THE DISTRIBUTION AND RETURN OF SOCIAL CAPITAL: EVIDENCE FROM SWEDEN

ABSTRACT

This paper studies the role of social capital in the status attainment process and examines the link between the hiring process and the potential pool of social capital embedded in a person's network. The analysis is based on a sample of people newly employed by the municipal services in Malmö, the third largest city in Sweden. Jobs in this sector of labour market are mainly low-paid, and are dominated by women and immigrants. The position generator method is used to measure social capital, understood as assets captured by individuals in social networks. The findings demonstrate that access to social capital is positively related to work experience, a higher educational level, having a partner, and active membership of voluntary associations. It is also apparent that being an immigrant is associated with a substantial social capital deficit. Regarding the return on capital, the results show that both human capital and social capital were rewarded with higher wages and more adequate jobs. Furthermore, we found that social capital is associated with better labour market outcomes, whether or not respondents reported that they obtained their current jobs using informal job-search methods. Results also show similar returns on access to social capital for natives and immigrants.

Keywords:

• social capital
• hiring process
• immigrants

Acknowledgements

Many thanks to the anonymous reviewers of European Societies for their helpful comments and suggestions. Thanks also to Carl-Ulrik Scheirup, Mahmood Arai and Anders Neergaard for valuable comments.

Notes

¹Countries defined as NW in this study are: Denmark, Finland, Norway, Island, UK, France, Italy, Germany, The Netherlands, Belgium, Spain, Austria, Ireland, Luxemburg, Switzerland, Japan, Canada, Australia, New Zealand and the United States. The rest of the world is defined as ONW. The main criteria for division is that members of the first group (NW) each have an annual GNI per capita (formerly GNP per capita) of more than US 20,000 (see the World Bank, http://www.worldbank.org/data/, GNI per capital 2004, Atlas method and PPP). The rest of the world is defined as ONW.

²Jenkins (1986:47) defines ‘acceptability’ as a (not explicitly described or formally specified) spectrum of criteria ranging from appearance, attitude, personality, ability to ‘fit in’, to a manager's ‘gut feeling’. Acceptability is also all ‘criteria which depend to a greater or lesser degree upon shared cultural competence’ and therefore ‘seems probable that the realm of acceptability is the most likely setting for the operation of discrimination in employment recruitment’ (Ibid: 50).

³Regarding the third characteristic, Erickson (1996) argues that people with more varied networks have more varied cultural repertories, which help them to build smoother working relationships with a wider variety of other people, including those in different levels in their own firm.

⁴It would be more eligible to use a status score for various occupations in Sweden. In the absence of such specific occupational prestige scale we use SIOPS which seems to be not very far from the Swedish context.

⁵See also Appendix Table 2 for a summary of position-generated variables in the sample.

⁶As Bollen (1989) suggests, latent variables (or unobserved and unmeasured variables) correspond to particular abstract concepts such as power or social class that are central to many social science theories but at the same time are only indirectly measurable. Social capital, which is the central concept of this study, has the same feature. To test theories about these concepts, researchers try to collect observable measures of these latent variables. It is the theoretical definition that provides guidance in the selection of measures.

⁷Coefficient for education years in an OLS regression, as in Table 3, with this variable rather than educational level is 0.089 and significant at 1 per cent level. (Results of all not shown here estimations are available from the author on request.)

⁸The mean age of immigrants at the time of immigration has been 27.5 years.

⁹I have not included the variable ‘active membership of voluntary associations’ as a control variable here because there is no theoretical support for the inclusion of this variable in labour market outcome estimations.

¹⁰To get a sense of the relative impact of social capital, education and job experience, I redefined social capital, following Erickson (2001), as network variety (or the simple count of the number of different occupational categories in which the respondent reported knowing someone) and ran the same logistic regression as in Model 3. For social capital so defined (as network variety), knowing someone in one additional line of work multiplies the odds of being in the highest-wage group (BIHWG) by 1.070. Knowing someone in two additional lines of work multiplies the odds of BIHWG by 1.15, which is nearly the same effect as having one additional year of work experience (1.149). Knowing someone in six additional lines of work multiplies the odds by 1.52, nearly the same effect as having one additional year of education (1.485).

¹¹We have in addition used Ordinary least-squaring regression to estimate the effect of social capital on wages, while taking into account the same control variables as here and with the same models as Table 6. The exogenous variable in the estimation was ordinal, with six wage categories (less than 10 000 SEK = 1; 10,000–15,0000 SEK = 2; 15,001–20,000 SEK = 3; 20,001–25,000 SEK = 4; 25,001–30,000 SEK = 5; and more than 30,000 = 6). The results appear in Appendix Table 4 and demonstrate that the significance and the strength of association for the variable of main interests is roughly the same as here.

¹²About 80 per cent in our sample are trade union members. Membership tends to be higher among those with permanent jobs, full-time jobs and higher education.

¹³To answer the question about the net effect of the variables education, experience and social capital on the probability of having an adequate job, squared semi-partial correlation sr_i ² for each variable was estimated, because of the correlation between the variables of main interest among independent variables (see Tabachnick and Fidell 2001: 141). The squared semipartial correlation (sr_i ²) for the variables education, experience and social capital are 0.286, 0.190 and 0.101, respectively.

¹⁴The negative coefficient of being born in NW countries is not statistically significant in this estimation.

¹⁵Since information on years of completion of highest attained degree in not available in many cases, we have not been able to present results for this estimation (and for other estimations in this paper) for the subsample of immigrants who attained their final degree in Sweden.