Social Exclusion Among Older Rural Residents in China: Does Either Non-Agricultural Work or Living Away from Their Hometowns over the Courses of Their Lives Make any Difference?

Abstract

This paper analyses the prevalence of social exclusion experienced by older rural residents in China and investigates the impact of work and residence history on rural residents’ social exclusion in later life. Data are from the China Health and Retirement Longitudinal Study (CHARLS). Logistic regression models are used. Results show that rural males with non-agricultural work experience are less likely to experience exclusion from social participation, financial products and common consumer goods and report perceived exclusion than those who have only engaged in agricultural work over their life course. However, no significant associations between work history and the five domains of social exclusion were found among rural females. The change in place of residence itself leads to higher levels of perceived exclusion for both males and females. Policymakers should focus on promoting social participation and the psychological health for return migrants, particular for rural females.

Keywords

gender migration quantitative methods rural return migrants work experience changes in place of residence

Introduction

The large-scale population flow from rural to urban China has been a notable phenomenon since the mid-1980s (Cai et al., 2008). According to the Migrant Workers Monitoring Survey Report (National Bureau of Statistics, 2020), the number of migrant workers reached 290 million by the end of 2019. Meanwhile, emerging studies indicate that returning migrants have increased significantly over the past decade. Using data from the China Labour-Force Dynamic Survey, Xie and Zhou. (2017) and Liu (2017) found that approximately 25–35% of migrant workers returned to their place of origin (or hometown) permanently. It is challenging for return migrants to adapt to an environment that they were once familiar with after returning from a long period of absence and social isolation, which is often a significant issue for return migrants (Barrett & Mosca, 2013).

How is Return Migration Unfolding in China?

Within an international context, return migration refers to ‘migrants who return to their country of origin after a significant period of time abroad’ (Dustmann & Weiss, 2007; Lee, 1966). In the Chinese context, return migrants comprise two groups of rural residents: those who move to urban areas to take up job opportunities and return to rural areas to live, and those who change from agricultural to non-agricultural roles (Liu et al., 2020).

Similar to immigrants worldwide, higher salaries in the non-agricultural sector and better living conditions in cities are important drivers for migrant workers in China. Prior to the economic opening up in 1978, rural residents were restricted from travelling or moving to urban areas and across different regions in China due to limited work opportunities and resources (i.e. food) in urban areas (Cai et al., 2008).

The economic reform greatly expanded the non-agricultural sector in urban areas and generated a continuous need for labour. In response to such changes, the authorities eased relevant policies to facilitate rural–urban migration. These actions triggered a large-scale migration from the agricultural sector to the non-agricultural sector, from rural to urban areas. Thus, the proportion of all employment within the agricultural sector decreased from 70.5% to 47.8%, from 1978 to 2015 (National Bureau of Statistics, 2015).

Return migration was less prevalent compared to the dominant flow of out-migration, which was around 2000 (Zhao, 2002). Most early literature attributes migrants’ return to their hometown to personal reasons, such as caring for family members and health conditions (voluntary choices), or a lack of necessary education and skills to secure a job in urban areas (Wang & Fan, 2006; Zhang, 2011). With the expansion of this group, most recent studies begun to focused on how policies could lead to the involuntary return of migrant workers. For example early temporary policies, including stricter checks of residence permits due to decreased job opportunities during the lay-off wave in the 1990s and the global financial crisis in 2008, have led to a significant trend in return migration (Xu et al., 2017). More importantly, discrimination against migrant workers and their families, in terms of access to social insurance, housing, education and other social rights became the main reason why migrant workers choose to return to their hometown in later life (Huang & Guo, 2017). Obtaining permanent residence and long-term working positions in urban areas, especially in large cities, thus remains challenging for many migrant workers.

Regarding the impact of this increasing urban-rural migration group in China, existing research focuses on ‘entrepreneurs’ who start up their own businesses after their return, and highlights that these return members of villages bring capital and work positions that contribute to local economy (Zhou et al., 2017). However, given that the return migration is a heterogeneous group in terms of age and education, studies on the later lives of ordinary return migration remain limited.

Social Exclusion of Return Migrants in Later Life

Social exclusion refers to the separation of individuals and groups from mainstream society, which includes six domains: (1) material and financial resources; (2) services, amenities and mobility; (3) social relations; (4) neighbourhood and community; (5) sociocultural aspects and (6) civic participation (Burchardt et al., 1999; Kneale, 2012; Levitas, 2005; Nolan & Whelan, 1996; Walsh et al., 2017).

There is no doubt that return migrants experience social exclusion from mainstream society after their return, which may extend further into later life (Barrett & Mosca, 2013). A theoretical framework to explain the factors that may lead to social exclusion of return migrants includes at least two perspectives: cumulative challenges with reintegration (the longer the time of absence, the more challenging the reintegration); and the reserved attitude of locals (Barrett & Mosca, 2013). Return migrants’ work and residence histories which are related to their acquisition of urban modernity and to potential adjustments due to their relocation could form different habits or interests. These may develop more difficulties for them to re-adapt to the local community after their return, which could have impact on social exclusion in later life.

