Residential and industrial enclaves and labor market outcomes among migrant workers in Shenzhen, China

ABSTRACT

This study examines the association between participation in localistic enclaves and labour market outcomes among internal migrants in Shenzhen, China. Using data from the 2005 mini-census, we examine native-place residential enclaves and industrial enclaves on the basis of migrants’ counties of origin, residential neighbourhoods, and two-digit industrial sectors, and report four main findings. First, migrants who live in native-place residential enclaves tend to earn less than those who do not, but migrants who work in industrial enclaves tend to earn more than those working in the open economy. Second, the earnings disadvantage of new arrivals relative to established migrants is smaller in a residential enclave than that outside an ethnic enclave, and the advantage of working in industrial enclaves disappears for those new arrivals, compared to migrants who have left their place of origin more than five years. Third, the earnings return to years of schooling is smaller for migrants who live in or work in native-place enclaves than for those who do not. Finally, migrants in residential or industrial enclaves are more likely to be self-employed or to be employers. Our findings suggest that residential and economic enclaves may play different roles in labour market outcomes.

KEYWORDS:

• Migration
• enclaves thesis
• residential enclaves
• industrial enclaves
• labor market outcomes

Acknowledgements

The work described in this study was supported by the Research Grants Council of the Hong Kong Special Administrative Region, China (General Research Fund 16604516, Collaborative Research Fund C6011-16G to Xiaogang Wu). The authors would also like to thank the Shenzhen Municipal Bureau of Statistics for access to the 2005 min-census data, and Albert Park of HKUST for their valuable comments and suggestions.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Correction Statement

This article has been republished with minor changes. These changes do not impact the academic content of the article.

Notes

1 In 2005, there were a total of 2,862 counties in mainland China (National Bureau of Statistics 2006).

2 County is also the practical unit to define dialect groups (e.g., Mandarin, Wu, Min, Xiang, Gan, Hakka, and Cantonese). It allows us to distinguish among migrants by place of origin and to perceive them as different ‘ethnic’ groups.

3 In our sample, migrants coming from Guangdong, Hunan, Hubei, Sichuan, Jiangxi, Guangxi, and Chongqing accounted for 27.2%, 13.6%, 10.3%, 9.1%, 6.6%, 6.5%, and 3.6%, respectively, of all migrants in Shenzhen.

4 The 72 migrant groups consisted of 52,109 migrants, with at least 350 migrants in each group.

5 In 2005, the official number of residents in Shenzhen was 8.28 million, and there were 622 residential communities (Shenzhen Municipal Bureau of Statistics 2006). There were 95 two-digit industrial sectors listed in the China Standard Industrial Classification System in 2005 (GB/T 4754-2002).

6 Our data do not include information on the location of business with co-ethnic concentration, and therefore we are unable to construct the measure of ethnic economy used by Portes and associates (Portes and Jensen 1989; Wilson and Portes 1980). In this study, the industrial enclaves were defined as virtual economic communities in the labor market.

7 The thresholds are set to be large enough that only major residential neighborhoods or industrial sectors are included as enclaves. For determining residential enclaves, the threshold values of 3, 4, 5, 6, and 8 yield similar results. For industrial enclaves, the threshold is the same as that used by Logan, Alba, and McNulty (1994). We obtain similar results with cut-off values of 1, 2, and 3.

8 The positive association between working in an industrial enclave and earnings does not exist for new arrivals. This may be because new arrivals would just select jobs that are not well-paid, rather than being unemployed.

9 In the Chinese context, a schooling of 7.3 years usually refers to educational attainment at primary school, in which at least 6 years of schooling is officially required. In other words, migrants whose educational attainment are primary school or less tend to earn less if they live in ethnic enclaves. In our analytical sample (n=37,184), around 10% (i.e., 3,877/37,184) only received schooling at primary level or below, with 34% (i.e., 1,316/3,877) living in enclaves and 66% (i.e., 2,551/3,877) living elsewhere, respectively.

10 We conducted an additional analysis to look at the differences in earnings between migrants (inside and outside enclaves) and the local residents. We ran the regression models with an analytical sample consisting of local residents, migrants outside enclaves, and migrants in enclaves. While living in residential enclaves tends to enlarge the gaps between migrants and local residents, working in industrial enclaves tends to narrow the gaps between migrants and local residents. These results are consistent with our main arguments that living in residential enclaves is negatively associated with earnings and working in industrial enclaves leads to higher earnings compared to working in the open economy among migrants.

11 Unfortunately, we were not able to obtain any previous dataset that can be used to measure enclaves by county of origins. Further research is needed to investigate the causation between enclave participation and labor market outcome when appropriate datasets are available.

Additional information

Funding

The work described in this study was supported by the Research Grants Council of the Hong Kong Special Administrative Region, China (General Research Fund 16604516, Collaborative Research Fund C6011-16G to Xiaogang Wu).