The economic returns of bonding and bridging social capital for immigrant men in Germany

Abstract

Using longitudinal data, this paper analyses the effect of different forms of social capital on the likelihood of employment and the occupational status of first generation immigrant men in Germany. This allows me to examine to what extent social capital of the bonding and the bridging types yield different returns. The study considers how contacts with natives, co-ethnic ties and family-based social capital are beneficial to the economic position of immigrant men. Random effects and fixed effects models show that strong inter-ethnic ties are beneficial both for employment and occupational status. There is no effect of co-ethnic ties and family-based social capital. It is concluded that, when using panel data, bridging social capital contributes to a better economic position and bonding social capital does not.

Keywords:

• Social capital
• labour market
• immigrants
• longitudinal analysis
• Germany

Notes

1. Immigrants are either included in the ‘foreigner’ sample, (started in 1984, consisting of Turks, Yugoslavians, Greeks, Italians, and Spanish), in the ‘immigrant sample’ (started in 1994–1995, consisting of households in which at least one member moved from abroad to Germany), or in the ‘refreshment’ sample (1998 and 2000).

2. The attrition rate is most likely lower because only the working age population is included.

3. Persons with German nationality were not included because the construct ‘visiting Germans at home’ was not asked in all waves to people with German nationality. Furthermore, they were not included to make sure that the ties mentioned are measured as bridging and bonding in the same way for the whole sample.

4. This classification effectively separates native Germans and ethnic minorities. In one case it is less clear. Ties that belong to the second generation and have German citizenship could be classified as either ‘from another country’ or as ‘from the old/new Federal states’. In the follow-up question (‘Do you come from the same country [as the person mentioned]’), it appeared that in 98 per cent of cases, people indicate them to be from the same country as themselves (for people born in Germany but no German citizenship, this percentage is 93). That is, the country referred to is the country of origin. This is most likely the same when immigrants classify ties that belong to the second generation and with German nationality.

5. The categories are slightly collapsed: included are inadequately/general elementary (originally separate categories), basic vocational, intermediate vocational/general (originally separate categories), general or vocational maturity certificate and tertiary education. Models were also estimated including parental education. This did not yield different results.

6. Cronbach's alpha ranging from .83 to .86 across waves.

7. Due to the low N in some of the states several categories are collapsed: Mecklenburg-Western Pomerania, Brandenburg, Saxony, Saxony-Anhalt and Thuringia are collapsed into one category.

8. However, the Hausman test favours the FE model. Furthermore, the standard errors of the FE model are robust.