Pursuing better economic prospects is one key reasons for immigration, since return migrants are more likely to have a higher income from employment in urban areas than they would have received from agricultural work, and are likely to have accrued more savings than local people (Barrett & Mosca, 2013). In addition, return migrants are generally more highly skilled than local residents (Bai & He, 2003; Liu, 2017; National Development and Reform Commission, 2017). Therefore, in terms of the domain one listed above, return migrants would be less likely to be excluded from the material and financial resources than those who have never lived outside of their hometown.

On the one hand, out-migration impacted negatively on social relations with migrants’ parents or their children (Van Regenmortel et al., 2016) as migrants could not provide physical and emotional support to their family whilst they were away. On the other hand, in-migration had a negative impact on social relations with neighbours, because neighbours felt that migrants were not making contributions to their communities (Van Regenmortel et al., 2016). Return migrants could also be perceived as being in-migrants by the local community, and as a result, may face hostility. Therefore, in terms of social relations (Domain three listed above), return migrants may form tenuous relationships with local people, which could then lead to social disengagement in later life (Urbaniak & Walsh, 2019). In-migrants also need to work harder to re-establish close relationships with their family members. In addition, return migrants may experience a change in values and norms during their stay in urban areas, and the longer the time they have lived away, the more difficult it is to re-adapt to the local community after their return. Their subjective feelings about their social environments may lead to higher levels of perceived exclusion (Domains 4, 5 and 6 listed above) (Liu, 2017). All of these factors may contribute to conflicting feelings and experiences of social exclusion after their return.

Based on traditional gender roles, males and females play different roles in terms of household tasks and in employment earlier in the life course, which can lead to different experiences of social exclusion in later life. Females are less likely to be migrant workers than males and more likely to undertake caring responsibilities and become ‘left-behind’ family members (Hooper, 1998; Liu, 2014). Thus, females’ experiences of migration and resulting experience of social exclusion more often are different to males.

Research on social exclusion among return migrants in China is limited. It is unclear how return migrants living and working in urban areas, engaging in non-agricultural roles could affect their social exclusion when they return to their hometown in rural areas in their later life. Return migrants face a range of different types of social exclusion when they return to their hometown. However, return migrants may also gain experience in non-agricultural roles or whilst living in urban areas that could help them to re-adjust to living in their hometown. This study uses the life-course perspective to investigate the social exclusion of older return migrants, and explores whether work and residence history in early life has a cumulative impact on the occurrence of social exclusion by gender.

Data and Method

Data

Data obtained from the China Health and Retirement Longitudinal Study (CHARLS), a nationally representative sample of adults aged 45 and over living in China, contains rich information on respondents’ demographic backgrounds, as well as details of employment, financial status and social insurance. The first wave of CHARLS was conducted in 2011, which included more than 17,000 individuals from 450 villages, and respondents were re-interviewed in each following wave. At present, there are five waves of CHARLS: 2011, 2013, 2015 and 2018, and a life history survey, which was collected in 2014. In this study, we use the information about individuals’ lifetime residence and work history from the life history survey (2014), and social exclusion status and other key individual characteristics from Wave 2 (2013), as these two waves have the highest matched number of respondents. The merged dataset contains 8528 respondents aged 60 and over. We further limit our sample to rural residents only (N = 5093), which are identified by holding agricultural Hukou over their life courses. The 726 respondents were excluded with missing information on work history, residence history or any domain of social exclusion. In addition, individuals with military experience and with no surviving children were also excluded. The final number of observations was 4367.

Outcome Variables

Five domains of social exclusion are available from the CHARLS data: relationships with children, social participation, financial products, common consumer goods and perceived exclusion. These five domains of social exclusion were largely based on Kneale’s (2012) and Feng et al.’s (2018) construction of social exclusion.

• Relationships with children. This stands for the domain of ‘social relations’ in the conceptual framework. It reflects whether respondents report having relationships with children, and the extent to which these relationships are maintained through meeting in person, or through remote communication, for example using the telephone (Kneale, 2012). There are three questions regarding respondents’ contact with their children: co-residence with children, frequency of meeting up with non-co-resident children and frequency of contact with non-co-resident children (by telephone, text, etc.). For the first question, respondents are allocated 1.5 points if they live with their children, and those who do not live with their children are allocated 0 points. For the second question, different frequencies of meeting up or contact with non-co-resident children are transferred into points: 1, 0.8, 0.6, 0.4, 0.2 and 0 respectively (1 = almost every day; 0.8 = two to three times a week or once a week; 0.6 = once or twice a month; 0.4 = once every 3 to 6 months; 0.2 = once a year; 0 = almost never). For the third question, the frequency of contact with non-co-resident children is allocated 0.5, 0.4, 0.3, 0.2, 0.1 and 0 points, respectively (0.5 = almost every day; 0.4 = two to three times a week or once a week; 0.3 = once or twice a month; 0.2 = once every 3 or 6 months; 0.1 = once a year; 0 = almost never). If respondents have more than one child, the child with whom they have the most contact is selected to represent the frequency of meeting up and contact with children. Scores of relationships with children are summed by the unweighted points from all three questions. The range is from 0 to 3. We dichotomised this into 1 (‘excluded from relationships with children’) for scores smaller than 1, and 0 for scores 1 and above (‘not excluded’). Exclusion from relationships with children may be treated as an individual’s inability to maintain relationships (Kneale, 2012).

