- Email: [email protected]
Bringing urban governance back in: Neighborhood conflicts and depression
Social Science & Medicine 196 (2018) 1–9
Contents lists available at ScienceDirect
Social Science & Medicine journal homepage: www.elsevier.com/locate/socscimed
Bringing urban governance back in: Neighborhood conflicts and depression
MARK
Qiang Fu Department of Sociology, The University of British Columbia, V6T 1Z1, Vancouver, BC, Canada
A R T I C L E I N F O
A B S T R A C T
Keywords: Depression Neighborhood conflicts Urban transformation Community building Relations Urban governance Contentious politics China
Urban governance and its impact on contentious politics have received remarkably little attention in existing studies on mental health. Drawing on a measure of neighborhood conflicts developed in a survey of thirty-nine urban neighborhoods in Guangzhou, China, this article investigates the potential link between urban governance and mental health. Net of sociodemographic, relational, and environmental measures, it finds that among residents' conflicts with different entities of urban governance, only those with local/grassroots governments are significantly associated with more depressive symptoms. Moreover, these subgroups of government-oriented conflicts associated with more depressive symptoms are related to neighborhood planning and communal properties, reflecting a dilemma in the Chinese model of urban governance. By offering a relational interpretation of neighborhood conflicts, this study not only challenges the previous view that community building in China improves mental health, but calls attention to the significance of urban governance in research on mental health.
1. Introduction Mental illness in the city has long been an intriguing topic for social scientists. Simmel ([1903] 1964) deemed “profusion of sensory stimuli,” including sights, sounds, smells, and interference, as the city's most profound effects. Individuals must adapt, “react with their heads instead of hearts,” and become aloof when dealing with urban life. Until recently, however, the issue of urban governance as a key mechanism for the restructuring of urban space (Brenner, 2004; Jessop, 2002; Pierre, 1999; Wu, 2002) has largely been neglected by health scholars. Brenner (2004: 455) describes urban governance as “the broad constellation of social, political and economic forces that mold the process of urban development within modern capitalism.” While its scope is not necessarily restricted to modern capitalism, urban governance clearly involves a dynamic interplay of state authorities (e.g., local officials and urban planners), market forces (e.g., real estate developers and property managers), and social entities (e.g., residents and territorial civic organizations). Yet, the few epidemiological studies that have considered urban governance (Burris et al., 2007; Muennig, 2014; Shen, 2014) only treat it as being prescribed a priori by policy makers, and focus on the existing repertoire of strategies (e.g., ways of resource mobilization and regulatory measures) used by local authorities to promote healthier urban life. The relational nature of urban governance and its ramification on health have not been sufficiently theorized. This study considers neighborhood conflicts as another form of stressors for depressive symptoms. From a relational perspective,
neighborhood conflicts are conceptualized as contentious expressions of dyadic interplay between urban residents and other entities of urban governance (e.g., property managers, local authorities and real estate developers). It should be noted that the neighborhood conflicts analyzed here are different from the established measures of neighborhood disorder/disadvantage from the poverty literature, such as crime, the concentration of single-parent families, vandalism, and graffiti (Browning and Cagney, 2002; Sampson et al., 1999), nor do they include domestic violence within a family, daily complaints between neighbors, or routine strife among residents. By emphasizing a relational definition of neighborhood conflicts, this study argues that urban governance is not merely a set of rules made and enforced by these who are capable of imposing their will on the will of others. Instead, urban governance is experienced, realized, and made by interconnected entities, each with its own interests in urban space. The presence of neighborhood conflicts implies that existing rules of urban governance (material or abstract, traditional or transitional, moral or political, important or trivial, broad or narrow) are being contested, resisted or rejected by participants in urban governance, which may have ramifications for mental health. Today we see the process of urban transformation most vigorously in China, where cities are undergoing fundamental changes in less than a generation (Hsing, 2010). Owing to its sheer scale, China's great urban transformation can have a far-reaching impact on the global burden of mental disorders (Gong et al., 2012; Hsieh, 2015; Jin et al., 2012; L. Song, 2015). Given that the lower rates of depression reported
E-mail address: [email protected]. http://dx.doi.org/10.1016/j.socscimed.2017.10.035 Received 5 March 2017; Received in revised form 27 September 2017; Accepted 31 October 2017 0277-9536/ © 2017 Elsevier Ltd. All rights reserved.
Social Science & Medicine 196 (2018) 1–9
Q. Fu
Tomba, 2014; Vogel, 1989; Wu, 2002). This goal means that the state not only determines the landscape and planning of the city but also makes rules that govern the everyday life of urban residents (Friedmann, 2007). As noted by Wu (2001), state control over urban space has been weakened in the reform era: with the rise of the private sector, market transformation has created resources and personnel beyond the direct reach of state authorities. Given that decision-making power has been decentralized to the local/grassroots level, it becomes increasingly difficult and sometimes impossible for state authorities to dictate what urbanites should or should not do. To restore state control over urban space, the community-building campaign has several distinct political imperatives (Bray, 2006). Grassroots governments with defined territories have been endowed with a clear institutional identity; staffed with professional cadres; provided financial resources; and empowered to intervene neighborhood issues, using a wide repertoire of political instruments. A recent cross-sectional study argues that, through the building of community capacity, the community-building campaign promoted the communal grassroots organizations and amenities, which subsequently led to fewer depressive symptoms (Shen, 2014). A follow-up commentary corroborated this enthusiasm for community building and argued that this policy solution to health disparities, if proven to be effective, should be advocated worldwide (Muennig, 2014). However, the argument that community building promotes communal resources is questionable because the latter can be the consequence of other sociospatial changes, such as the rise of territory-based civic organizations, healthcare reforms, or simply the fact that all boats rise on China's urban-transformation tide. Without key measures reflecting the possible downside of the Chinese model of urban governance (e.g., neighborhood conflicts), it is too early to conclude that a policy aiming to strengthen the Party rule at the grassroots level can improve mental health. Moreover, the optimistic view of community building and its ramifications for neighborhood well-being is actually at odds with conclusions drawn by scholars with a keen interest in China's urban transformation (e.g., Bray, 2006; Fu and Lin, 2014; Hsing, 2010; Read, 2003; Tomba, 2014). As noted by Tomba (2005: 948-950), even inherently market-oriented relations such as dealing with PMCs are subject to government interventions, which frustrates urban residents, undermines residential satisfaction, and leads to direct confrontation. Moreover, power relations in Chinese urban neighborhoods are shown to discourage, let alone to promote, civic engagement and communal grassroots organizations (Fu and Lin, 2014). We thus have two conflicting pictures: in the view of a few medical sociologists and epidemiologists, community building promotes mental health, whereas the majority of human geographers, urban sociologists, political scientists, and cultural anthropologists see community building as leading to neighborhood unrest and contentious actions (e.g., Bray, 2006; Fu and Lin, 2014; Hsing, 2010; Read, 2003; Tomba, 2014). This study contributes to this debate by critically examining neighborhood conflicts and their association with depression from an urban-governance perspective.
previously tend to suggest denial of depression among Chinese (Parker et al., 2001), scholars need to understand the broader cultural, sociopolitical and historical context of China's urban transformation in order to identify meaningful stressors that explain the onset of depression. Since the late-1990s, China's central government has initiated a nationwide community-building campaign to restore state control over urban space, which leads to a further restructuring of the triangular relations among state, market, and society. Being officially recognized as “a process that is guided by the Party and the state, …and that harnesses community resources for the purposes of strengthening community functions” (Ministry of Civil Affairs, 2000), the purpose of this community-building model of urban governance is to convey the message that state authorities will do anything in its power to maintain governable urban space (Bray, 2006; Fu and Lin, 2014; Tomba, 2014). From an urban-governance perspective, we next explore how community building comes into being and its link with neighborhood conflicts. We often observe market (e.g., developers and property managers) or social (e.g., civic territorial NGOs) forces in Western urban neighborhoods, but Chinese neighborhoods provide scholars a great opportunity to observe state forces expressed in vibrant power relations. Since this particular constellation of power relations has rarely been addressed by previous studies on urban health, this article is among the first to critically examine the mental-health consequence of contentious urban politics. 2. Background 2.1. Community building as a way of urban governance Urban governance refers to the process through which local authorities, in conjunction with private entities, seek to achieve overarching goals in the city (Jessop, 2002; Pierre, 1999; Wu, 2002). To coordinate the interests of organizational or individual entities, this political process defines, reproduces and transforms relations among various entities in the urban space. In China, these entities involved in urban governance mainly include residents, local/grassroots governments, property management companies (PMCs) and real estate developers (Fu and Lin, 2014; Read, 2003; Tomba, 2014). Grassroots governments, including street offices (jiedao ban) and residents committees (juwei hui), compose subdistrict branches of local governments. They extend the roots of state power into urban neighborhoods through direct interactions with citizens (Lee and Zhang, 2013). While PMCs and developers appear to be nongovernmental, they often maintain various connections with governments due to reasons such as state ownership of enterprises, shared economic interests in land development, or personal networks. Drawing on evidence from cities in North America and Europe, Keating (1991) maintains that urban governance in different sociospatial contexts can be assessed along two aspects: democratic participation and managerial efficiency. Following this line of reasoning, Pierre (1999) eloquently defines four models (or, more precisely, ideal types) of urban governance according to their overarching goals—the managerial, corporatist, progrowth, and welfare models—and argues that real-world urban governance can resemble one or a mixture of the four models. Yet these frameworks primarily derived from Western cities appear to be ineffective in conveying the rich substance of China's urban governance against the backdrop of urban transformation and the persistence of political power (Hsing, 2010; Tomba, 2014; Wu, 2002). Despite the rise of territorial civic organizations (Fu and Lin, 2014) and fierce market-oriented competition within and between Chinese cities (He and Wu, 2009), the conventional democratic and managerial aspects fail to provide a comprehensive instrument for assessing the Chinese model of urban governance. One overarching goal of urban governance in China is the maintenance of state control over urban space, which is featured by a specific constellation of power relations (Bray, 2005; Lee and Zhang, 2013;
2.2. Neighborhood conflicts and community building To understand the link between neighborhood conflicts and community building, we must explore fundamental sociopolitical changes in urban China and answer the following two questions. First, what institutional changes account for the emergence of neighborhood conflicts in a socialist state? Second, with these institutional changes, how does community building lead to a dramatic increase in neighborhood conflicts? The first question should be explored in the context of China's great urban transformation, or more precisely, from workplace (work unit, or danwei) housing to commodity housing. Marxist ideology views unequal homeownership as displaced class struggles between capitalists 2
Social Science & Medicine 196 (2018) 1–9
Q. Fu
3. Research hypotheses
and the working class, and fails to recognize the significance of housing consumption in economic growth (Saunders, 1984). Before the housing reforms in the 1980s, workplaces were the all-encompassing microfoundations of communist rule bestowing economic benefits, social identity, and political rights on their employees (Bray, 2005). Grassroots politics in urban China were then determined by workplaces. As the owner of a housing compound, a workplace not only dictated the allocation of housing but maintained patriarchal authority over its employees who lived in the same workplace compound. Given that most, if not all, urban residents had to rely on their corresponding workplace for housing occupancy, they did not perceive themselves as property owners and played a trivial role in urban governance. However, this welfare housing system turned out to be a major fiscal burden to the socialist regime. Since the early 1980s the central Chinese government was determined to change urban housing from welfare benefits to commodities. After several major waves of housing reforms, most urban households have now achieved homeownership in two ways: acquiring existing workplace housing with price discounts, or purchasing commodity housing in the open market (Fu, 2016). To answer the second question, we must understand the differential effects of housing reforms on urban residents, workplaces, and local/ grassroots governments respectively. Because housing is no longer a privilege conferred by workplace affiliation, urban residents have gradually come to see themselves as housing consumers. Moreover, they also become stakeholders of communal properties, which is formally referred to as partitioned ownership (qufen suoyouquan). In a Chinese urban neighborhood the cost of communal properties (e.g., elevators, stairs, sidewalks, and parking space) is divided and then included in the sale price of each housing unit as shared areas (gongtan). As it stands, China's Real Right Law (National People's Congress, 2007: Article 76) stipulates that decisions pertaining to communal properties must be approved by at least one-half of homeowners in a neighborhood. With these institutional changes, Chinese homeowners are ready to assert their private/communal property rights and fight against violations of their rights. In addition, the demise of workplace-housing compounds also means that the socialist regime can no longer control China's vast urban space through workplaces. As predicted by Vogel at the beginning of China's urban transformation (Vogel, 1989), the regime must rely on territory-based governmental agencies to fill the power vacuum created by the declining influence of workplaces, which makes the community-building campaign necessary. One dilemma created by community building is ill-defined property rights. Although urban residents are now entitled to housing and communal property rights, community building ironically allows grassroots governments to intervene in neighborhood affairs and even allocate communal resources. For example, although homeowners' property rights over housing and communal properties are acknowledged by the Regulation on Realty Management, it also stipulates that homeowners and homeowners' associations should “support the residents' committee to carry out the work, and subject themselves to direction and supervision by the residents' committee” (Article 20: State Council, 2007 [2003]). The 2007 Revision of Regulation on Realty Management also explicitly specifies street offices as the supervising unit of civic homeowners congress/associations; this Revision is thus in line with the earlier Guidelines for Homeowners Congress (Ministry of Housing and Urban-Rural Development, 2003). Such ill-defined property rights across Chinese neighborhoods appear to account for a flood of urban unrest. A study by Tomba (2014) demonstrates that the deregulation of China's urban housing market has actually been accompanied by more unanticipated government interventions in residential neighborhoods. The peculiar mixture of housing commodification and community building then becomes a source of neighborhood grievances and conflicts. Claims of territorial legitimacy are constantly asserted, contested, and denied among entities of urban governance, which provides justification for the deployment of political coercion or collective resistance.
