ARCHIVED – Social Capital and Wages - Outcome of Recent Immigrants to Canada

6. Empirical Results

The estimation of the log of real weekly wages is undertaken in longitudinal models, including random effects, fixed effects, Hausman-Taylor models and instrumental variables (IV) model for panel data. Table 2 shows the estimation and relevant test results of the log of real weekly wages for male and female immigrants.[Note 17] 

As the estimated effects of the non-social capital variables are consistent with the theoretical explanations and the findings of related empirical literature, they will not be discussed in detail, but presented in Appendix instead.

6.1. Social capital effects

Looking at social capital variables included in Table 2, it is clear that there are significant relationships between social networks and weekly wages. The results are robust across different statistical models. The directions of the relationships between social capital indicators within various types of social networks and weekly wages are mixed. The channels through which a newcomer gets a job do not make much difference for male immigrants in terms of wages, which is reflected by the always non-significant coefficients of personal contacts as job-found channels. In contrast, jobs found through family ties provide higher wages for female newcomers than those found through other methods, varying from 4.5% (random effects model) to 15.1% (panel IV model), but the significance is not observed in the panel IV model. The size of kinship network always has a negative coefficient for both male and female newcomers, but only sometimes statistically significant (in fixed effects model and Hausman-Taylor model). The size of the friendship network has different impacts on wages of different genders. It has a non-significant positive effect on male immigrants’ wages but has a significant negative impact on female newcomers’ wages. While a more diverse workplace network is associated with higher employment earnings for both male and female newcomers, the magnitude of the effect differs across models.

The coefficients from panel IV models are quite different from those obtained from non IV ones and Hausman-Taylor models, not only in the magnitude but also in the significance. This difference may result from different assumptions about endogeneity of social capital: with unobserved heterogeneity or with disturbances term. Hausman tests between different pairs of estimates provide evidence towards identifying assumptions and choosing among models.

The p-values of the Hausman tests are shown in the bottom row of Table 2. For male and female immigrants, the Hausman tests for (fixed effects – random effects) are X² (44) = 269.71 with p-value of 0.0000 and X² (44) = 254.35 with p-value of 0.0000, respectively. Thus we can reject the null hypothesis that assumption (3) holds and fixed effects model is favoured over random effects model for both men and women samples. In contrast, the Hausman tests for (Hausman-Taylor estimator – fixed effects) show that we cannot reject the null hypothesis of assumption (5) for both gender samples; therefore the set of instruments x1 and z1 chosen are legitimate and the Hausman-Taylor estimator is consistent and efficient.

The Hausman tests for (FE2SLS – EC2SLS) are X² (43) = 108.02 with p-value of 0.0000 and X² (43) = 151.53 with p-value of 0.0000 for male and female immigrants, respectively. This result shown in Table 4 rejects the null hypothesis that EC2SLS yields a consistent estimator. An additional Hausman test based on the difference between fixed effects 2SLS and ordinary fixed effects estimators fail to reject the null hypothesis that ordinary fixed effects estimators are consistent. In other words, there is no significant evidence that social capital variables are correlated with the disturbances term  in the wages equation. Recall that the Hausman-Taylor estimators are consistent and efficient compared with fixed effects. Thus the Hausman-Taylor (HT) estimators are favoured among all panel data models in the current research, and the rest of the paper will focus on the results from the HT estimators.

* p<0.1; ** p<0.05; *** p<0.01.

Note: The Hausman-Taylor estimates assume the endogeneity of social capital variables, education, skill level, job tenure and working hours with unobserved heterogeneity. The estimations also include control variables for immigration category, demographic characteristics (age, marital status), province of residence, region of birth, ethnic group, education, official language skills, previous experience or attachment in Canada and occupational characteristics. See Appendix for complete results.

Data source: Longitudinal Survey of Immigrants to Canada (2005).

For male immigrants, the elements within social networks which play a role in determining employment wages are workplace ethnic diversity, kinship size and frequency of contact with family sponsors. An increase in workplace ethnic diversity from a total concentration in one ethnic group to a total diverse workplace network would increase the log of real weekly wage by 13.8%. The frequency of contact with sponsors shows a return of 5.6% in wages. The geographic closeness of relatives and participation in organizations also shows marginally significant (at 10% level) effects on real weekly wages, though the directions of the effects differ. Size of kinship has a significant negative return for male newcomers. One more relative[Note 18] in Canada is associated with a 10.3% wage discount.

The effect of using personal contacts (i.e. family or friends) to obtain work on wages is generally positive but not significant for male immigrants.

For female newcomers, while a job found through family members or relatives pays 7.1% higher than jobs found through other methods, the kinship size does not do the same: one more relative in Canada relates to 11.9% less wages. Friendship size also has a significant and negative effect on real weekly wages, though the effect is small (-1%). Again, a totally ethnic diverse workplace implies a big wage gain of 14.5%, compared to 13.8% for male newcomers.

