Generalized trust among rural-to-urban migrants in China: Role of relative deprivation and neighborhood context

Abstract

China has experienced sizeable internal migration in the past several decades. This research examines the implications of internal migration on generalized trust in China. Using data from the 2014 China Labor-force Dynamics Survey (CLDS), a nationally representative survey, we compared the level of self-rated generalized trust in strangers among Chinese rural-to-urban migrants, rural nonmigrants, and urban nonmigrants. We further assessed the mediating effects of relative deprivation and neighborhood context in the relationship between migration status and generalized trust. The results show that migrants are less trusting than rural and urban nonmigrants after adjusting for relevant sociodemographic characteristics. The mediation analysis suggests that relative deprivation partially explains the difference in generalized trust between migrants and rural nonmigrants, and it greatly mediates the trust gap between migrants and urban nonmigrants. The neighborhood context largely explains the difference in generalized trust between migrants and rural nonmigrants. Moreover, urban nonmigrants are less trusting than migrants when taking neighborhood context into consideration, which indicates a suppression effect of neighborhood context on the relationship between migrant status and social trust in urban China. This study contributes to a deeper understanding of how social trust changes in the process of internal migration. Policy implications are discussed.

Introduction

Generalized trust, which refers to one’s disposition to place trust in strangers or general others (Yamagishi & Yamagishi, 1994), is a key underpinning for effective social functioning and social capital (Fukuyama, 1995, Putnam, 1993), and thereby has attracted a great deal of attention from both policymakers and scholars. As populations in most regions worldwide are becoming increasingly diverse caused by immigration and other social forces, it has become a critical question as to how immigration and population mobility affects generalized trust (Fukuyama, 1995, Putnam, 1993). While most research investigates the relationship between diversity (typically measured in terms of the racial and ethnic composition of the population) and social cohesion, only a few studies have assessed what determines the generalized trust of immigrants (typically measured based on place of birth and/or nationality). Moreover, research on how internal migration affects generalized trust in regions outside Western societies is scarce. For instance, in the past three decades, China has witnessed an unprecedented influx of labor migrants relocating from rural to urban areas. How does internal migration affect trust, and what factors explain the effects? Using a national representative survey, this study is among the first attempts to examine the impact of migration on social trust and investigate the factors that may account for the potential differences between migrants and their urban and rural counterparts.

While both Western and Chinese societies share similar sentiments that social change resulting from migration may disturb social cohesion and trust, there are multiple differences between the two types of migration. First, Chinese migrants are more powerfully and systematically influenced by institutional arrangements and constraints (i.e., the hukou system), while social segregation and exclusions in Western societies are largely based on race/ethnicity. Much of the discrimination and prejudices towards minority groups in the western regions are activated by immediate reactions to physical appearances (e.g., whites vs. blacks), while in the context of China, such sentiments come from socially constructed differences. In this sense, the research can be suggestive for studies on discrimination experienced among social groups similar in physical appearances but differ in socially constructed categories (e.g., blacks vs. Hispanics).

Second, while a majority of studies on the relationship between immigration and generalized trust focus on migrants from low trust (less economically developed) to the high trust (more economically developed) areas, Chinese labor migrants are characterized by relocating from high trust societies of acquaintances to relatively low trust city areas. Using large-scale national surveys in China, previous studies found that rural residents reported a higher level of generalized trust than their urban counterparts (Jiang et al., 2019; Yang & Zeng, 2016). It may be because rural life in China is oriented towards traditionalism, featuring high levels of interpersonal trust based on local networks and reputation systems. Urban China, by contrast, has largely transited to modernity in which people are largely anonymous in social interactions and therefore trust others less. Moreover, political trust or perception of local governance may affect generalized social trust in China (Steinhardt, 2012). Tao et al. (2014) ’s research confirmed a causal relationship between political and social trust in rural China, with political trust positively influencing social trust. A study of over 5,000 participants in China found that urban residents have a lower level of trust in government (both central and local government). Rural residents may have a higher level of social trust due to their higher political/institutional trust. This study provides an opportunity to look at how rural-to-urban migrants change their levels of social trust in adapting to urban culture and life as compared to their counterparts of rural and urban local residents. It also has general implications for studying migration from high trust to low trust areas.