• Social participation. This indicates respondents’ contact with friends, or participation in social and cultural activities, and it also reflects their attachment to their neighbourhood and community. There are five types of social activity in the questionnaire: visiting friends, offering help to friends, recreational games with friends (i.e. cards or Mahjong), playing sports games with friends, and other social activities with social groups (e.g. club activities, volunteering and charities). Respondents were allocated one point if they did not participate in any social activities (‘excluded’) and 0 points if they participated at least one social activity (‘not excluded’). Exclusion from social participation reflects alienation and distance from mainstream society.

• Financial products. This emphasises the availability of financial resources, including stable income, access to social insurance and house tenure that older people are able to access to help with managing their finances (Kneale, 2012). There are three items relating to this domain: having more than 100 Yuan in cash or bank deposit at the time of interview (short-term), having pension income (longer term) and owning any property. Respondents were allocated one point if they had none of the three items (‘excluded’) and 0 points if they reported they had any of the items (‘not excluded’).

• Common consumer goods. This indicates the ownership of goods which are owned by the majority of the population (Kneale, 2012). There are six items relating to this domain: the possession of fridges, TVs, mobile telephones, access to water, electricity and heating devices at home. Respondents received one point for each type of household durable goods and 0 points if they did not possess them. The total score ranged from 0 to 6. Respondents with scores lower than 5 were regarded as ‘excluded’ (value 1). Exclusion from possessing common consumer goods is widely accepted as a direct indicator of material poverty.

• Perceived exclusion. There are five questions about the frequency of respondents’ feelings of being ignored, isolated and lonely in the last week, including ‘I feel depressed’; ‘I feel scared/afraid’; ‘I feel that it is hard to do everything’; ‘I feel lonely’ and ‘I feel that I could not get going’. These refer to perceived absence of intimate relationships. For each question, respondents report the frequency of their feelings: less than 1 day per week, one or 2 days per week, three or 4 days per week and five or more days per week. For each question, respondents received one point if they reported these feelings fewer than 3 days a week and 0 points if they reported these feelings more than 3 days a week. The total score of perceived exclusion ranged from 0 to 5. Respondents with scores lower than 3 were regarded as ‘excluded’ (value 1).

Explanatory variables

Non-agricultural work experience and experiences of residence in urban areas are used to identify return migration in empirical studies (Liu et al., 2020). This study uses work and residence history as two key explanatory variables. Work history is a binary variable: value 0 represents never having carried out non-agricultural work over their life course, and value 1 represents ever having carried out non-agricultural work. Non-agricultural work includes non-agricultural employment (i.e. work in state-owned, collective, private-owned or foreign-funded enterprises) and non-agricultural self-employment (i.e. running small-sized enterprises or engagement in family business). Agricultural work includes household agricultural activities and agricultural employment (i.e. working in collective farms or other landowners). Work history is constructed based on respondents’ work trajectories from ages 20 to 60 taken from the life history survey. If respondents experienced any lengths of non-agricultural work (6 months or more), they are grouped as ever having done non-agricultural work.

Residence history includes three categories: (1) Respondents who have lived in the same rural village over their life course; (2) Respondents who have moved between villages and (3) Respondents who had lived in urban areas and have returned to rural areas by the time of interview in 2013. This variable was constructed using information on respondents’ place of birth and changes in place of residence since birth from the life history survey.

Control variables include age group, education, ethnicity, household expenditure, whether they have any ADL limitations, number of surviving children and the province of origin.

Method

We apply logistic regression models to explore the impact of residence history and work history on social exclusion among rural residents. For each domain of social exclusion, four models are applied: Model 1 includes residence history as the explanatory variable; Model 2 includes work history as the explanatory variable; Model 3 includes both residence history and work history as the explanatory variable; and Model 4 additionally includes interaction between work and residence history as the explanatory variable. All models include all control variables. Gender-specific analyses are adopted in the aforementioned models.

Results

Table 1 shows the descriptive statistics between rural men and women. Regarding work histories, 38.4% of rural men and 10.7% of rural women have ever carried out non-agricultural work. Regarding residence history, over 90% of respondents lived in rural areas over their life course; only 6.5% of rural men and 1.9% of rural women have ever lived in urban areas and returned to rural areas. These results reflect that non-agricultural work experience and changing residence between rural and urban areas could represent two different life course of return migrants.

Table 1.

Characteristics of the Sample (N = 4367).