Although neighborhood conflicts are probably less intense than labor protests and deal with local rather than global issues, their relevance to depressive symptoms should not be underrated. In 2003, only five years after the official deregulation of China's urban housing market, problems related to housing and urban development ranked as the third-leading cause (after peasant burdens and mass layoffs) of all petitions received by the central government and accounted for 25 percent of all lawsuits in Zhejiang Province (Z. Song et al., 2003). Despite the scarcity of recent statistics, several qualitative studies suggest increasing neighborhood conflicts in China (Fu and Lin, 2014; Hsing, 2010; Tomba, 2014). Low residential mobility resulting from skyrocketing housing prices also makes the situation worse (S.-m. Li and Zhu, 2014). Residents have to live with long-lasting conflicts since moving seldom appears to be a viable solution. The perception of neighborhood conflicts is thus deemed as one stressor in urban China. Moreover, neighborhood conflicts inducing problem-solving actions are possibly correlated with depressive symptoms for two possible reasons. First, engagement in problem solving suggests the severity of neighborhood conflicts. Urban residents in a (post-)socialist state tend to stay away from civic engagement because of previous stressful experiences of mass mobilization, political pressure from state agencies, and reliance on private networks (Fu and Lin, 2014; Howard, 2003). Trivial or innocent neighborhood conflicts may not warrant engagement in problem solving and have little impact on mental health. Second, contentious actions aiming at solving neighborhood conflicts are often bounded by state-prescribed channels (e.g., petitions or mediation), which involve political pressure and appear to be arduous (Lee and Zhang, 2013; Perry, 2008). The first hypothesis is formulated as follows: Hypothesis 1. Net of other sociodemographic, relational, and environmental measures, both the perception of neighborhood conflicts and conflicts inducing problem-solving actions are significantly associated with more depressive symptoms. Nevertheless, an intriguing question is probably not whether neighborhood conflicts matter to depression but rather what aspect of neighborhood conflicts matters. According to Perry (2008), the Chinese notion of government legitimacy stems from its capacity to provide social security and address citizens' livelihood needs. Since community building paradoxically allows local/grassroots governments to intervene in neighborhood affairs and allocate communal resources belonging to urban residents, the ill-defined property rights and subsequent neighborhood instability caused by state intervention contradict the very origin of government legitimacy perceived by its own citizens. Since this dilemma resulting from community building is fueled by neighborhood conflicts, there has been an increasing public outcry for government accountability. Meanwhile, proximity to local authorities explains why other entities of urban governance, including developers and PMCs, can achieve power over residents. For example, once grassroots governments make PMCs responsible for the neighborhoods they manage, property managers can dictate the use of revenues generated from communal amenities (e.g., parking lots, gyms, clubs, swimming pools) regardless of collective decisions made by residents (Fu and Lin, 2014; Tomba, 2014). Given that the predominant role of state power in shaping neighborhood affairs is restored by community building, the state's presence is also experienced as a constraining influence on perceiving and solving neighborhood conflicts. For example, conflicts with developers or PMCs and related problemsolving actions do not necessarily trigger the deployment of state tactics for maintaining grassroots stability but these with local/grassroots governments do (Fu and Lin, 2014; Tomba, 2014). The repertoire of tactics then revolves around intimidation, coercion, and even the use of force meant to quell any (potential) threat to political stability (Lee and Zhang, 2013). Therefore, conflicts with local/grassroots governments 3
Social Science & Medicine 196 (2018) 1–9
Q. Fu
the dependent variable instead of a dichotomous variable of clinical depression.
are less likely to be translated to problem-solving actions and appear to be more stressful than these with developers or PMCs: Hypothesis 2. Among residents' conflicts with other entities of neighborhood governance, those with local/grassroots governments are most relevant to depression.
4.2.2. Sociodemographic variables Various sociodemographic variables, either with known consequences on depression or recognized as important indicators of social stratification in Chinese society, were considered (Bray, 2005; Hsieh, 2015; Jin et al., 2012; Parker et al., 2001; Yen et al., 2000): sex (female versus male); age (in years); marital status (currently married versus others); years of schooling; the logarithm of annual income (in 10,000 China Yuan); perceived socioeconomic status; unemployment (being employed, retired, or at school as the reference); housing tenure (homeownership versus tenancy); length of residence (in years); Communist Party membership; and state-sector employment (working in governments, institutes, and military versus others). Perceived socioeconomic status is an ordinal variable with five categories: lower class (coded as 1), lower-middle class, middle class, upper-middle class, and upper class (coded as 5).
Moreover, our understanding of neighborhood conflicts with governments cannot be isolated from the political economy of place. The manipulation of neighborhood planning, including changes in floorarea ratios (FARs), house demolition, and land expropriation, has been demonstrated as a lucrative way for local governments to acquire wealth at their own disposal (Fu and Lin, 2014; Hsing, 2010; Logan and Molotch, 2007). Communal properties (e.g., parking lots, shops, and outdoor advertising) can also generate substantial revenues given that Chinese urban neighborhoods are often sizable (Fu and Lin, 2014). Since community building in China provides an institutionalized way for those with political connections to interfere in neighborhood affairs, the enormous revenues generated by changes in neighborhood planning and communal properties are then used to consolidate their control over urban neighborhoods (Fu and Lin, 2014; Hsing, 2010; Tomba, 2014). Due to this coalition between political power and economic capital, these specific conflicts often perpetuate themselves and resist change, which generates feelings of loss of control and helplessness among urban residents.
4.2.3. Neighborhood conflicts A measure of neighborhood conflicts is developed in the survey. Prior to the survey, three focus groups, were formed comprised of local scholars, NGO leaders, neighborhood activists, and urban residents. The groups met to identify conflicts over a wide range of neighborhood issues and assess the questionnaires to be used in the survey. These conflicts were further organized by three entities of urban governance in China: developers, PMCs, and local/grassroots governments. Moreover, survey respondents were viewed as conscious and active agents making their lives instead of passive subjects overwhelmed by stressors (Thoits, 1994). They were asked whether they were aware of different types of neighborhood conflicts in the most recent five years, and if yes, whether they had participated in solving the conflict.
Hypothesis 2A. Among different subgroups of conflicts with governments, those relevant to neighborhood planning and communal properties are more likely to be associated with depression. 4. Data and methods 4.1. Data
4.2.4. Relational and environmental variables This survey used the position-generator method to measure network-based social capital (Lin and Dumin, 1986). Respondents were asked to report whether they knew someone in a list of 23 occupations (e.g., journalists, doctors, lawyers, janitors) and each occupation was assigned an occupational prestige score (C. Li, 2005). Three measures of social networks were constructed (L. Song and Lin, 2009): network extensity (the number of occupations accessed), upper reachability (the highest occupational prestige score accessed) and the range of occupational prestige scores accessed. As neighborhood acquaintances provide social leverage for gaining information, resources and influence (Briggs, 1998), another relational variable is the logarithm of the selfreported number of acquaintances (measured as people one can talk to about private matters) within a neighborhood. Four variables neighborhood attachment, informal neighborhood control, neighborhood cohesion and appraisal of neighborhood environment were constructed to measure neighborhood environment. Based on results from focus groups, survey items related to these variables were adapted and revised from existing measures (e.g., Lewicka, 2011; Sampson et al., 1999) so that they are clear to potential respondents in Guangzhou. Neighborhood attachment, consisting of both relational and environmental dimensions (Lewicka, 2011), was measured by six items (see Table 1). Respondents chose answers from a five-point scale ranging from “strongly agree” to “strongly disagree.” Survey items related to informal neighborhood control and neighborhood cohesion were adapted from research on collective efficacy (Sampson et al., 1999). For informal neighborhood control, respondents were asked about the likelihood (ranging from “very likely” to “very unlikely”) that intervention into several socially deviant behaviors (see Table 1) within their neighborhood would occur. Neighborhood cohesion was measured by eight survey items using a five-point Likert scale ranging from “strongly agree” to “strongly disagree” (also see Table 1 for these survey
This study is based on data from a face-to-face survey of thirty-nine urban neighborhoods in Guangzhou, China. The survey, conducted in from October 2012 to January 2013, adopted a multistage stratified random sampling method to recruit respondents. First, three primarysampling-unit (PSU) strata within the border of the outer-ring road of Guangzhou were specified by land-use purpose and population density: the inner core area, the inner suburb area, and the outer suburb area. Next, street offices within each stratum were selected by a GIS sampling procedure. Within each selected street office, one target neighborhood and a list of adjoining neighborhoods were obtained by GIS sampling. Local scholars or NGO leaders referred the survey team in order to access a target neighborhood. Despite the assistance from local universities and civic organizations, gatedness and intervention from property managers prevented data collection in one-third of the target neighborhoods; these were replaced by adjoining ones from the list. Within a selected urban neighborhood, households were selected by an interval sampling procedure according to the residential distribution of that neighborhood. Trained interviewers interviewed the household head or, if unavailable, his/her spouse or another adult family member familiar with the neighborhood. The survey's response rate was 65 percent; 1470 respondents from this survey are included in the analysis. 4.2. Measures 4.2.1. Dependent variable The full-length twenty-item version of the Center for Epidemiologic Studies Depression Scale (CES-D) was used to measure depressive symptoms in the last week, and the CES-D scores were calculated based on Radloff (1977). The Cronbach's alpha of the twenty CES-D items in the overall survey sample was 0.85. Given the low levels of depressive symptoms in China (Hsieh, 2015; Parker et al., 2001), most CES-D scores calculated stay below the clinical cut-off score of sixteen. Thus the total CES-D score (with a possible range from 0 to 60) was used as 4
Social Science & Medicine 196 (2018) 1–9
Q. Fu
Table 1 Principal component analysis of survey items related to neighborhood environment.a Neighborhood attachment
I feel attached to this neighborhood As a dwelling place, I like my neighborhood I do not want to move out of this neighborhood If I had to move out, I would miss people in this neighborhood If I had to move out, I would miss the neighborhood environment here My connection with other people in this neighborhood means a lot to me
Informal neighborhood control
Neighborhood cohesion
0.406 0.401 0.424 0.423 0.414 0.381
Vandalism of public property such as elevators and fire hydrants Damage to neighborhood environment (e.g., spitting and graffiti) Verbal or physical fights among residents
0.572 0.608 0.551
People in this neighborhood can get together to deal with problems This is a close-knit neighborhood People around here are willing to help each other I can count on others to pick up my mail, milk, and newspapers People in this neighborhood get along with each other People in this neighborhood can be trusted. People have very different opinions of neighborhood affairs a,b No one in this neighborhood cares much about what happens to others b
0.394 0.430 0.434 0.327 0.366 0.377 0.086 0.284
Location and transportation accessibility Landscape and architecture Sanitation service Responsiveness of property managers to individual requests Maintenance of public property Security service Parking management Parking availability Eigenvalue associated with the principal component Variance explained by the principal component a b
Appraisal of neighborhood environment
0.269 0.341 0.417 0.404 0.410 0.393 0.351 0.174 3.372 56.2%
2.098 69.9%
2.241 42.2%
3.520 44.0%
This survey item's effect on the principal component was removed due to its small factor loading. This item was reverse-coded.
items). The appraisal of the neighborhood environment, a key determinant of residential experience in urban China (Zhu et al., 2012), corresponds to residents' subjective evaluation of eight aspects of the neighborhood environment: location and transportation accessibility, landscape and architecture, sanitation service, responsiveness of property managers, maintenance of collective property, security service, parking management, and parking availability, using a five-point scale ranging from “very good” to “very bad.”