Allowing social network variables to be endogenous with unobserved heterogeneity has a large effect on the estimated returns to social capital. It can be seen that introducing the endogeneity of social capital indicators with individual effects reduces or increases the return to social capital by about 20 to 30 percent for both sexes. Comparing column 1 and column 3 in Table 2, the return to workplace diversity decreases from 18% using random effects model to 13.8% using the Hausman-Taylor estimator for male immigrants, a decrease of 23%. For females, the return to social capital estimated with random effects model is also very different from the return using the Hausman-Taylor estimator: the return to family ties as a job search channel increases to 7.1% from 4.5% while the return to workplace diversity drops from 18.6% to 14.5%, a drop of 22%. When taking into account the endogeneity of social capital variables with individual effects, the negative effects of kinship size also become bigger and statistically significant, for both male and female immigrants. 

6.2. Differential social capital effects

To further explore the differential effects of social capital indicators, interaction terms of social network variables with education, language skills, ethnic group and immigration class are included separately in the Hausman Taylor models. The final specifications presented in Table 3 put all significant interaction terms together, excluding non-significant interactions for men and women immigrants respectively. In Table 3, the specifications of interaction models for male and female immigrants are presented based on the same assumption that social capital variables, education, skill level, job tenure and working hours are endogenous with individual specific effects. This closer look at the coefficients of interaction terms reveals that social capital affects wages of immigrants with varied characteristics quite differently.

For male immigrants, social capital, especially family networks, has stronger effects on less-educated immigrants’ earnings. Specifically, when interacted with education attainment, the size of family ties plays different roles on weekly real wages for immigrants with varied education levels. Relative to those with university degrees, those less educated benefit more from the number of relatives – having much higher returns to the number of relatives in Canada: 20.9% higher for those who had some post-secondary education, 14.1% more for those with a college diploma and 2.2% more gain (though not significant) for those who had a high school diploma or less. Furthermore, the return to family ties as a channel for finding a job is also larger for those with less education: 21.1% more gain for those high school graduates, 18.5% more for those with some post-secondary schooling and 13% more for college graduates, relative to the jobs found through non family members or relatives. It is interesting that those with higher education are also more likely to benefit from family ties in terms of 16.1% higher return in the jobs found through relatives than jobs found through other ways, though the effect is only marginally significant at the 10% level.

Jobs found through relatives earn 25% more and those obtained through co-ethnic friends pay 9.7% less for francophone immigrants than for those without any French knowledge. Despite that the pure effect of number of relatives is significantly negative (- 42.5%), the effects of these close ties are more positive (or less negative) for almost all visible minorities, especially for Chinese, West Asian and Arab male immigrants (33.1% and 50.5% more, respectively) compared to White immigrants. When looking at disparity among immigration categories, kinship size has less negative effects on wages of economic immigrants (skilled workers and business immigrants), particularly on those of skilled worker principal applicants (-20.8% = - 42.5% + 21.7%, compared to other classes).

For female newcomers, there is not much differential effect across ethnic groups or immigration categories. Using family ties as a method for finding a job is 16.2% less beneficial to those with English language skills.

However, workplace diversity seems to be most beneficial for those with a university degree, and not for the less educated, neither more educated. For example, workplace diversity has 35.5% weaker effects for those with a high school diploma or less, 46% less returns for those with some post-secondary education, and 34.3% weaker for female immigrants with a master’s degree or a PhD.

The literature indicates that there are differences in social capital impacts among groups of immigrants - social capital effect on wages is amplified for unskilled workers or undocumented migrants (e.g. Beine, Docquier and Ozden 2007; Aguilera and Massey 2003). Most of the results in this paper seem to support the previous literature that social capital effects, especially strong ties with family members and relatives, are amplified for immigrants with disadvantaged human capital, such as education and official language skills.

* p<0.1; ** p<0.05; *** p<0.01.

Note: The interaction effects are estimated with Hausman-Taylor estimator assuming the endogeneity of social capital variables, education, skill level, job tenure and working hours with unobserved heterogeneity. The estimations also include control variables for immigration category, demographic characteristics (age, marital status), province of residence, region of birth, ethnic group, education, official language skills, previous experience or attachment in Canada and occupational characteristics. See Appendix for complete results.

Data source: Longitudinal Survey of Immigrants to Canada (2005).

Notes

¹⁷ Table 2 only shows the coefficient estimates for social capital variables. For complete estimation results, see Appendix.

¹⁸ As the number of relatives is measured by the number of types of relatives in Canada, this result should be interpreted with caution.

• Previous
• 1
• 2
• 3
• 4
• 5
• 6
• 7 (current)
• 8
• 9
• 10
• Next