	Males (N=2069)	Females (N=2298)
Age group (%)
60–64	39.8	41.6
65–69	26.4	27.1
70–74	18.4	17.3
75+	15.4	14.0
Ethnicity (%)
Han	94.4	93.1
Other	5.6	6.9
Education attainment (%)
No formal education	20.2	58.1
Did not finish primary school	27.4	21.7
Primary school	34.2	15.4
Middle school	14.9	4.4
High school or above	3.2	0.5
Marital status* (%)
Married	88.9	76.2
Widowed	11.1	23.8
Have any ADL difficulties (%)
No	19.4	28.2
Number of surviving children (%)
1 surviving child	5.7	3.9
2 surviving children	23.4	19.7
3 surviving children	28.3	27.0
4+ surviving children	42.7	49.4
Work history (binary)
Never having carried out non-agricultural work	61.6	89.3
Ever have non-agricultural work	38.4	10.7
Residence history
Lived in the same village since birth	79.3	76.1
Have moved between villages	14.2	22.0
Lived in urban areas and returned to rural areas	6.5	1.9
Social exclusion (%)
Exclusion from relationships with children	17.8	14.7
Exclusion from social participation	50.2	50.0
Exclusion from financial products	46.6	51.9
Exclusion from common consumer goods	40.1	42.7
Perceived exclusion	20.4	31.4

Note. *No respondents reported having any other marital status.

Regarding the prevalence of social exclusion among rural males and females, males are slightly more likely to be excluded from relationships with their children than females (17.8% vs. 14.7%), whereas the levels of exclusion from social participation are almost the same for both genders (50.2% vs. 50%). Males are less likely to be excluded from financial products and common consumer goods than females (46.6% compared to 51.9%, and 40.1% compared to 42.7%, respectively). The most significant difference exists in perceived exclusion: 20.4% of males and 31.4% of females perceived themselves as excluded. To summarise, rural males have more financial resources and are less likely to perceive themselves as excluded, with slightly weaker relationships with their children than females.

Table 2 reports the logistic regression results. The results of Model 1 show that rural males with non-agricultural work experience are significantly less likely to be excluded from social participation, financial products, household durable goods and report perceived exclusion than those who have only engaged in agricultural work over their life course (ORs = 0.72, 0.75, 0.69 and 0.79, respectively). However, no significant associations between work history and the five domains of social exclusion were found among rural females.

Table 2:

Logistic regression model results for five domains of social exclusion experienced by older rural residents.

		Males					Females
	Key Explanatory Variable	Exclusion from relationships with children	Exclusion from social participation	Exclusion from financial products	Exclusion from common consumer goods	Perceived exclusion	Exclusion from relationships with children	Exclusion from social participation	Exclusion from financial products	Exclusion from common consumer goods	Perceived exclusion
Model 1	(Ref. group: Never having carried out non-agricultural work)
	Ever having carried out non-agricultural work	1.01	0.72***	0.75***	0.69***	0.79**	1.02	0.92	0.94	0.90	0.96
	Ever having carried out non-agricultural work	(0.787 – 1.299)	(0.597 – 0.870)	(0.620 – 0.916)	(0.564 – 0.848)	(0.616 – 1.009)	(0.670 – 1.544)	(0.688 – 1.234)	(0.700 – 1.252)	(0.660 – 1.230)	(0.695 – 1.338)
Model 2	(Ref. group: Lived in the same rural village over their life course)
	Have moved between villages	1.14	0.83	0.99	1.15	1.37*	0.86	0.73***	0.98	0.87	1.00
	Have moved between villages	(0.803 – 1.609)	(0.638 – 1.085)	(0.757 – 1.303)	(0.865 – 1.517)	(0.991 – 1.890)	(0.623 – 1.173)	(0.586 – 0.903)	(0.787 – 1.208)	(0.692 – 1.086)	(0.793 – 1.264)
	Had lived in urban areas and have returned to rural areas	0.97	0.73*	0.79	0.57***	1.12	1.48	0.84	0.74	1.34	1.75**
	Had lived in urban areas and have returned to rural areas	(0.607 – 1.549)	(0.505 – 1.049)	(0.542 – 1.156)	(0.375 – 0.863)	(0.714 – 1.767)	(0.684 – 3.203)	(0.444 – 1.591)	(0.391 – 1.395)	(0.690 – 2.620)	(0.899 – 3.388)
Model 3	(Ref. group: Never having carried out non-agricultural work)
	Ever having carried out non-agricultural work	1.00	0.74***	0.76***	0.71***	0.75**	0.99	0.94	0.95	0.89	0.93
	Ever having carried out non-agricultural work	(0.773 – 1.306)	(0.611 – 0.902)	(0.623 – 0.931)	(0.578 – 0.881)	(0.578 – 0.964)	(0.652 – 1.518)	(0.700 – 1.263)	(0.710 – 1.276)	(0.651 – 1.218)	(0.665 – 1.292)
	(Ref. group: Lived in the same village)
	Have moved between villages	1.14	0.87	1.04	1.21	1.43**	0.86	0.73***	0.98	0.87	1.00
	Have moved between villages	(0.798 – 1.615)	(0.667 – 1.139)	(0.789 – 1.364)	(0.912 – 1.608)	(1.036 – 1.988)	(0.623 – 1.174)	(0.587 – 0.904)	(0.788 – 1.209)	(0.693 – 1.089)	(0.794 – 1.266)
	Lived in urban areas and returned to rural areas	0.97	0.83	0.90	0.66*	1.28	1.48	0.86	0.75	1.38	1.78**
	Lived in urban areas and returned to rural areas	(0.595 – 1.574)	(0.572 – 1.215)	(0.607 – 1.326)	(0.433 – 1.018)	(0.803 – 2.052)	(0.679 – 3.235)	(0.450 – 1.625)	(0.394 – 1.420)	(0.707 – 2.701)	(0.913 – 3.487)
Model 4	(Ref. Never having carried out non-agricultural work and Lived in the same rural village over their life course)
	Never having carried out non-agricultural work and Have moved between villages	0.99	0.93	1.04	0.95	1.16	0.91	0.70***	0.97	0.90	1.02
		(0.596 – 1.655)	(0.650 – 1.330)	(0.725 – 1.505)	(0.648 – 1.384)	(0.750 – 1.785)	(0.652 – 1.268)	(0.558 – 0.883)	(0.774 – 1.219)	(0.711 – 1.143)	(0.798 – 1.305)
	Never having carried out non-agricultural work and Lived in urban areas and returned to rural areas	1.33	0.70	3.50**	0.32**	0.67	1.83	0.71	0.48*	0.81	2.16*
		(0.462 – 3.800)	(0.317 – 1.562)	(1.348 – 9.079)	(0.120 – 0.832)	(0.239 – 1.879)	(0.721 – 4.658)	(0.314 – 1.626)	(0.206 – 1.120)	(0.349 – 1.894)	(0.931 – 5.014)
	Ever having carried out non-agricultural work and Lived in the same rural village over their life course	0.98	0.75**	0.83*	0.62***	0.64***	1.18	0.85	0.89	0.90	1.00
		(0.730 – 1.315)	(0.605 – 0.935)	(0.660 – 1.036)	(0.489 – 0.788)	(0.474 – 0.863)	(0.731 – 1.906)	(0.599 – 1.195)	(0.630 – 1.249)	(0.623 – 1.290)	(0.679 – 1.470)
	Ever having carried out non-agricultural work and Have moved between villages	1.25	0.61***	0.83	1.03	1.25	0.60	0.83	0.92	0.58	0.85
		(0.794 – 1.975)	(0.419 – 0.883)	(0.566 – 1.212)	(0.701 – 1.523)	(0.801 – 1.950)	(0.241 – 1.471)	(0.478 – 1.433)	(0.536 – 1.591)	(0.305 – 1.087)	(0.449 – 1.603)
	Ever having carried out non-agricultural work and returned from urban areas	0.90	0.65**	0.51***	0.54***	1.08	1.03	1.04	1.39	3.35**	1.25
		(0.535 – 1.523)	(0.429 – 0.980)	(0.327 – 0.793)	(0.340 – 0.859)	(0.649 – 1.782)	(0.265 – 3.998)	(0.374 – 2.902)	(0.485 – 3.984)	(1.010 – 11.147)	(0.429 – 3.630)