Depressionij = β0 +
∑ β Dp Dp + ∑ βSESq SESq + ∑ βSNr SNr p=3
+
q=8
r=4
∑ βEnvironments Environments + ∑ βConflictt Conflictt + u0j s=4
+ εij The three demographic variables (sex, age, and marital status) are represented by Dp; SESq refers to each of the eight (dummy) variables measuring socioeconomic status; the four variables related to social networks are denoted by SNr. Environments corresponds to the four variables measuring neighborhood environment; the key variables of this study, neighborhood conflicts between residents and other three entities of urban governance are denoted by Conflictt. βs represents the corresponding coefficients of these covariates. u0j is the residual random effect of community j, which is normally distributed with variance τ0. The error term εij is normally distributed with variance τε. Given that Breusch-Pagan tests suggest heteroscedasticity (results not shown), heteroscedasticity-robust standard errors are reported.
4.3. Method Statistical analysis was implemented in two steps. In the first step, principal component analyses (PCA) were conducted on the four sets of survey items related to neighborhood attachment, informal neighborhood control, neighborhood cohesion, and appraisal of neighborhood environment. The four principal components generated (with eigenvalues greater than 1) explain 56.2 percent, 69.9 percent, 42.2 percent, and 44.0 percent of the variances of the four sets of survey items respectively (see Table 1). A data-reduction procedure was employed to remove the influence of survey items with trivial loadings (Land et al., 1990). In the second step, random-effect linear models were used to assess factors associated with depressive symptoms. The full model can be expressed as:
5. Results The mean CES-D score of the sample is 9.3, which is consistent with previous findings of low prevalence rates of depression in Chinese populations (Hsieh, 2015; Parker et al., 2001; Yen et al., 2000). Less than one-half (43.6 percent) of respondents were male, and most of them 5
Social Science & Medicine 196 (2018) 1–9
Q. Fu
Table 2 Neighborhood conflicts by entities of urban governance (N = 1470).
Neighborhood conflicts Conflicts with developers Housing amenities Housing quality Size of shared residential areas Housing ownership certificates Conflicts with property management companies Parking Property-management fees Sanitation and gardening Appropriation of homeowners' collective revenues Utilities Harassment from security personnel Conflicts with local and grassroots governments Changes in urban planning nearby Changes in neighborhood planning Violations to partitioned ownership Intervention in the operation of homeowners associations Appropriation of homeowners' collective revenues Housing ownership certificates Intervention in the election of homeowners associations (HOAs)
Perceived
Engaged in Problem solving
Ratio: Engaged/Perceived
46.5% 22.9% 15.6% 10.6% 10.0% 4.2% 39.3% 21.5% 21.3% 15.7% 14.5% 10.6% 5.0% 15.0% 7.7% 6.7% 5.9% 3.4% 3.0% 3.0% 2.9%
18.1% 7.5% 4.5% 2.4% 2.7% 1.1% 13.5% 5.5% 7.5% 3.8% 4.0% 3.0% 0.5% 3.8% 1.7% 1.1% 1.1% 1.1% 0.5% 0.5% 0.9%
0.39 0.33 0.29 0.23 0.27 0.26 0.34 0.26 0.35 0.24 0.28 0.28 0.10 0.25 0.22 0.16 0.19 0.32 0.17 0.17 0.31
Table 3 Random-effect regression of CES-D scores on socio-demographic background, network and perceived neighborhood environment (N = 1470).a Model 1
Male Age Married Years of schooling Log of income Perceived SES Unemployed Homeownership Length of residence Party membership State employment Network extensity Network upper reachability Network range Log of number of acquaintances Neighborhood attachment Informal neighborhood control Neighborhood cohesion Appraisal of neighborhood environ. Neighborhood conflicts (perceived) Neighborhood conflicts (problem solving) Constant Random-effects parameter σ0 t Random-effects parameter σε t χ2 Log pseudolikelihood
Model 2
Model 3
Model 4
Model 5
Model 6
Coeff.
S.E.
Coeff.
S.E.
Coeff.
S.E.
Coeff.
S.E.
Coeff.
S.E.
Coeff.
S.E.
0.126 −0.012 −1.015* 0.150 −0.228 −0.631*** 2.447** −0.179 −0.012 −0.548 0.170
0.273 0.012 0.473 0.077 0.218 0.170 0.804 0.393 0.026 0.407 0.398
0.132 −0.013 −0.862 0.155 −0.236 −0.595** 2.410** −0.252 0.005 −0.489 0.223 0.007 −0.007 −0.013 −0.404**
0.276 0.012 0.490 0.081 0.212 0.173 0.795 0.394 0.024 0.392 0.376 0.054 0.008 0.011 0.121
0.217 −0.010 −0.940* 0.135 −0.295 −0.523** 2.225** −0.152 −0.002 −0.485 0.321
0.267 0.012 0.442 0.072 0.195 0.171 0.825 0.380 0.025 0.401 0.358
−0.184 −0.283* −0.242* −0.223*
0.111 0.109 0.113 0.091
0.212 −0.010 −0.878 0.142 −0.285 −0.508** 2.255** −0.209 0.005 −0.449 0.331 0.021 −0.007 −0.012 −0.182 −0.177 −0.259* −0.211 −0.229*
0.270 0.012 0.459 0.077 0.192 0.176 0.807 0.379 0.024 0.391 0.350 0.051 0.008 0.011 0.119 0.113 0.107 0.110 0.092
0.193 −0.008 −0.858 0.127 −0.270 −0.527** 2.248** −0.321 −0.005 −0.469 0.318 0.021 −0.006 −0.013 −0.198 −0.179 −0.270* −0.212 −0.213* 0.742*
0.272 0.012 0.453 0.076 0.195 0.179 0.846 0.383 0.024 0.384 0.352 0.051 0.008 0.011 0.117 0.114 0.109 0.111 0.092 0.287
0.196 −0.011 −0.813 0.120 −0.261 −0.509** 2.258** −0.290 0.001 −0.455 0.334 0.013 −0.006 −0.011 −0.214 −0.183 −0.278** −0.216 −0.214*
0.272 0.012 0.461 0.074 0.193 0.171 0.825 0.379 0.024 0.388 0.354 0.050 0.008 0.011 0.118 0.113 0.105 0.111 0.090
11.067*** 0.820***
1.179 0.208
12.007*** 0.789***
1.167 0.221
10.780*** 0.837***
1.193 0.173
11.495*** 0.820***
1.181 0.180
11.404*** 0.841***
1.150 0.178
1.055* 11.626*** 0.832***
0.467 1.146 0.168
5.504***
0.122
5.479***
0.120
5.400***
0.114
5.392***
0.115
5.379***
0.112
5.377***
0.112
75.34*** −4604.2
165.44*** −4596.8
116.26*** −4576.8
174.82*** −4574.2
167.15*** −4571.1
179.20*** −4570.4
Note: ***p < 0.001 **p < 0.01 * p < 0.05. a Robust standard errors are reported.
listed (5.1 ± 4.5). The mean upper reachability was 73.5, roughly about the prestige scores of nurses, entrepreneurs, and engineers in Chinese society (C. Li, 2005). The average range of prestige scores was 23.3, about the distance from restaurant waitresses to nurses, or from photographers to university professors. The average number of acquaintances within a neighborhood was 5.8 ± 13.6. Table 2 shows the prevalence of neighborhood conflicts by entities of neighborhood governance. Related to Hypothesis 2, urban residents were less likely to take action if they perceive local/grassroots governments instead of developers or PMCs as responsible for the same
were married and above age thirty (the mean age was 45.0 ± 14.7). Their mean years of schooling were 10.3 years, and annual income was about 73,000 yuan. The average score of perceived socioeconomic status was 2.3, which falls between lower-middle class and middle class. The unemployed only compose 2.3 percent of the sample. Most of the respondents were homeowners (79.7 percent), and the average years of residence were 7.2 years. About 20 percent of the respondents were Party members (20.7 percent) or working in the state sector (21.5 percent). In terms of network measures, respondents on average had access to about five occupations out of the twenty-three occupations 6
Social Science & Medicine 196 (2018) 1–9
Q. Fu
Table 4 Random-effect regression of CES-D scores on neighborhood conflicts (N = 1470).a Model 1
Male Age Married Years of schooling Log of income Perceived SES Unemployed Homeownership Length of residence Party membership State employment Network extensity Network upper reachability Network range Log of number of acquaintances Neighborhood attachment Informal neighborhood control Neighborhood cohesion Appraisal of neighborhood environment Neighborhood conflicts (perceived) with developers with property management companies with local and grassroots governments Neighborhood conflicts (problem solving) with developers with property management companies with local and grassroots governments Conflicts with governments (perceived) Changes in urban planning nearby Changes in neighborhood planning Violations to partitioned ownership Intervention in the operation of HOAs Appropriation of collective revenues Housing ownership certificates Intervention in the election of HOAs Conflicts with governments (problem solving) Changes in urban planning nearby Changes in neighborhood planning Violations to partitioned ownership Intervention in the operation of HOAs Appropriation of collective revenues Housing ownership certificates Intervention in the election of HOAs Constant Random-effects parameter σ0 Random-effects parameter σε t χ2 Log pseudolikelihood
Model 2
Model 3
Model 4
Coeff.
S.E.
Coeff.
S.E.
Coeff.
S.E.
Coeff.
S.E.
0.164 −0.007 −0.835 0.121 −0.268 −0.511** 2.298** −0.368 −0.004 −0.417 0.302 0.014 −0.006 −0.014 −0.215 −0.208 −0.262* −0.192 −0.204*
0.263 0.012 0.441 0.075 0.196 0.176 0.829 0.389 0.024 0.387 0.352 0.051 0.008 0.012 0.116 0.111 0.107 0.110 0.092
0.183 −0.010 −0.811 0.122 −0.265 −0.521** 2.323** −0.312 0.001 −0.420 0.323 0.010 −0.006 −0.012 −0.227 −0.194 −0.271* −0.205 −0.214*
0.273 0.012 0.454 0.074 0.192 0.170 0.821 0.378 0.024 0.396 0.350 0.048 0.008 0.011 0.124 0.111 0.105 0.109 0.094
0.162 −0.008 −0.828 0.125 −0.276 −0.488** 2.310** −0.297 0.001 −0.450 0.414 0.010 −0.005 −0.012 −0.207 −0.214 −0.257* −0.190 −0.210*
0.266 0.012 0.442 0.073 0.195 0.166 0.766 0.372 0.025 0.391 0.352 0.050 0.008 0.011 0.111 0.111 0.105 0.109 0.094
0.189 −0.009 −0.887 0.137 −0.299 −0.522** 2.340** −0.289 −0.003 −0.403 0.318 0.013 −0.007 −0.011 −0.223 −0.194 −0.270* −0.191 −0.228*
0.275 0.012 0.453 0.076 0.195 0.172 0.808 0.377 0.024 0.401 0.335 0.047 0.008 0.010 0.122 0.111 0.105 0.109 0.100
0.527 0.266 1.094*
0.471 0.336 0.477 0.740 0.284 2.067**
0.613 0.462 0.748 0.882 1.585* −0.849 1.076 −0.100 −0.486 1.588
0.546 0.687 1.030 1.194 1.121 1.043 1.107 2.396 1.250 4.109** −1.530 3.613 −1.096 1.088 11.619*** 0.868*** 5.339***
1.239 1.272 1.409 2.452 2.315 2.854 2.647 1.142 0.175 0.117
11.376*** 0.855*** 5.358***
1.160 0.175 0.113
174.18*** −4565.9
11.653*** 0.852*** 5.363*** 183.70*** −4567.0
1.131 0.170 0.113
11.397*** 0.810*** 5.355*** 288.66*** −4563.9
1.153 0.182 0.114
223.63*** −4560.9
Note: ***p < 0.001 **p < 0.01 * p < 0.05. a Robust standard errors are reported.