There is substantial evidence that immigrants tend to have less trust in generalized others than native-born people, such as in European (Bjørnskov, 2008, De Vroome et al., 2013) and North American societies (Chávez et al., 2006, Wilkes and Wu, 2019). It is less clear, however, what may account for the trust gap between immigrants and non-immigrants. Past studies have highlighted the importance of both individual factors and neighborhood context when forming trust. One factor that has been cited as “arguably the strongest determinant of the level of social trust” (Bjørnskov, 2008:271) is socioeconomic inequality. It has been well-established that higher social class is associated with higher generalized trust (Qiang et al., 2021). Recent studies have further explored the role of subjective inequality and perceived fairness, which may capture the psychological components of socioeconomic status (e.g., relative deprivation, social comparison) in influencing generalized trust (Dinesen, 2012b, Dinesen, 2013, Hu, 2017, Nannestad et al., 2014).

In addition to individual characteristics, neighborhood contexts are also found to significantly predict generalized trust (Intravia et al., 2016). Neighborhoods and communities are where social interactions frequently take place and thus may shape one’s general psychological orientations. Previous research has indicated that social and economic characteristics of one’s neighborhood, such as demographic composition, neighborhood social order, and interactions with neighbors, will influence one’s decisions of whether to place trust in a stranger (Marschall and Stolle, 2004, Stolle et al., 2008). Informed by past studies, this study sets out to examine whether relative deprivation and neighborhood context may account for the gap in generalized trust between rural-to-urban migrants and rural and urban nonmigrants.

The current research aims to answer the following questions. First, how does China’s internal labor migration affect migrants’ social trust? To answer that question, we compare the self-reported level of generalized trust among three critical groups: rural-to-urban migrants, rural nonmigrant residents, and urban nonmigrant residents. Furthermore, whether relative deprivation and neighborhood context may explain the levels of social trust among the emerging social groups during the migration process? Below we first discuss general literature on migration and generalized trust, as well as the effect of relative deprivation and neighborhood context on generalized trust. Then we explain the Chinese context, focusing on the disadvantaged life conditions of rural-to-urban migrants. Next, we explain the methodology, followed by data analysis and results. Last, we discuss the implications of the study.

Generalized trust refers to one’s disposition to place trust in strangers or general others (Yamagishi & Yamagishi, 1994). Researchers have distinguished between generalized trust and particularized trust; the latter correspondents to trust extended to known others based on their reputation or trustworthiness (Bjørnskov, 2007). Researchers have paid special attention to generalized trust as it critically affects both individuals’ and society’s well-being. On the individual level, generalized trust correlates with a number of positive outcomes, such as self-rated health (Giordano et al., 2012) and life satisfaction (Bjørnskov, 2003, Uslaner, 2002). More importantly, on the societal level, the generalized trust serves as the backbone for social functioning and social capital (Putnam, 1995, 2007). Numerous studies have documented that generalized trust plays a positive role in economic prosperity (Beugelsdijk et al., 2004, Bjørnskov, 2012) and institutional performance (Boix and Posner, 1998, Knack, 2000). In fact, generalized trust has long been treated as the primary indicator for social capital, which refers to “features of social organization, such as trust, norms, and networks that can improve the efficiency of society by facilitating coordinated actions” (Putnam, 1993, p. 167). In social surveys, generalized trust is typically measured with the question, “Do you think people are generally trustworthy?” Though controversy exists regarding whether or not this question objectively measures generalized trust (Simpson et al., 2007), we will still use this measure in our analysis, and we address potential issues with using this measure in the method section below.

In explaining the roots of generalized trust, researchers have recognized both cultural and institutional/experiential explanations (Dinesen, 2012b). The cultural view argues that generalized trust is largely determined by cultural transmission from one generation to the next (Uslaner, 2002, Uslaner, 2008). Accordingly, generalized trust is a stable trait immune to new environments and experiences. Most researchers, however, insisted that generalized trust is an outcome of socialization and experiences (Dinesen, 2012a, Nannestad et al., 2014). They emphasize the critical role of everyday as well as institutional environments in shaping one’s generalized trust. In particular, previous literature has shown that the level of generalized trust can be subject to a number of experiences and environmental factors, such as one’s relative deprivation experiences and neighborhood contexts.

Empirically, researchers have treated immigration as a natural experiment contrasting the cultural vs. experiential views on generalized trust (Dinesen, 2012a). By comparing the level of trust of migrants to both residents in rural and urban areas also provides an effective way to examine two broad perspectives regarding the causes of individuals’ generalized trust, that is, cultural perspective (trust as a persistent cultural trait) and experiential perspective (trust as a response to the immediate environment and personal experiences) (Dinesen & Bekkers, 2017). Trust serves as a cultural persistence if migrants’ social trust continues to reflect the culture of trust in their places of origin. On the contrary, if migrants show similar trust levels to urban natives, then trust is largely shaped by environment and experience.