Note. *** p<0.001, ** p<0.01, * p<0.05. All models control for age group, education, ethnicity, household expenditure, whether the respondents have any ADL limitations, their number of surviving children and their province of origin. Results for control variables are not shown.

The results of Model 2 show that rural males who have ever lived in urban areas are less likely to experience exclusion from social participation and common consumer goods than respondents who have always lived in the same village or moved within rural areas (ORs = 0.73 and 0.57, respectively). Rural males who have moved within rural areas are more likely to perceive themselves as excluded than those who have lived in the same village and who have returned there from urban areas (OR = 1.37). For rural females, respondents who have moved within rural areas are less likely to be excluded from social participation than respondents who have always lived in the same village (OR = 0.73); and respondents who have lived in urban areas experience the highest level of perceived exclusion compared with those who have always lived in the same village (OR = 1.75).

The results of the impact of work and residence history in Model 3 are similar to those in Models 1 and 2, whilst the impact of residence history on social participation was insignificant for males. This indicates that work and residence history have different impacts on social exclusion and that these impacts are not linked.

Model 4 explores the interaction between work and residence history on social exclusion (respondents who only have agricultural work experience and have always lived in the same village are the reference group). For rural men, as compared to the reference group, having non-agricultural work experience is significantly related to lower levels of exclusion from social participation regardless different living history (ORs = 0.75, 0.61 and 0.65, respectively). Regarding exclusion from financial products, males who engaged in agricultural work and returned from urban areas were more likely to be excluded than the reference group (OR = 3.5); whilst, respondents who engaged in non-agricultural work and always lived in the same village, and those who engaged in non-agricultural work and returned from urban areas were less likely to be excluded than the reference group (ORs = 0.83 and 0.51, respectively). Respondents who engaged in agricultural work and returned from urban areas, who engaged in non-agricultural work and always lived in the same village, and those who engaged in non-agricultural work and returned from urban areas were less likely to be excluded from common consumer goods than the reference group (ORs = 0.32, 0.62 and 0.54, respectively). Respondents who engaged in non-agricultural work and always lived in the same village reported lower levels of perceived social exclusion (OR = 0.64).