were included in Model 4. Both informal neighborhood control and appraisal of neighborhood environment had significantly negative associations with depressive symptoms, and these negative associations remained significant when relational measures were included (p < 0.05 in Model 4). Net of other sociodemographic, relational, and environmental measures, the perception of neighborhood conflicts and conflicts inducing problem-solving actions were significantly associated with more depressive symptoms (p < 0.05 in both Model 5 and Model 6). This finding lends support to Hypothesis 1. Regression analysis shown in Table 4 further considers different types of neighborhood conflicts. Among neighborhood conflicts with three entities of urban governance, only those with local/grassroots governments were significantly associated with more depressive symptoms (p < 0.05 in Model 1). This conclusion remained the same if perceived conflicts were replaced by conflicts inducing problem-solving actions (p < 0.01 in Model 2), which supports Hypothesis 2. Models 3
type of conflicts: about one-third of perceived conflicts with developers and PMCs led to problem-solving actions; in contrast, only one-quarter of conflicts with local/grassroots governments resulted in problemsolving actions. The lower ratios between perceived conflicts with governments and related problem-solving actions also exist for subgroups of conflicts, such as problems related to housing ownership certificates and appropriation of homeowners' collective revenues. The first set of random-effect regression models considers the influence of sociodemographic background, social networks, and neighborhood environment on depressive symptoms (see Table 3). Across all six models in Table 3, perceived socioeconomic status was significantly and negatively associated with depressive symptoms, whereas unemployment was linked with more depressive symptoms. The only significant relational measure is the number of neighborhood acquaintances (p < 0.01 in Model 2), but its significant association with depression disappeared when variables of neighborhood environment 7
Social Science & Medicine 196 (2018) 1–9
Q. Fu
understanding urban governance, which is inherently path dependent and context specific, as something given merely blinds us to the complexity of politics and thereby helps justify the status quo. According to Harvey (1978), urban spatiality is simultaneously a presupposition, a medium, and a product of ever-evolving social relations. Since urban governance is constrained by its social, historical, and cultural contingencies, a “one measure fits all” approach, regardless of the specific research context, is probably less productive. In this regard, the community-building campaign does not substantially deviate from previous top-down, authoritarian, and campaign-like policies in socialist regimes, which determine the well-being of people according to their proximity to political power. Community building will probably continue to consolidate the state's rule over urban space, but its multifaceted effects on mental health warrant further investigation.
and 4 in Table 3 further investigate the link between subgroups of government-oriented conflicts and depressive symptoms. For perceived conflicts with governments, the variable changes in neighborhood planning had a positive and significant association with depression (p < 0.05 in Model 3); for government-oriented conflicts inducing problem-solving actions, the variable violations to partitioned ownership showed a significant association with depression (p < 0.01 in Model 4). In other words, these government-oriented conflicts that are significantly associated with more depressive symptoms were all related to neighborhood planning or communal properties, which supports Hypothesis 2A. Across all models in Table 4, the results pertaining to sociodemographic background, social relations, and neighborhood environment remained virtually the same as those reported in Table 3. 6. Discussion
Acknowledgements Drawing on data from a community-based survey, this article investigates important sociopolitical changes over China's urban space and their implications on depression. It finds that: 1. net of other sociodemographic, relational, and environmental measures, neighborhood conflicts are significantly associated with more depressive symptoms (Hypothesis 1); 2. among neighborhood conflicts with three entities of urban governance, only those with local/grassroots governments have a significant and positive association with depressive symptoms (Hypothesis 2); 3. if we further examine subgroups of conflicts with governments, those conflicts associated with more depressive symptoms appear to be related to neighborhood planning and communal properties (Hypothesis 2A). The fact that China's new urban space in the era of globalization, decentralization and marketization could be restored to a governable space subject to political power testifies authoritarian dominance at its height. Exposure to neighborhood conflicts is stressful, but equally stressful is an understanding that these conflicts are deeply embedded in urban governance. Under the community-building model of urban governance, many came to believe that neighborhood conflicts are the rule rather than the exception (Fu and Lin, 2014; Tomba, 2014). The limitations of this research should be acknowledged. First, because both the exact timing of conflicts and trajectories of depressive symptoms are absent in this cross-sectional survey, no definite causal conclusion can be drawn. The CES-D score only reflects a one-week snapshot of depression and is thus highly context dependent. Moreover, reverse causality from depression to conflicts may exist if individuals with mental disorders are more (or less) likely to solve neighborhood conflicts. Second, the proposed measures of neighborhood conflicts are far from exhaustive. A severity scale of neighborhood conflicts would also be useful in future research. Finally, given that the sample was collected in one Chinese city, scholars should be cautious in applying these conclusions to other urban settings with different structures of governance and built-environment characteristics (e.g., gatedness, size of the neighborhood, housing affordability, and amount of communal properties). Scholars tend to stay away from a Foucauldian approach in searching for policy solutions to social problems, fearing that the analysis of the specific power relations at stake could compromise the generalizability of policies being pursued. This study instead suggests that our understanding of mental illness can be advanced by a closer investigation into urban politics. One striking feature of research on mental health is the degree to which urban politics has been neglected. Earlier thinkers such as Max Weber, Karl Marx, and Bertrand Russell have meticulously demonstrated power as a fundamental concept in the social sciences. Today we often focus on established proxies for power (e.g., education, income, and social class) but fail to examine how power can be expressed in more subtle ways. Since urban residents are always positioned within a web of power relations, the urban-governance perspective allows us to view the urban space as a spatial manifestation of power and further reveal the politics of mental illness. Yet
The author gratefully acknowledges financial support from The Chiang Ching-kuo Foundation for International Scholarly Exchange and The Hampton New Faculty Award at The University of British Columbia. References Bray, D., 2005. Social Space and Governance in Urban China: the Danwei System from Origins to Reform. Stanford University Press, Palo Alto, CA. Bray, D., 2006. Building ‘community’: new strategies of governance in urban China. Econ. Soc. 35, 530–549. Brenner, N., 2004. Urban governance and the production of new state spaces in Western Europe, 1960–2000. Rev. Int. Political Econ. 11, 447–488. Briggs, X.d.S., 1998. Brown kids in white suburbs: housing mobility and the many faces of social capital. Hous. Policy Debate 9, 177–221. Browning, C.R., Cagney, K.A., 2002. Neighborhood structural disadvantage, collective efficacy, and self-rated physical health in an urban setting. J. Health Soc. Behav. 43, 383–399. Burris, S., Hancock, T., Lin, V., Herzog, A., 2007. Emerging strategies for healthy urban governance. J. Urban Health 84, 154–163. Friedmann, J., 2007. Reflections on place and place-making in the cities of China. Int. J. Urban Regional Res. 31, 257–279. Fu, Q., 2016. The persistence of power despite the changing meaning of homeownership: an age-period-cohort analysis of urban housing tenure in China, 1989-2011. Urban Stud. 53, 1225–1243. Fu, Q., Lin, N., 2014. The weaknesses of civic territorial organizations: civic engagement and homeowners associations in urban China. Int. J. Urban Regional Res. 38, 2309–2327. Gong, P., Liang, S., Carlton, E.J., Jiang, Q., Wu, J., Wang, L., et al., 2012. Urbanisation and health in China. Lancet 379, 843–852. Harvey, D., 1978. The urban process under capitalism: a framework for analysis. Int. J. Urban Regional Res. 2, 101–131. He, S., Wu, F., 2009. China's emerging neoliberal urbanism: perspectives from urban redevelopment. Antipode 41, 282–304. Howard, M.M., 2003. The Weakness of Civil Society in Post-communist Europe. Cambridge University Press, Cambridge. Hsieh, N., 2015. Economic security, social cohesion, and depression disparities in posttransition societies a comparison of older adults in China and Russia. J. Health Soc. Behav. 56, 534–551. Hsing, Y.-t., 2010. The Great Urban Transformation: Politics of Land and Property in China. Oxford University Press, Oxford, UK. Jessop, B., 2002. Liberalism, neoliberalism, and urban governance: a state–theoretical perspective. Antipode 34, 452–472. Jin, L., Wen, M., Fan, J.X., Wang, G., 2012. Trans-local ties, local ties and psychological well-being among rural-to-urban migrants in Shanghai. Soc. Sci. Med. 75, 288–296. Keating, M., 1991. Comparative Urban Politics: Power and the City in the United States, Canada, Britain, and France. Edward Elgar Publishing, Aldershot, UK. Land, K.C., McCall, P.L., Cohen, L.E., 1990. Structural covariates of homicide rates: are there any invariances across time and social space? Am. J. Sociol. 95, 922–963. Lee, C.K., Zhang, Y., 2013. The power of instability: unraveling the microfoundations of bargained authoritarianism in China. Am. J. Sociol. 118, 1475–1508. Lewicka, M., 2011. Place attachment: how far have we come in the last 40 years? J. Environ. Psychol. 31, 207–230. Li, C., 2005. Prestige Stratification in the Contemporary China: occupational prestige measures and socio-economic index. Sociol. Res. 74–102. Li, S.-m., Zhu, Y., 2014. Residential mobility within Guangzhou city, China, 1990–2010: local residents versus migrants. Eurasian Geogr. Econ. 55, 313–332. Lin, N., Dumin, M., 1986. Access to occupations through social ties. Soc. Netw. 8, 365–385. Logan, J.R., Molotch, H.L., 2007. Urban Fortunes: the Political Economy of Place. University of California Press, Berkeley, CA.
8
Social Science & Medicine 196 (2018) 1–9
Q. Fu
capital, comparative reference group, and depression in two societies. Soc. Sci. Med. 136, 117–127. Song, L., Lin, N., 2009. Social capital and health inequality: evidence from Taiwan. J. Health Soc. Behav. 50, 149–163. Song, Z., Zhou, G., Cui, L., 2003. Urban Redevelopment: the Third Cause of Petitions. Xinhua News Agency, Beijing People.com.cn. State Council. (2007 [2003]). Regulation on Realty Management. Beijing: Document No. 379. Thoits, P.A., 1994. Stressors and problem-solving: the individual as psychological activist. J. Health Soc. Behav. 35, 143–160. Tomba, L., 2005. Residential space and collective interest formation in Beijing's housing disputes. China Q. 184, 934–951. Tomba, L., 2014. The Government Next Door: Neighborhood Politics in Urban China. Cornell University Press, Ithaca, NY. Vogel, E.F., 1989. One Step Ahead in China: Guangdong under Reform. Harvard University Press, Cambridge, MA. Wu, F., 2001. China's recent urban development in the process of land and housing marketisation and economic globalisation. Habitat Int. 25, 273–289. Wu, F., 2002. China's changing urban governance in the transition towards a more market-oriented economy. Urban Stud. 39, 1071–1093. Yen, S., Robins, C.J., Lin, N., 2000. A cross-cultural comparison of depressive symptom manifestation: China and the United States. J. Consult. Clin. Psychol. 68, 993–999. Zhu, Y., Breitung, W., Li, S.-m, 2012. The changing meaning of neighbourhood attachment in Chinese commodity housing estates: evidence from Guangzhou. Urban Stud. 49, 2439–2457.
Ministry of Civil Affairs, 2000. Circular on the Promotion of Community Building in Urban China. Beijing. Ministry of Housing and Urban-Rural Development, 2003. Guidelines for Homeowners Congress. Beijing. Muennig, P., 2014. What China's experiment in community building can tell us about tackling health disparities. Soc. Sci. Med. 107, 217–220. National People's Congress, 2007. Real Right Law of the People's Republic of China. Beijing. Parker, G., Gladstone, G., Chee, K.T., June 2001. Depression in the planet's largest ethnic group: the Chinese. Am. J. Psychiatry 158 (6), 857–864. Perry, E.J., 2008. Chinese conceptions of “rights”: from Mencius to Mao—and now. Perspect. Polit. 6, 37–50. Pierre, J., 1999. Models of urban governance: the institutional dimension of urban politics. Urban Aff. Rev. 34, 372–396. Radloff, L.S., 1977. The CES-D scale: a self-report depression scale for research in the general population. Appl. Psychol. Meas. 1, 385–401. Read, B.L., 2003. Democratizing the neighbourhood? New private housing and homeowner self-organization in urban China. China J. 31–59. Sampson, R.J., Morenoff, J.D., Earls, F., 1999. Beyond social capital: spatial dynamics of collective efficacy for children. Am. Sociol. Rev. 633–660. Saunders, P., 1984. Beyond housing classes: the sociological significance of private property rights in means of consumption†. Int. J. Urban Regional Res. 8, 202–227. Shen, Y., 2014. Community building and mental health in mid-life and older life: evidence from China. Soc. Sci. Med. 107, 209–216. Simmel, G. (Ed.) ([1903] 1964). The Metropolis and Mental Life. New York: Free Press. Song, L., 2015. Does who you know in the positional hierarchy protect or hurt? Social
9
Contents lists available at ScienceDirect
Social Science & Medicine journal homepage: www.elsevier.com/locate/socscimed
Bringing urban governance back in: Neighborhood conflicts and depression
MARK
Qiang Fu Department of Sociology, The University of British Columbia, V6T 1Z1, Vancouver, BC, Canada
A R T I C L E I N F O
A B S T R A C T
Keywords: Depression Neighborhood conflicts Urban transformation Community building Relations Urban governance Contentious politics China
Urban governance and its impact on contentious politics have received remarkably little attention in existing studies on mental health. Drawing on a measure of neighborhood conflicts developed in a survey of thirty-nine urban neighborhoods in Guangzhou, China, this article investigates the potential link between urban governance and mental health. Net of sociodemographic, relational, and environmental measures, it finds that among residents' conflicts with different entities of urban governance, only those with local/grassroots governments are significantly associated with more depressive symptoms. Moreover, these subgroups of government-oriented conflicts associated with more depressive symptoms are related to neighborhood planning and communal properties, reflecting a dilemma in the Chinese model of urban governance. By offering a relational interpretation of neighborhood conflicts, this study not only challenges the previous view that community building in China improves mental health, but calls attention to the significance of urban governance in research on mental health.
