing a methodological illustration about testing a working hypothesis like the one posed here. The economic meaning of the results is to be interpreted within the limitations of the data-set.

6. Heterogeneity within the ethnic entrepreneurs exists, but is only partially addressable through the data-set. Namely, indeed we have five categories of non-Dutch background in the data-set, among which the Moroccans were most widely presented. Yet, the data-set contains concrete sub-ethnic characteristics (country of birth) that can allow us to capture this cultural background heterogeneity only among the ethnic entrepreneurs from first-generation immigrants. We performed such a heterogeneity test, introducing as controls dummies for four of the five sub-ethnic types in first-generation ethnic entrepreneurs. The results show generally the same trends, and the main culture-related results survive. However, this test on itself is partial, as the sub-ethnic types in the second generation are not accounted for. Thus, we do not envisage that these results are significantly distinct for the results presented in Tables 2a and 2b and Tables 3a and 3b.

7. We have an interaction of bonding and bridging once with the availability of bank loans – which to a large extent is an objective rational expression of local cultural milieu openness. Meanwhile, alternatively, we have an interaction between bonding and bridging with an attitudinal variable which is more related with the bounded rationality aspect of the local milieu.

8. Sector specialization sect_IT has a very strong effect on the results and when introduced almost takes over the human capital effect. Firm-age-related control – dummy variable old -- is not significant. We present here the results with all controls, yet for the next step of the estimations, we drop both sectoral specialization and firm age. The first due to a high correlation with human capital, and the second due to its reported statistical insignificance.

9. The economic significance seems also to be quite important when one compares the parameters of the cultural variables, especially the ones gravity is related and the parameters of the rest of the explanatory variables. Yet, we would with over-interpreting these results due to the previously discussed data limitations.

Additional information

Notes on contributors

Annie Tubadji

Dr. Annie Tubadji is a post-doctoral researcher at the University of Bologna, Italy and the University of the Aegean, Greece. She received her PhD in culture and labor market productivity from University of Regensburg, Germany. Her areas of research include entrepreneurship, innovation, economic growth, regional economics, migration, and cultural diversity.

Karima Kourtit

Dr. Karima Kourtit is a post-doctoral researcher at the Royal Institute of Technology in Stockholm and A. Mickiewicz University, Poznan, Poland. She received her PhD from the VU University, Amsterdam, the Netherlands, as well as from the A. Mickiewicz University in Poznan. Her areas of research include entrepreneurship, creativity, urban strategic analysis, and planning.

Peter Nijkamp

Prof. Peter Nijkamp is associated with both the VU University, Amsterdam, the Netherlands and the A. Mickiewicz University, Poznan, Poland. His research interests cover plan evaluation, multicriteria analysis, regional and urban planning, transport systems analysis, mathematical modeling of spatial systems, technological innovation, and environmental and resource management. In his long research career he has focused his research in particular on quantitative methods for policy analysis, as well as on behavioral analysis of economic agents. He has a broad expertise in the area of public policy, services planning, infrastructure management and environmental management.