Among rural females, respondents who engaged in agricultural work and moved within rural areas had a lower level of exclusion from social participation than the reference group (OR = 0.70). Respondents who engaged in agricultural work but who moved within rural areas were less likely to be excluded from financial products and common consumer goods than the reference group (ORs = 0.48 and 0.58, respectively); whilst respondents who engaged in non-agricultural work and returned from urban areas were more likely to be excluded from common consumer goods (OR = 3.35). For perceived exclusion, respondents who did not engage in agricultural work and returned from urban areas experienced higher level of perceived exclusion than the reference group (OR = 2.16).

Discussion and Conclusion

This paper investigates the prevalence of social exclusion among rural residents, and using a life course perspective, explores how return migrants’ earlier work history and residence history could impact on their social exclusion in later life. From the descriptive analyses, rural females were in a disadvantaged position regarding social exclusion. Males had more financial resources (financial products and common consumer goods) and lower levels of subjective feelings of exclusion than females; females had slightly improved circumstances in terms of their relationships with their children than males.

Work and residence history of migrant workers had a more positive impact for male return migrants than for females. Firstly, non-agricultural work experiences were related to lower levels of exclusion from social participation for males but not for females. This means that the social networks of return migrants do not deteriorate, as migrant workers are able to keep in contact with local people when they were living away from their hometowns. As some studies have shown, connections with people from the same village are the most important social network members for migrant workers (Li et al., 2008; Zhao & Jin, 2019), which may explain the positive impact of non-agricultural work experience on social attachment with other villagers. In contrast, female’s social participation in later life is shaped by residence history rather than work history: females who moved between villages and worked in non-agricultural roles had the lowest levels of exclusion from social participation, such as married women who moved from their hometown to their husband’s villages. Their lower levels of exclusion from social participation may result from a combination of experiencing difficulties in establishing a new social network after marriage, and losing contact with family members in their hometown since they had less contact with family members in their original place of residence (Connelly et al., 2010).

The impact of early work and residence trajectories on exclusion from benefiting from financial products and from common consumer goods are different, although both domains reflect the availability of financial resources. For males, non-agricultural work roles were related to a lower level of exclusion from financial products; in contrast, change of residency (either having moved across rural areas or living in urban areas) did not have a significant impact on exclusion from financial products. In addition, both non-agricultural work experience and changes in residence were related to a lower level of exclusion from common consumer goods. These results indicate that the impact of work experience on respondents’ financial well-being is stronger than changes in residence. Migrant workers may begin to participate in formal pension programmes when they participate in non-agricultural work, and the pension benefits can greatly improve their possession of financial products in later life. For females, the impact of work and residence history on exclusion from financial products and exclusion from common consumer goods were not significant. These results indicate that the work position of female migrant workers may result in lower income and fewer opportunities for participation in pension programmes, so that non-agricultural work experiences do not contribute to female’s possession of financial products.

Regarding perceived exclusion, rural males with non-agricultural work experience had lower levels of perceived exclusion; whereas males who moved within rural areas had higher levels of perceived exclusion. These results indicate that the change in residence itself may lead to a higher level of perceived exclusion, as return migrants may lose contact with friends in the villages where they used to live in. However, non-agricultural work experience had a positive effect on perceived exclusion: male return migrants may have more optimistic attitudes due to their experience than local people, since they have been influenced by different values and norms. Rural women who lived in urban areas and returned to rural areas were more likely to report perceived exclusion than those who had always lived in the same village. This may reflect that their status as involuntary return migrants, who have negative feelings about their reluctance returning to their hometown from cities.

Work and residence trajectories do not influence return migrants’ relationships with their children for either males or females. As filial piety is still the norm in modern China (Du, 2013), children of return migrants would still maintain contact with their parents and support them financially and emotionally, even if they may have lived elsewhere as migrant workers (Phillips & Feng, 2015).

This study has its limitations. The length of work experience and residence, and when it happened in people’s lives could influence their social exclusion in later life in various ways. Respondents have different experiences during their lives, and the number of similar experiences is too small to be estimated in the models, which limits further exploration. Secondly, we are not able to differentiate between involuntary or voluntary return migrants.

This paper nevertheless contributes to existing literature in three ways. First, although existing studies have confirmed the impact of a series of work and residence trajectories on critical life domains at older ages (e.g. quality of life, financial status and the timing of retirement (Han & Moen, 1999; Sefton et al., 2011; Wahrendorf, 2015), the majority of literature is about the importance of proximate factors (i.e. the conditions around the occurrence of social exclusion), such as health (Jose & Cherayi, 2017) and social-economic status (Feng et al., 2018; MacLeod et al., 2019; Van Regenmortel et al., 2016; Yuan & Ngaiyum, 2012). This study captures the long-term impact of early experience of work and residence across participants’ life courses focussing on social exclusion in later life. Second, few studies have focused on the later lives of return migrants. This study highlights that their gains from early work experience incorporates both financial and non-financial aspects and should be evaluated comprehensively. Third, the results of this study further contribute to the policy debate regarding improving the circumstances of older rural residents in China. In recent years, the focus of the government is on poverty alleviation programmes aiming for the increase of income and the improvement of living standards. However, our results indicate that the government should focus on promoting social participation and the psychological health of rural residents, which should be incorporated into China’s Active Ageing Agenda proposed in year 2021. This is especially the case for older rural females, as our results suggest that they experience higher levels of social exclusion in general. In addition, it is noteworthy that the group of return migration is heterogeneous. Whilst the authority is willing to encourage and support a small number of ‘return entrepreneurs', the potential needs of the majority of return migrants, who are middle-aged or older, are largely absent from policy intervention. Lastly, as return migrants represent a very large group not only in modern China but also in many countries with differing levels of economic development, the findings of this study also add value to the evaluation of, and policy intervention towards return migrants in a wider context.