1. Introduction Mental illness in the city has long been an intriguing topic for social scientists. Simmel ([1903] 1964) deemed “profusion of sensory stimuli,” including sights, sounds, smells, and interference, as the city's most profound effects. Individuals must adapt, “react with their heads instead of hearts,” and become aloof when dealing with urban life. Until recently, however, the issue of urban governance as a key mechanism for the restructuring of urban space (Brenner, 2004; Jessop, 2002; Pierre, 1999; Wu, 2002) has largely been neglected by health scholars. Brenner (2004: 455) describes urban governance as “the broad constellation of social, political and economic forces that mold the process of urban development within modern capitalism.” While its scope is not necessarily restricted to modern capitalism, urban governance clearly involves a dynamic interplay of state authorities (e.g., local officials and urban planners), market forces (e.g., real estate developers and property managers), and social entities (e.g., residents and territorial civic organizations). Yet, the few epidemiological studies that have considered urban governance (Burris et al., 2007; Muennig, 2014; Shen, 2014) only treat it as being prescribed a priori by policy makers, and focus on the existing repertoire of strategies (e.g., ways of resource mobilization and regulatory measures) used by local authorities to promote healthier urban life. The relational nature of urban governance and its ramification on health have not been sufficiently theorized. This study considers neighborhood conflicts as another form of stressors for depressive symptoms. From a relational perspective,
neighborhood conflicts are conceptualized as contentious expressions of dyadic interplay between urban residents and other entities of urban governance (e.g., property managers, local authorities and real estate developers). It should be noted that the neighborhood conflicts analyzed here are different from the established measures of neighborhood disorder/disadvantage from the poverty literature, such as crime, the concentration of single-parent families, vandalism, and graffiti (Browning and Cagney, 2002; Sampson et al., 1999), nor do they include domestic violence within a family, daily complaints between neighbors, or routine strife among residents. By emphasizing a relational definition of neighborhood conflicts, this study argues that urban governance is not merely a set of rules made and enforced by these who are capable of imposing their will on the will of others. Instead, urban governance is experienced, realized, and made by interconnected entities, each with its own interests in urban space. The presence of neighborhood conflicts implies that existing rules of urban governance (material or abstract, traditional or transitional, moral or political, important or trivial, broad or narrow) are being contested, resisted or rejected by participants in urban governance, which may have ramifications for mental health. Today we see the process of urban transformation most vigorously in China, where cities are undergoing fundamental changes in less than a generation (Hsing, 2010). Owing to its sheer scale, China's great urban transformation can have a far-reaching impact on the global burden of mental disorders (Gong et al., 2012; Hsieh, 2015; Jin et al., 2012; L. Song, 2015). Given that the lower rates of depression reported
E-mail address: [email protected]. http://dx.doi.org/10.1016/j.socscimed.2017.10.035 Received 5 March 2017; Received in revised form 27 September 2017; Accepted 31 October 2017 0277-9536/ © 2017 Elsevier Ltd. All rights reserved.
Social Science & Medicine 196 (2018) 1–9
Q. Fu
Tomba, 2014; Vogel, 1989; Wu, 2002). This goal means that the state not only determines the landscape and planning of the city but also makes rules that govern the everyday life of urban residents (Friedmann, 2007). As noted by Wu (2001), state control over urban space has been weakened in the reform era: with the rise of the private sector, market transformation has created resources and personnel beyond the direct reach of state authorities. Given that decision-making power has been decentralized to the local/grassroots level, it becomes increasingly difficult and sometimes impossible for state authorities to dictate what urbanites should or should not do. To restore state control over urban space, the community-building campaign has several distinct political imperatives (Bray, 2006). Grassroots governments with defined territories have been endowed with a clear institutional identity; staffed with professional cadres; provided financial resources; and empowered to intervene neighborhood issues, using a wide repertoire of political instruments. A recent cross-sectional study argues that, through the building of community capacity, the community-building campaign promoted the communal grassroots organizations and amenities, which subsequently led to fewer depressive symptoms (Shen, 2014). A follow-up commentary corroborated this enthusiasm for community building and argued that this policy solution to health disparities, if proven to be effective, should be advocated worldwide (Muennig, 2014). However, the argument that community building promotes communal resources is questionable because the latter can be the consequence of other sociospatial changes, such as the rise of territory-based civic organizations, healthcare reforms, or simply the fact that all boats rise on China's urban-transformation tide. Without key measures reflecting the possible downside of the Chinese model of urban governance (e.g., neighborhood conflicts), it is too early to conclude that a policy aiming to strengthen the Party rule at the grassroots level can improve mental health. Moreover, the optimistic view of community building and its ramifications for neighborhood well-being is actually at odds with conclusions drawn by scholars with a keen interest in China's urban transformation (e.g., Bray, 2006; Fu and Lin, 2014; Hsing, 2010; Read, 2003; Tomba, 2014). As noted by Tomba (2005: 948-950), even inherently market-oriented relations such as dealing with PMCs are subject to government interventions, which frustrates urban residents, undermines residential satisfaction, and leads to direct confrontation. Moreover, power relations in Chinese urban neighborhoods are shown to discourage, let alone to promote, civic engagement and communal grassroots organizations (Fu and Lin, 2014). We thus have two conflicting pictures: in the view of a few medical sociologists and epidemiologists, community building promotes mental health, whereas the majority of human geographers, urban sociologists, political scientists, and cultural anthropologists see community building as leading to neighborhood unrest and contentious actions (e.g., Bray, 2006; Fu and Lin, 2014; Hsing, 2010; Read, 2003; Tomba, 2014). This study contributes to this debate by critically examining neighborhood conflicts and their association with depression from an urban-governance perspective.
previously tend to suggest denial of depression among Chinese (Parker et al., 2001), scholars need to understand the broader cultural, sociopolitical and historical context of China's urban transformation in order to identify meaningful stressors that explain the onset of depression. Since the late-1990s, China's central government has initiated a nationwide community-building campaign to restore state control over urban space, which leads to a further restructuring of the triangular relations among state, market, and society. Being officially recognized as “a process that is guided by the Party and the state, …and that harnesses community resources for the purposes of strengthening community functions” (Ministry of Civil Affairs, 2000), the purpose of this community-building model of urban governance is to convey the message that state authorities will do anything in its power to maintain governable urban space (Bray, 2006; Fu and Lin, 2014; Tomba, 2014). From an urban-governance perspective, we next explore how community building comes into being and its link with neighborhood conflicts. We often observe market (e.g., developers and property managers) or social (e.g., civic territorial NGOs) forces in Western urban neighborhoods, but Chinese neighborhoods provide scholars a great opportunity to observe state forces expressed in vibrant power relations. Since this particular constellation of power relations has rarely been addressed by previous studies on urban health, this article is among the first to critically examine the mental-health consequence of contentious urban politics. 2. Background 2.1. Community building as a way of urban governance Urban governance refers to the process through which local authorities, in conjunction with private entities, seek to achieve overarching goals in the city (Jessop, 2002; Pierre, 1999; Wu, 2002). To coordinate the interests of organizational or individual entities, this political process defines, reproduces and transforms relations among various entities in the urban space. In China, these entities involved in urban governance mainly include residents, local/grassroots governments, property management companies (PMCs) and real estate developers (Fu and Lin, 2014; Read, 2003; Tomba, 2014). Grassroots governments, including street offices (jiedao ban) and residents committees (juwei hui), compose subdistrict branches of local governments. They extend the roots of state power into urban neighborhoods through direct interactions with citizens (Lee and Zhang, 2013). While PMCs and developers appear to be nongovernmental, they often maintain various connections with governments due to reasons such as state ownership of enterprises, shared economic interests in land development, or personal networks. Drawing on evidence from cities in North America and Europe, Keating (1991) maintains that urban governance in different sociospatial contexts can be assessed along two aspects: democratic participation and managerial efficiency. Following this line of reasoning, Pierre (1999) eloquently defines four models (or, more precisely, ideal types) of urban governance according to their overarching goals—the managerial, corporatist, progrowth, and welfare models—and argues that real-world urban governance can resemble one or a mixture of the four models. Yet these frameworks primarily derived from Western cities appear to be ineffective in conveying the rich substance of China's urban governance against the backdrop of urban transformation and the persistence of political power (Hsing, 2010; Tomba, 2014; Wu, 2002). Despite the rise of territorial civic organizations (Fu and Lin, 2014) and fierce market-oriented competition within and between Chinese cities (He and Wu, 2009), the conventional democratic and managerial aspects fail to provide a comprehensive instrument for assessing the Chinese model of urban governance. One overarching goal of urban governance in China is the maintenance of state control over urban space, which is featured by a specific constellation of power relations (Bray, 2005; Lee and Zhang, 2013;
2.2. Neighborhood conflicts and community building To understand the link between neighborhood conflicts and community building, we must explore fundamental sociopolitical changes in urban China and answer the following two questions. First, what institutional changes account for the emergence of neighborhood conflicts in a socialist state? Second, with these institutional changes, how does community building lead to a dramatic increase in neighborhood conflicts? The first question should be explored in the context of China's great urban transformation, or more precisely, from workplace (work unit, or danwei) housing to commodity housing. Marxist ideology views unequal homeownership as displaced class struggles between capitalists 2
Social Science & Medicine 196 (2018) 1–9
Q. Fu
3. Research hypotheses
and the working class, and fails to recognize the significance of housing consumption in economic growth (Saunders, 1984). Before the housing reforms in the 1980s, workplaces were the all-encompassing microfoundations of communist rule bestowing economic benefits, social identity, and political rights on their employees (Bray, 2005). Grassroots politics in urban China were then determined by workplaces. As the owner of a housing compound, a workplace not only dictated the allocation of housing but maintained patriarchal authority over its employees who lived in the same workplace compound. Given that most, if not all, urban residents had to rely on their corresponding workplace for housing occupancy, they did not perceive themselves as property owners and played a trivial role in urban governance. However, this welfare housing system turned out to be a major fiscal burden to the socialist regime. Since the early 1980s the central Chinese government was determined to change urban housing from welfare benefits to commodities. After several major waves of housing reforms, most urban households have now achieved homeownership in two ways: acquiring existing workplace housing with price discounts, or purchasing commodity housing in the open market (Fu, 2016). To answer the second question, we must understand the differential effects of housing reforms on urban residents, workplaces, and local/ grassroots governments respectively. Because housing is no longer a privilege conferred by workplace affiliation, urban residents have gradually come to see themselves as housing consumers. Moreover, they also become stakeholders of communal properties, which is formally referred to as partitioned ownership (qufen suoyouquan). In a Chinese urban neighborhood the cost of communal properties (e.g., elevators, stairs, sidewalks, and parking space) is divided and then included in the sale price of each housing unit as shared areas (gongtan). As it stands, China's Real Right Law (National People's Congress, 2007: Article 76) stipulates that decisions pertaining to communal properties must be approved by at least one-half of homeowners in a neighborhood. With these institutional changes, Chinese homeowners are ready to assert their private/communal property rights and fight against violations of their rights. In addition, the demise of workplace-housing compounds also means that the socialist regime can no longer control China's vast urban space through workplaces. As predicted by Vogel at the beginning of China's urban transformation (Vogel, 1989), the regime must rely on territory-based governmental agencies to fill the power vacuum created by the declining influence of workplaces, which makes the community-building campaign necessary. One dilemma created by community building is ill-defined property rights. Although urban residents are now entitled to housing and communal property rights, community building ironically allows grassroots governments to intervene in neighborhood affairs and even allocate communal resources. For example, although homeowners' property rights over housing and communal properties are acknowledged by the Regulation on Realty Management, it also stipulates that homeowners and homeowners' associations should “support the residents' committee to carry out the work, and subject themselves to direction and supervision by the residents' committee” (Article 20: State Council, 2007 [2003]). The 2007 Revision of Regulation on Realty Management also explicitly specifies street offices as the supervising unit of civic homeowners congress/associations; this Revision is thus in line with the earlier Guidelines for Homeowners Congress (Ministry of Housing and Urban-Rural Development, 2003). Such ill-defined property rights across Chinese neighborhoods appear to account for a flood of urban unrest. A study by Tomba (2014) demonstrates that the deregulation of China's urban housing market has actually been accompanied by more unanticipated government interventions in residential neighborhoods. The peculiar mixture of housing commodification and community building then becomes a source of neighborhood grievances and conflicts. Claims of territorial legitimacy are constantly asserted, contested, and denied among entities of urban governance, which provides justification for the deployment of political coercion or collective resistance.