However, previous studies on the relationship between immigration and trust have been far from conclusive. Some studies found that trust among recent immigrants or descendants of immigrants is rooted in the cultural footprints of their ancestors and does not change with their new experiences in the receiving society (e.g., Uslaner, 2008; Bilodeau & White, 2016; Moschion & Tabasso, 2014). Uslaner (2008) compares the level of trust among Americans from different ethnic backgrounds and found that Americans whose grandparents came from Nordic countries, Britain, and Germany tend to report higher levels of trust. Conversely, Italians, Latinos, and African Americans tend to be less trusting. Considering that Western European countries have higher levels of trust compared with countries in Latin America, Africa, and other parts of Europe, the author concluded that Americans’ trust is influenced more by their ethnic heritage than by their new experiences in the country. Also, trust among immigrants in Canada is strongly correlated with trust in their countries of origin (Bilodeau & White, 2016). Similarly, a positive relationship exists between the trust of immigrants in Australia and the trust in their home country (Moschion & Tabasso, 2014). These studies suggest that immigrants tend to retain the cultural fingerprints of their ancestral roots, and thus trust is culturally determined.

On the other hand, some scholars find stronger evidence for the experiential theory of trust, which contends that immigrants’ trust is subject to change in response to their new social experiences (Dinesen and Hooghe, 2010, Dinesen, 2012a, Dinesen, 2012b, Nannestad et al., 2014, Dinesen and Sønderskov, 2018). Dinesen and Hooghe (2010) found that non-Western immigrants tend to adjust to the level of trust of native residents in Western European countries after controlling for the level of trust of immigrants’ country of origin. As for Danish society, Nannestad et al. (2014) found that it is immigrants’ socioeconomic status and their trust in Danish institutions that affect their trust in others rather than immigrants’ cultural proposition. Using data from European Social Survey (ESS), Dinesen, 2012a, Dinesen, 2012b suggested that Turks, Poles, and Italians in Northern European countries display significantly higher levels of trust than their counterparts in their countries of origin. In these studies, trust is found to be malleable in response to new experiences in different social contexts.

While most studies on the relationship between migration and trust focus on international migration, some recent studies have examined the trust of internal migrants compared to natives in the destination. Wu’s (2021) analysis of internal migrants in Canada showed support for the cultural theory that an individual’s trust in others is stabilized early in life. He found that Quebecers (typically less trusting) that emigrated to live in other parts of the country continue to show a lower level of trust, whereas English Canadians who migrated to live in Quebec remain more trusting than local natives. Wu’s (2020) research on internal migrants’ trust in the U.S. drives a similar conclusion, which argues that irrespective of where they move after age 16, Americans’ trust changes very little. The overall results lend significant support for the cultural socialization theory of trust. That is, people learn to trust early in life from cultural heritage and socialization, and their learned trust does not respond to new experiences and changing circumstances.

Despite that most studies on the relationship between migration and trust focus on international migration, some recent studies have compared the level of trust between internal migrants with natives in the destination. Wu’s (2021) analysis of internal migrants in Canada supported the cultural theory that trust is formed early in life. Specifically, Quebecers (generally less trusting) who migrated to other parts of the country continue to show a lower level of trust, while English Canadians who migrated to Quebec remain more trusting than natives. Wu’s (2020) research on internal migrants’ trust in the U.S. leads to a similar conclusion, arguing that Americans’ trust changes very little, regardless of where people move after the age 16. These findings seem to support the cultural socialization theory of trust, which holds that trust is determined by cultural heritage and early socialization, not by personal experiences, particularly in adulthood.

While most previous research only focuses on the comparison between immigrants and the native population in the destination, studies that adopt a double comparative research design, with both origin and destination countries, will be more suited to distinguish origin and destination effects and will also provide a better examination of the persistence or experience hypotheses (De Vroome et al., 2013). To that end, this study provides an opportunity to further examine the roots of generalized trust in the context of Chinese internal migration.

Formally conceptualized by Runciman (1966), relative deprivation refers to the discontent people feel when they compare their positions to others and realize that others in the group possess something they do not have. People’s satisfaction with their rewards is not influenced by the absolute value of their actual rewards, but rather by the relative value of their social or historical comparisons. The experience of feeling relatively deprived may involve subjective cognition and affective experience that trigger negative emotions such as anger and frustration (Smith et al., 2012; Xiong and Ye, 2016).