Footnotes

Acknowledgements

The authors acknowledge the assistance of the institutes and individuals aforementioned in providing data. We are grateful to Dr. Kathryn Wicks’and Margarida Cheung Vieira’s helpful suggestions on the revisions.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by Sichuan Social Science Foundation Youth Project of China (SC21C070), National Social Science Foundation of China (21ZDA103, 20CSH014) and National Natural Science Foundation of China (42171196). These funders have no role in the research design, the writing of the article, or the decision to submit the article. This paper uses data from the China Health and Retirement Longitudinal Study (CHARLS) (see ). The CHARLS research project was conducted by the National School of Development of Peking University.

Ethics statement

We use secondary data from the China Health and Retirement Longitudinal Study (CHARLS). The CHARLS has received ethical approval from the Peking University Institution Review Board; the participation of respondents in the survey is subject to informed consent.

ORCID iDs

Maodi Xu, PhD

Zhixin Feng, PhD

References

Bai

(2003). Returning to the countryside versus continuing to work in the cities: A study on rural urban migrants and their return to the countryside of China. Social Sciences in China, 24(4), 149–159.

Barrett

Mosca

(2013). Social isolation, loneliness and return migration: evidence from older irish adults. Journal of Ethnic and Migration Studies, 39(10), 1659–1677. https://doi.org/10.1080/1369183X.2013.833694.

Burchardt

Le Grand

Piachaud

(1999). Social exclusion in Britain 1991-1995. Social Policy and Administration, 33(3), 227–244. https://doi.org/10.1111/1467-9515.00148.

Cai

Park

Zhao

(2008). The Chinese labor market in the reform Era. In Brandt

Rawski

T. G.

(Eds.), China’s Great Economic Transformation (pp. 167–214). Cambridge University Press. https://doi.org/10.1017/CBO9780511754234.

Connelly

Roberts

Zheng

(2010). The impact of circular migration on the position of married women in rural China. Feminist Economics, 16(1), 3–41. https://doi.org/10.1080/13545700903382752.

(2013). Intergenerational solidarity and old-age support for the social inclusion of elders in Mainland China: the changing roles of family and government. Ageing and Society, 33(1), 44–63. https://doi.org/10.1017/S0144686X12000773.

Dustmann

Weiss

(2007). Return migration: Theory and empirical evidence from the UK. British Journal of Industrial Relations, 45(2), 236–256. https://doi.org/10.1111/j.1467-8543.2007.00613.x.

Feng

Phillips

D. R.

Jones

(2018). A geographical multivariable multilevel analysis of social exclusion among older people in China: Evidence from the China Longitudinal Aging Social Survey ageing study. Geographical Journal, 184(4), 413–428. https://doi.org/10.1111/geoj.12274.

Han

S. K.

Moen

(1999). Clocking out : Temporal patterning of retirement. American Journal of Sociology, 105(1), 191–236. https://doi.org/10.1086/210271.

10.

Hooper

(1998). Flower vase and housewife: Women and consumerism in post-Mao China. In Sen

Stivens

(Eds.), Gender and Power in Affluent Asia (pp. 167–193). Routledge. https://doi.org/10.1525/ae.2000.27.2.529.

11.

Huang

Guo

(2017). Welfare programme participation and the wellbeing of non-local rural migrants in metropolitan china: A social exclusion perspective. Social Indicators Research, 132(1), 63–85. https://doi.org/10.1007/s11205-016-1329-y.

12.

Jose

J. P.

Cherayi

(2017). Social exclusion of older persons: The role of socio-demographic characteristics. Ageing International, 42(4), 447–465. https://doi.org/10.1007/s12126-016-9255-5.

13.

Kneale

(2012). Is social exclusion still important for older people? In Age UK and International Longevity Centre (Issue September). https://doi.org/10.1529/biophysj.106.098269.

14.

Lee

E. S.

(1966). A theory of migration. Demography, 3(1), 47–57. https://doi.org/doi.org/10.2307/2060063.

15.

Levitas

(2005). The inclusive society? Social exclusion and new labour. Springer.

16.

Ren

Jin

Feldman

M. W.

(2008). [中国农民工的社会融合及其影响因素研究基于社会支持网络的分析.[A research on the social integration of Chinese migrant rural workers and its influencing factors:An analysis based on the social support network. 人口与经济, 2(2), 170–178.

17.