Although neighborhood conflicts are probably less intense than labor protests and deal with local rather than global issues, their relevance to depressive symptoms should not be underrated. In 2003, only five years after the official deregulation of China's urban housing market, problems related to housing and urban development ranked as the third-leading cause (after peasant burdens and mass layoffs) of all petitions received by the central government and accounted for 25 percent of all lawsuits in Zhejiang Province (Z. Song et al., 2003). Despite the scarcity of recent statistics, several qualitative studies suggest increasing neighborhood conflicts in China (Fu and Lin, 2014; Hsing, 2010; Tomba, 2014). Low residential mobility resulting from skyrocketing housing prices also makes the situation worse (S.-m. Li and Zhu, 2014). Residents have to live with long-lasting conflicts since moving seldom appears to be a viable solution. The perception of neighborhood conflicts is thus deemed as one stressor in urban China. Moreover, neighborhood conflicts inducing problem-solving actions are possibly correlated with depressive symptoms for two possible reasons. First, engagement in problem solving suggests the severity of neighborhood conflicts. Urban residents in a (post-)socialist state tend to stay away from civic engagement because of previous stressful experiences of mass mobilization, political pressure from state agencies, and reliance on private networks (Fu and Lin, 2014; Howard, 2003). Trivial or innocent neighborhood conflicts may not warrant engagement in problem solving and have little impact on mental health. Second, contentious actions aiming at solving neighborhood conflicts are often bounded by state-prescribed channels (e.g., petitions or mediation), which involve political pressure and appear to be arduous (Lee and Zhang, 2013; Perry, 2008). The first hypothesis is formulated as follows: Hypothesis 1. Net of other sociodemographic, relational, and environmental measures, both the perception of neighborhood conflicts and conflicts inducing problem-solving actions are significantly associated with more depressive symptoms. Nevertheless, an intriguing question is probably not whether neighborhood conflicts matter to depression but rather what aspect of neighborhood conflicts matters. According to Perry (2008), the Chinese notion of government legitimacy stems from its capacity to provide social security and address citizens' livelihood needs. Since community building paradoxically allows local/grassroots governments to intervene in neighborhood affairs and allocate communal resources belonging to urban residents, the ill-defined property rights and subsequent neighborhood instability caused by state intervention contradict the very origin of government legitimacy perceived by its own citizens. Since this dilemma resulting from community building is fueled by neighborhood conflicts, there has been an increasing public outcry for government accountability. Meanwhile, proximity to local authorities explains why other entities of urban governance, including developers and PMCs, can achieve power over residents. For example, once grassroots governments make PMCs responsible for the neighborhoods they manage, property managers can dictate the use of revenues generated from communal amenities (e.g., parking lots, gyms, clubs, swimming pools) regardless of collective decisions made by residents (Fu and Lin, 2014; Tomba, 2014). Given that the predominant role of state power in shaping neighborhood affairs is restored by community building, the state's presence is also experienced as a constraining influence on perceiving and solving neighborhood conflicts. For example, conflicts with developers or PMCs and related problemsolving actions do not necessarily trigger the deployment of state tactics for maintaining grassroots stability but these with local/grassroots governments do (Fu and Lin, 2014; Tomba, 2014). The repertoire of tactics then revolves around intimidation, coercion, and even the use of force meant to quell any (potential) threat to political stability (Lee and Zhang, 2013). Therefore, conflicts with local/grassroots governments 3
Social Science & Medicine 196 (2018) 1–9
Q. Fu
the dependent variable instead of a dichotomous variable of clinical depression.
are less likely to be translated to problem-solving actions and appear to be more stressful than these with developers or PMCs: Hypothesis 2. Among residents' conflicts with other entities of neighborhood governance, those with local/grassroots governments are most relevant to depression.
4.2.2. Sociodemographic variables Various sociodemographic variables, either with known consequences on depression or recognized as important indicators of social stratification in Chinese society, were considered (Bray, 2005; Hsieh, 2015; Jin et al., 2012; Parker et al., 2001; Yen et al., 2000): sex (female versus male); age (in years); marital status (currently married versus others); years of schooling; the logarithm of annual income (in 10,000 China Yuan); perceived socioeconomic status; unemployment (being employed, retired, or at school as the reference); housing tenure (homeownership versus tenancy); length of residence (in years); Communist Party membership; and state-sector employment (working in governments, institutes, and military versus others). Perceived socioeconomic status is an ordinal variable with five categories: lower class (coded as 1), lower-middle class, middle class, upper-middle class, and upper class (coded as 5).
Moreover, our understanding of neighborhood conflicts with governments cannot be isolated from the political economy of place. The manipulation of neighborhood planning, including changes in floorarea ratios (FARs), house demolition, and land expropriation, has been demonstrated as a lucrative way for local governments to acquire wealth at their own disposal (Fu and Lin, 2014; Hsing, 2010; Logan and Molotch, 2007). Communal properties (e.g., parking lots, shops, and outdoor advertising) can also generate substantial revenues given that Chinese urban neighborhoods are often sizable (Fu and Lin, 2014). Since community building in China provides an institutionalized way for those with political connections to interfere in neighborhood affairs, the enormous revenues generated by changes in neighborhood planning and communal properties are then used to consolidate their control over urban neighborhoods (Fu and Lin, 2014; Hsing, 2010; Tomba, 2014). Due to this coalition between political power and economic capital, these specific conflicts often perpetuate themselves and resist change, which generates feelings of loss of control and helplessness among urban residents.
4.2.3. Neighborhood conflicts A measure of neighborhood conflicts is developed in the survey. Prior to the survey, three focus groups, were formed comprised of local scholars, NGO leaders, neighborhood activists, and urban residents. The groups met to identify conflicts over a wide range of neighborhood issues and assess the questionnaires to be used in the survey. These conflicts were further organized by three entities of urban governance in China: developers, PMCs, and local/grassroots governments. Moreover, survey respondents were viewed as conscious and active agents making their lives instead of passive subjects overwhelmed by stressors (Thoits, 1994). They were asked whether they were aware of different types of neighborhood conflicts in the most recent five years, and if yes, whether they had participated in solving the conflict.
Hypothesis 2A. Among different subgroups of conflicts with governments, those relevant to neighborhood planning and communal properties are more likely to be associated with depression. 4. Data and methods 4.1. Data
4.2.4. Relational and environmental variables This survey used the position-generator method to measure network-based social capital (Lin and Dumin, 1986). Respondents were asked to report whether they knew someone in a list of 23 occupations (e.g., journalists, doctors, lawyers, janitors) and each occupation was assigned an occupational prestige score (C. Li, 2005). Three measures of social networks were constructed (L. Song and Lin, 2009): network extensity (the number of occupations accessed), upper reachability (the highest occupational prestige score accessed) and the range of occupational prestige scores accessed. As neighborhood acquaintances provide social leverage for gaining information, resources and influence (Briggs, 1998), another relational variable is the logarithm of the selfreported number of acquaintances (measured as people one can talk to about private matters) within a neighborhood. Four variables neighborhood attachment, informal neighborhood control, neighborhood cohesion and appraisal of neighborhood environment were constructed to measure neighborhood environment. Based on results from focus groups, survey items related to these variables were adapted and revised from existing measures (e.g., Lewicka, 2011; Sampson et al., 1999) so that they are clear to potential respondents in Guangzhou. Neighborhood attachment, consisting of both relational and environmental dimensions (Lewicka, 2011), was measured by six items (see Table 1). Respondents chose answers from a five-point scale ranging from “strongly agree” to “strongly disagree.” Survey items related to informal neighborhood control and neighborhood cohesion were adapted from research on collective efficacy (Sampson et al., 1999). For informal neighborhood control, respondents were asked about the likelihood (ranging from “very likely” to “very unlikely”) that intervention into several socially deviant behaviors (see Table 1) within their neighborhood would occur. Neighborhood cohesion was measured by eight survey items using a five-point Likert scale ranging from “strongly agree” to “strongly disagree” (also see Table 1 for these survey
This study is based on data from a face-to-face survey of thirty-nine urban neighborhoods in Guangzhou, China. The survey, conducted in from October 2012 to January 2013, adopted a multistage stratified random sampling method to recruit respondents. First, three primarysampling-unit (PSU) strata within the border of the outer-ring road of Guangzhou were specified by land-use purpose and population density: the inner core area, the inner suburb area, and the outer suburb area. Next, street offices within each stratum were selected by a GIS sampling procedure. Within each selected street office, one target neighborhood and a list of adjoining neighborhoods were obtained by GIS sampling. Local scholars or NGO leaders referred the survey team in order to access a target neighborhood. Despite the assistance from local universities and civic organizations, gatedness and intervention from property managers prevented data collection in one-third of the target neighborhoods; these were replaced by adjoining ones from the list. Within a selected urban neighborhood, households were selected by an interval sampling procedure according to the residential distribution of that neighborhood. Trained interviewers interviewed the household head or, if unavailable, his/her spouse or another adult family member familiar with the neighborhood. The survey's response rate was 65 percent; 1470 respondents from this survey are included in the analysis. 4.2. Measures 4.2.1. Dependent variable The full-length twenty-item version of the Center for Epidemiologic Studies Depression Scale (CES-D) was used to measure depressive symptoms in the last week, and the CES-D scores were calculated based on Radloff (1977). The Cronbach's alpha of the twenty CES-D items in the overall survey sample was 0.85. Given the low levels of depressive symptoms in China (Hsieh, 2015; Parker et al., 2001), most CES-D scores calculated stay below the clinical cut-off score of sixteen. Thus the total CES-D score (with a possible range from 0 to 60) was used as 4
Social Science & Medicine 196 (2018) 1–9
Q. Fu
Table 1 Principal component analysis of survey items related to neighborhood environment.a Neighborhood attachment
I feel attached to this neighborhood As a dwelling place, I like my neighborhood I do not want to move out of this neighborhood If I had to move out, I would miss people in this neighborhood If I had to move out, I would miss the neighborhood environment here My connection with other people in this neighborhood means a lot to me
Informal neighborhood control
Neighborhood cohesion
0.406 0.401 0.424 0.423 0.414 0.381
Vandalism of public property such as elevators and fire hydrants Damage to neighborhood environment (e.g., spitting and graffiti) Verbal or physical fights among residents
0.572 0.608 0.551
People in this neighborhood can get together to deal with problems This is a close-knit neighborhood People around here are willing to help each other I can count on others to pick up my mail, milk, and newspapers People in this neighborhood get along with each other People in this neighborhood can be trusted. People have very different opinions of neighborhood affairs a,b No one in this neighborhood cares much about what happens to others b
0.394 0.430 0.434 0.327 0.366 0.377 0.086 0.284
Location and transportation accessibility Landscape and architecture Sanitation service Responsiveness of property managers to individual requests Maintenance of public property Security service Parking management Parking availability Eigenvalue associated with the principal component Variance explained by the principal component a b
Appraisal of neighborhood environment
0.269 0.341 0.417 0.404 0.410 0.393 0.351 0.174 3.372 56.2%
2.098 69.9%
2.241 42.2%
3.520 44.0%
This survey item's effect on the principal component was removed due to its small factor loading. This item was reverse-coded.
items). The appraisal of the neighborhood environment, a key determinant of residential experience in urban China (Zhu et al., 2012), corresponds to residents' subjective evaluation of eight aspects of the neighborhood environment: location and transportation accessibility, landscape and architecture, sanitation service, responsiveness of property managers, maintenance of collective property, security service, parking management, and parking availability, using a five-point scale ranging from “very good” to “very bad.”