Differences in generalized trust can be attributed to perceived relative deprivation and relevant social comparison processes. For example, a survey of over 24,000 individuals in 25 nations found that those with lower relative positions were less trusting than those that were comparatively better off (Fischer & Torgler, 2006). A sociopsychological explanation of the role of relative deprivation in social trust argues that people who feel relatively deprived may widen the psychological distance to and distrust many social members (Hu, 2017). Individuals with lower relative positions may be envious of those with higher positions and feel frustrated that they are unable to attain a similar status and consequently, be less trusting (Fischer & Torgler, 2006). Perceiving that others are enjoying great wealth and power but not themselves, they tend to believe that others are somehow cheating the system and thus that people generally cannot be trusted (Fairbrother & Martin, 2013).

Furthermore, people not satisfied with their socioeconomic position are more likely to question institutional fairness. Unfair institutions may undermine generalized trust by providing fewer incentives for trustworthy behavior, making trust a risky enterprise (Kumlin and Rothstein, 2007, Rothstein and Stolle, 2008). People who have experienced institutional discrimination may expect that institutions will favor the native population over ethnic/immigrant groups. Consequently, they may lose faith in others and expect others to take advantage of the system to their misfortune. Using data obtained in a large-scale community survey in England and Wales, Letki (2008) found that when the degree of economic deprivation was taken into account, the negative effects of ethnic diversity on social interactions disappeared.

Over the past decades, research on internal migration in China has embraced the concept of relative deprivation (Jin, 2016). The perception of relative deprivation is dependent on reference groups (Pettigrew, 2002). Migration changes people’s social context so that their reference groups and sense of relative deprivation may also change. Due to labor market segregation, the wages of rural-to-urban migrants in China are lower than what urban residents earn. Rural migrants have to work extremely long hours to compensate for their low wages (Frijters et al., 2010). Thus, the hourly wages of rural migrants are even lower than what urban residents earn after taking working time into consideration (Wu & Zhang, 2015). When rural-to-urban migrants adopt an urban frame of reference and compare their socioeconomic conditions to those of urban residents, they may perceive lower status than urban residents and generate a feeling of unfairness and injustice (Jin, 2016, Knight and Gunatilaka, 2010). They may also place themselves lower on the social hierarchy than their rural counterparts because of institutional barriers and social discrimination in cities. The relatively low perceived status may be the reason that migrants report lower happiness and more emotional problems affecting role functioning than both rural and urban nonmigrants (Jin, 2016, Knight and Gunatilaka, 2010). To our best knowledge, no studies have examined how relative deprivation may affect generalized trust among rural-to-urban migrants in China.

Neighborhoods and communities are where social interactions frequently take place and thus may shape one’s general psychological orientations. Previous research has indicated that social and economic characteristics of one’s neighborhood, such as demographic composition, neighborhood social order, and interactions with neighbors, will influence one’s decisions of whether to place trust in a stranger (Marschall and Stolle, 2004, Stolle et al., 2008).

Previous studies have shown that the frequency of neighborhood social interactions will be positively associated with individuals’ propensity to trust (Bledsoe et al., 1995, Putnam, 1995). Moreover, social ties, no matter strong or weak, not only provide resources for migrants to successfully settle in receiving cities but play important roles in affecting migrants’ psychological well-being, such as generalized trust. When the degree and quality of neighborhood contacts are low, diversity may primarily have a detrimental effect on generalized trust, yet when they are high, diversity may have little or even positive effects.

Community-level analysis of generalized trust has also indicated that perceived neighborhood safety, an indicator of social disorganization, is negatively associated with the development of generalized trust (Intravia et al., 2016). Residents perceiving low levels of neighborhood safety tend to spend more time indoors, distrust their neighbors, and regard them as threats rather than as sources of support or assistance (Massey, 1993). Residents may increasingly stay indoors and withdraw from interpersonal interactions. In particular, the effect of perceived neighborhood disorder is stronger among those who are already socially and economically disadvantaged. Ross and colleagues (2001), for instance, found that residents with disadvantaged socioeconomic backgrounds report low levels of social trust in the presence of high levels of neighborhood disorder. They were more likely to feel a sense of powerlessness and threat, which indirectly intensified the negative effects of social disorder.