Liu

(2014). Ageing, migration and familial support in rural China. Geoforum, 51(1), 305–312. https://doi.org/10.1016/j.geoforum.2013.04.013.

18.

Liu

(2017). 农村劳动力外出务工经历与现代性获得.[Rural labor force’s experience of going out to work and the acquisition of Modernity]. In Cai

(Ed.), Report on China Labor-Force Dynamic Survey 2017 (pp. 290–312). Social Sciences Academic Press.

19.

Liu

Dou

Perry

Angeles

(2020). Return migration in later life in Mainland China: motivations, patterns, and influences. Ageing International, 45(1), 30–49. https://doi.org/10.1007/s12126-018-9336-8.

20.

MacLeod

C. A.

Ross

Sacker

Netuveli

Windle

(2019). Re-thinking social exclusion in later life: A case for a new framework for measurement. Ageing and Society, 39(1), 74–111. https://doi.org/10.1017/S0144686X17000794.

21.

National Bureau of Statistics . (2015). Chinese Statistics Yearbook. Beijing, China: Chinese Statistics Press.

22.

National Bureau of Statistics (2020). 2019年农民工监测调查报告 [Migrant Workers Monitoring Survey Report 2019]. National Bureau of Statistics. http://www.stats.gov.cn/tjsj/zxfb/201804/t20180427_1596389.html (Accessed 6 August 2021).

23.

National Development and Reform Commission (2017). 农村一二三产业融合发展年度报告 [Annual report of the integration of primary, secondary and tertiary industries in rural areas]. National Development and Reform Commission. http://www.gov.cn/xinwen/2018-04/20/content_5284269.htm#1 (Accessed 6 August 2021).

24.

Nolan

Whelan

C. T.

(1996). Resources, deprivation, and poverty. Oxford University Press.

25.

Phillips

Feng

(2015). Challenges for the aging family in the people’s Republic of China. Canadian Journal on Aging/La Revue Canadienne Du Vieillissement, 34(3), 290–304. https://doi.org/10.1017/S0714980815000203.

26.

Sefton

Evandrou

Falkingham

Vlachantoni

(2011). The relationship between women’s work histories and incomes in later life in the UK, US and West Germany. Journal of European Social Policy, 21(1), 20–36. https://doi.org/10.1177/0958928710380475.

27.

Urbaniak

Walsh

(2019). The interrelationship between place and critical life transitions in later life social exclusion: A scoping review. Health and Place, 60(5), 102234. https://doi.org/10.1016/j.healthplace.2019.102234.

28.

Van Regenmortel

De Donder

Dury

Smetcoren

A.-S.

De Witte

Verté

(2016). Social exclusion in later life: A systematic review of the literature. Journal of Population Ageing, 9(4), 315–344. https://doi.org/10.1007/s12062-016-9145-3.

29.

Wahrendorf

(2015). Previous employment histories and quality of life in older ages: Sequence analyses using sharelife. Ageing and Society, 35(09), 1928–1959. https://doi.org/10.1017/S0144686X14000713.

30.

Walsh

Scharf

Keating

(2017). Social exclusion of older persons: A scoping review and conceptual framework. European Journal of Ageing, 14(1), 81–98. https://doi.org/10.1007/s10433-016-0398-8.

31.

Wang

W. W.

Fan

C. C.

(2006). Success or failure: Selectivity and reasons of return migration in Sichuan and Anhui, China. Environment and Planning A, 38(5), 939–958. https://doi.org/10.1068/a37428.

32.

Xie

Zhou

(2017). 农民工的返乡行为及其就业分化研究[The returning and employment differentiation of migrant workers]. 农业经济问题, 2(2), 92–112. https://doi.org/10.13246/j.cnki.iae.2017.02.012.

33.

Liu

(2017). Return migration in China: A case study of Zhumadian in Henan province. Eurasian Geography and Economics, 58(1), 114–140. https://doi.org/10.1080/15387216.2017.1280689.

34.

Yuan

Ngaiyum

(2012). Social exclusion and neighborhood support: A case study of empty-nest elderly in urban Shanghai. Journal of Gerontological Social Work, 55(7), 587–608. https://doi.org/10.1080/01634372.2012.676613.

35.

Zhang

(2011). 生命历程视角下的返乡农民工研究: 以湖南省沅江镇的返乡农民工为表述对象[The study of farmer workers under the perspective of life course theory]. Doctoral dissertation. Huazhong University of Science and Technology.

36.

Zhao

Jin

(2019). Migrant workers in Beijing: How hometown ties affect economic outcomes. Work, Employment and Society, 34(5), 789–808. https://doi.org/10.1177/0950017019870754.

37.

Zhao

(2002). Causes and consequences of return migration: Recent evidence from China. Journal of Comparitive Economics, 30(2), 376–394. https://doi.org/10.1006/jcec.2002.1781.

38.

Zhou

Tan

(2017). 外出务工经历有利于农民工返乡创业吗[Does Migration Experience Promote Entrepreneurship in Rural China]. China Economic Quartley, 16(2), 793–814. https://doi.org/10.13821/j.cnki.ceq.2017.01.15.