Depressionij = β0 +
∑ β Dp Dp + ∑ βSESq SESq + ∑ βSNr SNr p=3
+
q=8
r=4
∑ βEnvironments Environments + ∑ βConflictt Conflictt + u0j s=4
+ εij The three demographic variables (sex, age, and marital status) are represented by Dp; SESq refers to each of the eight (dummy) variables measuring socioeconomic status; the four variables related to social networks are denoted by SNr. Environments corresponds to the four variables measuring neighborhood environment; the key variables of this study, neighborhood conflicts between residents and other three entities of urban governance are denoted by Conflictt. βs represents the corresponding coefficients of these covariates. u0j is the residual random effect of community j, which is normally distributed with variance τ0. The error term εij is normally distributed with variance τε. Given that Breusch-Pagan tests suggest heteroscedasticity (results not shown), heteroscedasticity-robust standard errors are reported.
4.3. Method Statistical analysis was implemented in two steps. In the first step, principal component analyses (PCA) were conducted on the four sets of survey items related to neighborhood attachment, informal neighborhood control, neighborhood cohesion, and appraisal of neighborhood environment. The four principal components generated (with eigenvalues greater than 1) explain 56.2 percent, 69.9 percent, 42.2 percent, and 44.0 percent of the variances of the four sets of survey items respectively (see Table 1). A data-reduction procedure was employed to remove the influence of survey items with trivial loadings (Land et al., 1990). In the second step, random-effect linear models were used to assess factors associated with depressive symptoms. The full model can be expressed as:
5. Results The mean CES-D score of the sample is 9.3, which is consistent with previous findings of low prevalence rates of depression in Chinese populations (Hsieh, 2015; Parker et al., 2001; Yen et al., 2000). Less than one-half (43.6 percent) of respondents were male, and most of them 5
Social Science & Medicine 196 (2018) 1–9
Q. Fu
Table 2 Neighborhood conflicts by entities of urban governance (N = 1470).
Neighborhood conflicts Conflicts with developers Housing amenities Housing quality Size of shared residential areas Housing ownership certificates Conflicts with property management companies Parking Property-management fees Sanitation and gardening Appropriation of homeowners' collective revenues Utilities Harassment from security personnel Conflicts with local and grassroots governments Changes in urban planning nearby Changes in neighborhood planning Violations to partitioned ownership Intervention in the operation of homeowners associations Appropriation of homeowners' collective revenues Housing ownership certificates Intervention in the election of homeowners associations (HOAs)
Perceived
Engaged in Problem solving
Ratio: Engaged/Perceived
46.5% 22.9% 15.6% 10.6% 10.0% 4.2% 39.3% 21.5% 21.3% 15.7% 14.5% 10.6% 5.0% 15.0% 7.7% 6.7% 5.9% 3.4% 3.0% 3.0% 2.9%
18.1% 7.5% 4.5% 2.4% 2.7% 1.1% 13.5% 5.5% 7.5% 3.8% 4.0% 3.0% 0.5% 3.8% 1.7% 1.1% 1.1% 1.1% 0.5% 0.5% 0.9%
0.39 0.33 0.29 0.23 0.27 0.26 0.34 0.26 0.35 0.24 0.28 0.28 0.10 0.25 0.22 0.16 0.19 0.32 0.17 0.17 0.31
Table 3 Random-effect regression of CES-D scores on socio-demographic background, network and perceived neighborhood environment (N = 1470).a Model 1
Male Age Married Years of schooling Log of income Perceived SES Unemployed Homeownership Length of residence Party membership State employment Network extensity Network upper reachability Network range Log of number of acquaintances Neighborhood attachment Informal neighborhood control Neighborhood cohesion Appraisal of neighborhood environ. Neighborhood conflicts (perceived) Neighborhood conflicts (problem solving) Constant Random-effects parameter σ0 t Random-effects parameter σε t χ2 Log pseudolikelihood
Model 2
Model 3
Model 4
Model 5
Model 6
Coeff.
S.E.
Coeff.
S.E.
Coeff.
S.E.
Coeff.
S.E.
Coeff.
S.E.
Coeff.
S.E.
0.126 −0.012 −1.015* 0.150 −0.228 −0.631*** 2.447** −0.179 −0.012 −0.548 0.170
0.273 0.012 0.473 0.077 0.218 0.170 0.804 0.393 0.026 0.407 0.398
0.132 −0.013 −0.862 0.155 −0.236 −0.595** 2.410** −0.252 0.005 −0.489 0.223 0.007 −0.007 −0.013 −0.404**
0.276 0.012 0.490 0.081 0.212 0.173 0.795 0.394 0.024 0.392 0.376 0.054 0.008 0.011 0.121
0.217 −0.010 −0.940* 0.135 −0.295 −0.523** 2.225** −0.152 −0.002 −0.485 0.321
0.267 0.012 0.442 0.072 0.195 0.171 0.825 0.380 0.025 0.401 0.358
−0.184 −0.283* −0.242* −0.223*
0.111 0.109 0.113 0.091
0.212 −0.010 −0.878 0.142 −0.285 −0.508** 2.255** −0.209 0.005 −0.449 0.331 0.021 −0.007 −0.012 −0.182 −0.177 −0.259* −0.211 −0.229*
0.270 0.012 0.459 0.077 0.192 0.176 0.807 0.379 0.024 0.391 0.350 0.051 0.008 0.011 0.119 0.113 0.107 0.110 0.092
0.193 −0.008 −0.858 0.127 −0.270 −0.527** 2.248** −0.321 −0.005 −0.469 0.318 0.021 −0.006 −0.013 −0.198 −0.179 −0.270* −0.212 −0.213* 0.742*
0.272 0.012 0.453 0.076 0.195 0.179 0.846 0.383 0.024 0.384 0.352 0.051 0.008 0.011 0.117 0.114 0.109 0.111 0.092 0.287
0.196 −0.011 −0.813 0.120 −0.261 −0.509** 2.258** −0.290 0.001 −0.455 0.334 0.013 −0.006 −0.011 −0.214 −0.183 −0.278** −0.216 −0.214*
0.272 0.012 0.461 0.074 0.193 0.171 0.825 0.379 0.024 0.388 0.354 0.050 0.008 0.011 0.118 0.113 0.105 0.111 0.090
11.067*** 0.820***
1.179 0.208
12.007*** 0.789***
1.167 0.221
10.780*** 0.837***
1.193 0.173
11.495*** 0.820***
1.181 0.180
11.404*** 0.841***
1.150 0.178
1.055* 11.626*** 0.832***
0.467 1.146 0.168
5.504***
0.122
5.479***
0.120
5.400***
0.114
5.392***
0.115
5.379***
0.112
5.377***
0.112
75.34*** −4604.2
165.44*** −4596.8
116.26*** −4576.8
174.82*** −4574.2
167.15*** −4571.1
179.20*** −4570.4
Note: ***p < 0.001 **p < 0.01 * p < 0.05. a Robust standard errors are reported.
listed (5.1 ± 4.5). The mean upper reachability was 73.5, roughly about the prestige scores of nurses, entrepreneurs, and engineers in Chinese society (C. Li, 2005). The average range of prestige scores was 23.3, about the distance from restaurant waitresses to nurses, or from photographers to university professors. The average number of acquaintances within a neighborhood was 5.8 ± 13.6. Table 2 shows the prevalence of neighborhood conflicts by entities of neighborhood governance. Related to Hypothesis 2, urban residents were less likely to take action if they perceive local/grassroots governments instead of developers or PMCs as responsible for the same
were married and above age thirty (the mean age was 45.0 ± 14.7). Their mean years of schooling were 10.3 years, and annual income was about 73,000 yuan. The average score of perceived socioeconomic status was 2.3, which falls between lower-middle class and middle class. The unemployed only compose 2.3 percent of the sample. Most of the respondents were homeowners (79.7 percent), and the average years of residence were 7.2 years. About 20 percent of the respondents were Party members (20.7 percent) or working in the state sector (21.5 percent). In terms of network measures, respondents on average had access to about five occupations out of the twenty-three occupations 6
Social Science & Medicine 196 (2018) 1–9
Q. Fu
Table 4 Random-effect regression of CES-D scores on neighborhood conflicts (N = 1470).a Model 1
Male Age Married Years of schooling Log of income Perceived SES Unemployed Homeownership Length of residence Party membership State employment Network extensity Network upper reachability Network range Log of number of acquaintances Neighborhood attachment Informal neighborhood control Neighborhood cohesion Appraisal of neighborhood environment Neighborhood conflicts (perceived) with developers with property management companies with local and grassroots governments Neighborhood conflicts (problem solving) with developers with property management companies with local and grassroots governments Conflicts with governments (perceived) Changes in urban planning nearby Changes in neighborhood planning Violations to partitioned ownership Intervention in the operation of HOAs Appropriation of collective revenues Housing ownership certificates Intervention in the election of HOAs Conflicts with governments (problem solving) Changes in urban planning nearby Changes in neighborhood planning Violations to partitioned ownership Intervention in the operation of HOAs Appropriation of collective revenues Housing ownership certificates Intervention in the election of HOAs Constant Random-effects parameter σ0 Random-effects parameter σε t χ2 Log pseudolikelihood
Model 2
Model 3
Model 4
Coeff.
S.E.
Coeff.
S.E.
Coeff.
S.E.
Coeff.
S.E.
0.164 −0.007 −0.835 0.121 −0.268 −0.511** 2.298** −0.368 −0.004 −0.417 0.302 0.014 −0.006 −0.014 −0.215 −0.208 −0.262* −0.192 −0.204*
0.263 0.012 0.441 0.075 0.196 0.176 0.829 0.389 0.024 0.387 0.352 0.051 0.008 0.012 0.116 0.111 0.107 0.110 0.092
0.183 −0.010 −0.811 0.122 −0.265 −0.521** 2.323** −0.312 0.001 −0.420 0.323 0.010 −0.006 −0.012 −0.227 −0.194 −0.271* −0.205 −0.214*
0.273 0.012 0.454 0.074 0.192 0.170 0.821 0.378 0.024 0.396 0.350 0.048 0.008 0.011 0.124 0.111 0.105 0.109 0.094
0.162 −0.008 −0.828 0.125 −0.276 −0.488** 2.310** −0.297 0.001 −0.450 0.414 0.010 −0.005 −0.012 −0.207 −0.214 −0.257* −0.190 −0.210*
0.266 0.012 0.442 0.073 0.195 0.166 0.766 0.372 0.025 0.391 0.352 0.050 0.008 0.011 0.111 0.111 0.105 0.109 0.094
0.189 −0.009 −0.887 0.137 −0.299 −0.522** 2.340** −0.289 −0.003 −0.403 0.318 0.013 −0.007 −0.011 −0.223 −0.194 −0.270* −0.191 −0.228*
0.275 0.012 0.453 0.076 0.195 0.172 0.808 0.377 0.024 0.401 0.335 0.047 0.008 0.010 0.122 0.111 0.105 0.109 0.100
0.527 0.266 1.094*
0.471 0.336 0.477 0.740 0.284 2.067**
0.613 0.462 0.748 0.882 1.585* −0.849 1.076 −0.100 −0.486 1.588
0.546 0.687 1.030 1.194 1.121 1.043 1.107 2.396 1.250 4.109** −1.530 3.613 −1.096 1.088 11.619*** 0.868*** 5.339***
1.239 1.272 1.409 2.452 2.315 2.854 2.647 1.142 0.175 0.117
11.376*** 0.855*** 5.358***
1.160 0.175 0.113
174.18*** −4565.9
11.653*** 0.852*** 5.363*** 183.70*** −4567.0
1.131 0.170 0.113
11.397*** 0.810*** 5.355*** 288.66*** −4563.9
1.153 0.182 0.114
223.63*** −4560.9
Note: ***p < 0.001 **p < 0.01 * p < 0.05. a Robust standard errors are reported.