Over the past three decades, China’s economic reform has witnessed an unprecedented flow of labor migrants relocating from relatively underdeveloped, rural areas to more developed, urban regions in search of better job opportunities and life prospects. By 2018, the total volume of rural-to-urban migrant labor was estimated to be about 288.4 million (National Bureau of Statistics of China, 2019). The massive rural-to-urban migration has brought substantial economic benefits, but it also involves high social costs, including inequality, injustice, and social conflicts (Lu & Wang, 2013). In particular, rural-to-urban migrants in China are plagued by strong institutional barriers to movement due to the long-standing household registration (hukou) system. Since established in the 1950 s, the hukou system assigns an agricultural (rural) or nonagricultural (urban) residency status to individuals at birth. It entitles considerable rights and benefits to the urban hukou registration population. Although migrant workers are permitted to work and stay in cities, they can hardly change their official status from “rural” to “urban” under the current hukou system (Ma & Xiang, 1998). Without a local hukou, rural migrants are not entitled to most of the social welfare benefits enjoyed by urban residents, such as health care, education, and housing subsidy. As a result of that system, rural migrants are treated as second-class citizens who have limited access to subsidized housing, education, health care, and many other social services in urban China (Sun et al., 2015, Wu and Zhang, 2015). Rural-to-urban residents thus face greater barriers and institutional discrimination than what integration migrants would normally face in areas other than China (Chan, 2010, Chan and Zhang, 1999). The considerable personal and institutional discriminations experienced by rural-to-urban migrants in the cities greatly hinder their urban adaptation (Yuan et al., 2013) and may lower their trust in strangers.

Besides, rural-to-urban migrants are usually segregated into disadvantaged neighborhoods (Zhong & Broadhurst, 2007). Rural migrants without a local urban hukou are not entitled to receive social welfare in cities and have limited access to government-subsidized affordable housing, including government-sponsored low-rent housing (Wang, 2000, Wu, 2006). Moreover, due to low educational attainment and labor market segmentation, most rural migrants are concentrated in low-skilled and low-paying jobs and cannot afford commodity housing (Fan, 2002). They had little choice but to congregate in the low-end of the rental market – urbanized villages– characterized by crowded living spaces, a concentration of migrant population, poor hygiene conditions, short of basic public facilities, and lack of privacy and regulation (Liu et al., 2010).

Internal migration may simultaneously lead to a loss of social support networks. Due to the up-rooting process of migration, migrants risk losing their old social ties when they move to urban areas with a distance from their original family and friends in their rural hometowns (Zhong et al., 2017). There are also great difficulties for rural migrants to develop strong social ties with urban residents. It is not only because the market and residential segregation limit the opportunities for migrants to make friends with local people but also because of the pervasive social discrimination that stereotypes migrants as being poor, uncivilized and prone to crime (Lu & Ruan, 2013). Thus, it is reasonable to predict neighborhood context as a crucial factor that accounts for gaps in generalized trust between migrants and their urban local counterparts. We may expect that the social divisions and cleavages created in the process of Chinese internal migration will also largely erode rather than promote the generalized social trust of migrants.

The sharp rural-urban contrast is not the only cultural divide in China. The North and South are starkly different in terms of climatic conditions and geographical environment. Furthermore, the different types of major food crops in the North and South may affect social norms and behavior (Ren et al., 2021, Talhelm and English, 2020), which may in turn affect general trust in others. Some studies have shown that China’s historically rice-farming areas (South China) had tighter social norms (Talhelm & English, 2020) and have a higher level of collectivism (Ma et al., 2016) than wheat-farming areas (North China), after accounting for economic development and urbanization. However, higher collectivism and tighter social norms are not necessarily associated with higher coordination and social trust. For example, a study found that people from the rice-farming region of China were more likely than people from the wheat region to suspect that other people were secretly trying to undermine them (Liu et al., 2019).

Chinese migration research has moved beyond dichotomizing rural-urban migration to consider more complex the group relations process associated with crossing regional socioecological cultures (such as the North-South and rice-wheat divides). Ecological factors may affect how individuals cope with the acculturative stress of moving to a new environment. College students from South China (rice region) moving to a university in North China (wheat region) would use different strategies than students from the wheat-farming region moving to a university in the rice region (English & Geeraert, 2020). We are unaware of any studies that have explored how rural-to-urban migrants may adjust their trust in generalized others depending on the region they move to. As residents of the rice region are less trusting than residents of the wheat region in China, we may expect a decrease in generalized trust for rural-to-urban migrants moving from wheat regions to rice regions. In this study, we examined four patterns of migration (from wheat to wheat region, from wheat to rice region, from rice to wheat region, from rice to rice region) of rural-to-urban migrants.