were included in Model 4. Both informal neighborhood control and appraisal of neighborhood environment had significantly negative associations with depressive symptoms, and these negative associations remained significant when relational measures were included (p < 0.05 in Model 4). Net of other sociodemographic, relational, and environmental measures, the perception of neighborhood conflicts and conflicts inducing problem-solving actions were significantly associated with more depressive symptoms (p < 0.05 in both Model 5 and Model 6). This finding lends support to Hypothesis 1. Regression analysis shown in Table 4 further considers different types of neighborhood conflicts. Among neighborhood conflicts with three entities of urban governance, only those with local/grassroots governments were significantly associated with more depressive symptoms (p < 0.05 in Model 1). This conclusion remained the same if perceived conflicts were replaced by conflicts inducing problem-solving actions (p < 0.01 in Model 2), which supports Hypothesis 2. Models 3
type of conflicts: about one-third of perceived conflicts with developers and PMCs led to problem-solving actions; in contrast, only one-quarter of conflicts with local/grassroots governments resulted in problemsolving actions. The lower ratios between perceived conflicts with governments and related problem-solving actions also exist for subgroups of conflicts, such as problems related to housing ownership certificates and appropriation of homeowners' collective revenues. The first set of random-effect regression models considers the influence of sociodemographic background, social networks, and neighborhood environment on depressive symptoms (see Table 3). Across all six models in Table 3, perceived socioeconomic status was significantly and negatively associated with depressive symptoms, whereas unemployment was linked with more depressive symptoms. The only significant relational measure is the number of neighborhood acquaintances (p < 0.01 in Model 2), but its significant association with depression disappeared when variables of neighborhood environment 7
Social Science & Medicine 196 (2018) 1–9
Q. Fu
understanding urban governance, which is inherently path dependent and context specific, as something given merely blinds us to the complexity of politics and thereby helps justify the status quo. According to Harvey (1978), urban spatiality is simultaneously a presupposition, a medium, and a product of ever-evolving social relations. Since urban governance is constrained by its social, historical, and cultural contingencies, a “one measure fits all” approach, regardless of the specific research context, is probably less productive. In this regard, the community-building campaign does not substantially deviate from previous top-down, authoritarian, and campaign-like policies in socialist regimes, which determine the well-being of people according to their proximity to political power. Community building will probably continue to consolidate the state's rule over urban space, but its multifaceted effects on mental health warrant further investigation.
and 4 in Table 3 further investigate the link between subgroups of government-oriented conflicts and depressive symptoms. For perceived conflicts with governments, the variable changes in neighborhood planning had a positive and significant association with depression (p < 0.05 in Model 3); for government-oriented conflicts inducing problem-solving actions, the variable violations to partitioned ownership showed a significant association with depression (p < 0.01 in Model 4). In other words, these government-oriented conflicts that are significantly associated with more depressive symptoms were all related to neighborhood planning or communal properties, which supports Hypothesis 2A. Across all models in Table 4, the results pertaining to sociodemographic background, social relations, and neighborhood environment remained virtually the same as those reported in Table 3. 6. Discussion
Acknowledgements Drawing on data from a community-based survey, this article investigates important sociopolitical changes over China's urban space and their implications on depression. It finds that: 1. net of other sociodemographic, relational, and environmental measures, neighborhood conflicts are significantly associated with more depressive symptoms (Hypothesis 1); 2. among neighborhood conflicts with three entities of urban governance, only those with local/grassroots governments have a significant and positive association with depressive symptoms (Hypothesis 2); 3. if we further examine subgroups of conflicts with governments, those conflicts associated with more depressive symptoms appear to be related to neighborhood planning and communal properties (Hypothesis 2A). The fact that China's new urban space in the era of globalization, decentralization and marketization could be restored to a governable space subject to political power testifies authoritarian dominance at its height. Exposure to neighborhood conflicts is stressful, but equally stressful is an understanding that these conflicts are deeply embedded in urban governance. Under the community-building model of urban governance, many came to believe that neighborhood conflicts are the rule rather than the exception (Fu and Lin, 2014; Tomba, 2014). The limitations of this research should be acknowledged. First, because both the exact timing of conflicts and trajectories of depressive symptoms are absent in this cross-sectional survey, no definite causal conclusion can be drawn. The CES-D score only reflects a one-week snapshot of depression and is thus highly context dependent. Moreover, reverse causality from depression to conflicts may exist if individuals with mental disorders are more (or less) likely to solve neighborhood conflicts. Second, the proposed measures of neighborhood conflicts are far from exhaustive. A severity scale of neighborhood conflicts would also be useful in future research. Finally, given that the sample was collected in one Chinese city, scholars should be cautious in applying these conclusions to other urban settings with different structures of governance and built-environment characteristics (e.g., gatedness, size of the neighborhood, housing affordability, and amount of communal properties). Scholars tend to stay away from a Foucauldian approach in searching for policy solutions to social problems, fearing that the analysis of the specific power relations at stake could compromise the generalizability of policies being pursued. This study instead suggests that our understanding of mental illness can be advanced by a closer investigation into urban politics. One striking feature of research on mental health is the degree to which urban politics has been neglected. Earlier thinkers such as Max Weber, Karl Marx, and Bertrand Russell have meticulously demonstrated power as a fundamental concept in the social sciences. Today we often focus on established proxies for power (e.g., education, income, and social class) but fail to examine how power can be expressed in more subtle ways. Since urban residents are always positioned within a web of power relations, the urban-governance perspective allows us to view the urban space as a spatial manifestation of power and further reveal the politics of mental illness. Yet
The author gratefully acknowledges financial support from The Chiang Ching-kuo Foundation for International Scholarly Exchange and The Hampton New Faculty Award at The University of British Columbia. References Bray, D., 2005. Social Space and Governance in Urban China: the Danwei System from Origins to Reform. Stanford University Press, Palo Alto, CA. Bray, D., 2006. Building ‘community’: new strategies of governance in urban China. Econ. Soc. 35, 530–549. Brenner, N., 2004. Urban governance and the production of new state spaces in Western Europe, 1960–2000. Rev. Int. Political Econ. 11, 447–488. Briggs, X.d.S., 1998. Brown kids in white suburbs: housing mobility and the many faces of social capital. Hous. Policy Debate 9, 177–221. Browning, C.R., Cagney, K.A., 2002. Neighborhood structural disadvantage, collective efficacy, and self-rated physical health in an urban setting. J. Health Soc. Behav. 43, 383–399. Burris, S., Hancock, T., Lin, V., Herzog, A., 2007. Emerging strategies for healthy urban governance. J. Urban Health 84, 154–163. Friedmann, J., 2007. Reflections on place and place-making in the cities of China. Int. J. Urban Regional Res. 31, 257–279. Fu, Q., 2016. The persistence of power despite the changing meaning of homeownership: an age-period-cohort analysis of urban housing tenure in China, 1989-2011. Urban Stud. 53, 1225–1243. Fu, Q., Lin, N., 2014. The weaknesses of civic territorial organizations: civic engagement and homeowners associations in urban China. Int. J. Urban Regional Res. 38, 2309–2327. Gong, P., Liang, S., Carlton, E.J., Jiang, Q., Wu, J., Wang, L., et al., 2012. Urbanisation and health in China. Lancet 379, 843–852. Harvey, D., 1978. The urban process under capitalism: a framework for analysis. Int. J. Urban Regional Res. 2, 101–131. He, S., Wu, F., 2009. China's emerging neoliberal urbanism: perspectives from urban redevelopment. Antipode 41, 282–304. Howard, M.M., 2003. The Weakness of Civil Society in Post-communist Europe. Cambridge University Press, Cambridge. Hsieh, N., 2015. Economic security, social cohesion, and depression disparities in posttransition societies a comparison of older adults in China and Russia. J. Health Soc. Behav. 56, 534–551. Hsing, Y.-t., 2010. The Great Urban Transformation: Politics of Land and Property in China. Oxford University Press, Oxford, UK. Jessop, B., 2002. Liberalism, neoliberalism, and urban governance: a state–theoretical perspective. Antipode 34, 452–472. Jin, L., Wen, M., Fan, J.X., Wang, G., 2012. Trans-local ties, local ties and psychological well-being among rural-to-urban migrants in Shanghai. Soc. Sci. Med. 75, 288–296. Keating, M., 1991. Comparative Urban Politics: Power and the City in the United States, Canada, Britain, and France. Edward Elgar Publishing, Aldershot, UK. Land, K.C., McCall, P.L., Cohen, L.E., 1990. Structural covariates of homicide rates: are there any invariances across time and social space? Am. J. Sociol. 95, 922–963. Lee, C.K., Zhang, Y., 2013. The power of instability: unraveling the microfoundations of bargained authoritarianism in China. Am. J. Sociol. 118, 1475–1508. Lewicka, M., 2011. Place attachment: how far have we come in the last 40 years? J. Environ. Psychol. 31, 207–230. Li, C., 2005. Prestige Stratification in the Contemporary China: occupational prestige measures and socio-economic index. Sociol. Res. 74–102. Li, S.-m., Zhu, Y., 2014. Residential mobility within Guangzhou city, China, 1990–2010: local residents versus migrants. Eurasian Geogr. Econ. 55, 313–332. Lin, N., Dumin, M., 1986. Access to occupations through social ties. Soc. Netw. 8, 365–385. Logan, J.R., Molotch, H.L., 2007. Urban Fortunes: the Political Economy of Place. University of California Press, Berkeley, CA.
8
Social Science & Medicine 196 (2018) 1–9
Q. Fu
capital, comparative reference group, and depression in two societies. Soc. Sci. Med. 136, 117–127. Song, L., Lin, N., 2009. Social capital and health inequality: evidence from Taiwan. J. Health Soc. Behav. 50, 149–163. Song, Z., Zhou, G., Cui, L., 2003. Urban Redevelopment: the Third Cause of Petitions. Xinhua News Agency, Beijing People.com.cn. State Council. (2007 [2003]). Regulation on Realty Management. Beijing: Document No. 379. Thoits, P.A., 1994. Stressors and problem-solving: the individual as psychological activist. J. Health Soc. Behav. 35, 143–160. Tomba, L., 2005. Residential space and collective interest formation in Beijing's housing disputes. China Q. 184, 934–951. Tomba, L., 2014. The Government Next Door: Neighborhood Politics in Urban China. Cornell University Press, Ithaca, NY. Vogel, E.F., 1989. One Step Ahead in China: Guangdong under Reform. Harvard University Press, Cambridge, MA. Wu, F., 2001. China's recent urban development in the process of land and housing marketisation and economic globalisation. Habitat Int. 25, 273–289. Wu, F., 2002. China's changing urban governance in the transition towards a more market-oriented economy. Urban Stud. 39, 1071–1093. Yen, S., Robins, C.J., Lin, N., 2000. A cross-cultural comparison of depressive symptom manifestation: China and the United States. J. Consult. Clin. Psychol. 68, 993–999. Zhu, Y., Breitung, W., Li, S.-m, 2012. The changing meaning of neighbourhood attachment in Chinese commodity housing estates: evidence from Guangzhou. Urban Stud. 49, 2439–2457.
Ministry of Civil Affairs, 2000. Circular on the Promotion of Community Building in Urban China. Beijing. Ministry of Housing and Urban-Rural Development, 2003. Guidelines for Homeowners Congress. Beijing. Muennig, P., 2014. What China's experiment in community building can tell us about tackling health disparities. Soc. Sci. Med. 107, 217–220. National People's Congress, 2007. Real Right Law of the People's Republic of China. Beijing. Parker, G., Gladstone, G., Chee, K.T., June 2001. Depression in the planet's largest ethnic group: the Chinese. Am. J. Psychiatry 158 (6), 857–864. Perry, E.J., 2008. Chinese conceptions of “rights”: from Mencius to Mao—and now. Perspect. Polit. 6, 37–50. Pierre, J., 1999. Models of urban governance: the institutional dimension of urban politics. Urban Aff. Rev. 34, 372–396. Radloff, L.S., 1977. The CES-D scale: a self-report depression scale for research in the general population. Appl. Psychol. Meas. 1, 385–401. Read, B.L., 2003. Democratizing the neighbourhood? New private housing and homeowner self-organization in urban China. China J. 31–59. Sampson, R.J., Morenoff, J.D., Earls, F., 1999. Beyond social capital: spatial dynamics of collective efficacy for children. Am. Sociol. Rev. 633–660. Saunders, P., 1984. Beyond housing classes: the sociological significance of private property rights in means of consumption†. Int. J. Urban Regional Res. 8, 202–227. Shen, Y., 2014. Community building and mental health in mid-life and older life: evidence from China. Soc. Sci. Med. 107, 209–216. Simmel, G. (Ed.) ([1903] 1964). The Metropolis and Mental Life. New York: Free Press. Song, L., 2015. Does who you know in the positional hierarchy protect or hurt? Social
9