Opportunity-enriching environments or isolating gated communities? Assessing the chances of socio-economic integration of the populations of four shantytowns of Salvador, Brazil,

Oportunidades ampliadas ou condomínios fechados excludentes? Avaliando as chances de integração socioeconômica das populações de três bairros populares de Salvador, Brasil

Abstract

Grounded on semi-structured interviews, we seek to examine the impact of neighborhood effects on the individuals’ living conditions in four shanty-towns of Salvador, Brazil, addressing the question of under what conditions the proximity to affluent gated communities fosters their socio-economic integration. The research demonstrates that the relationship between spatial proximity and socio-economic integration is conditioned by the capacity of public space to promote (non)employment cross-class interactions, the impact of crime, and the gated communities' degree of securitization. Whereas in Calabar, large opportunities of socio-economic participation in its surroundings mitigate the negative impact of neighborhood effects, (non-)employment relationships sharply decline in the less centrally located Vale das Pedrinhas and Bate Facho, where the informal proletariat has been excluded from using the public space for commercial activities. The construction of the highly isolated gated community Alphaville II has neither fostered cross-class interactions nor benefitted the economic integration of the Vila Verde inhabitants. In all neighborhoods physical boundaries have been internalized by a similar discourse that emphasizes class-hierarchized opportunities for upward socio-economic mobility, particularly regarding the access to schools and public security. The study urges to reflect on a more holistic approach to social inequalities, comprising socially more integrative labor and housing policies.

Keywords
urban poverty; segregation; neighborhood effects; gentrification; Salvador; Brazil

Resumo

Com base em entrevistas semiestruturadas, examina-se o impacto do efeito-território nas condições de vida dos moradores de quatro bairros populares de Salvador, Brasil. Em particular, indaga-se sob quais condições a proximidade a condomínios fechados de elite produz sua integração socioeconômica. O estudo demonstra que a relação entre proximidade espacial e integração socioeconômica se vê condicionada à capacidade de o espaço público facilitar as articulações (não)empregatícias entre os grupos socialmente distantes, ao impacto do crime e ao grau de isolamento físico dos condomínios fechados. Enquanto, no bairro de Calabar, as amplas oportunidades de participação socioeconômica proporcionadas pelos condomínios fechados vizinhos mitigam o efeito-território, as relações (não)empregatícias se reduzem drasticamente no caso dos bairros menos centrais Vale das Pedrinhas e Bate Facho, onde o proletariado informal está impedido de utilizar o espaço público para atividades comerciais. A construção do condomínio fechado altamente isolado Alphaville II nem favoreceu as articulações entre os grupos socialmente distantes, nem beneficiou a integração econômica dos moradores do bairro Vila Verde. Em todos os bairros analisados, as fronteiras físicas foram internalizadas por um discurso similar que enfatiza o acesso hierarquizado às oportunidades de ascensão socioeconômica, nomeadamente no que tange ao acesso à escola e à segurança pública. O estudo urge refletir sobre uma aproximação mais holística às desigualdades sociais, orientada por políticas de trabalho e de habitação socialmente mais integrativas.

Palavras-chave
pobreza urbana; segregação; efeito-território; gentrificação; Salvador; Brasil

Introduction

Research on neighborhood effects has a long-standing tradition in US American scholarship and has influenced the academic debate about urban poverty and racial segregation since the 1990s, while giving important impulses for desegregation policies (Sampson, 2012; Wilson, 1987). Grounded on the assumption of opportunity-enriching environments, the spatial proximity to the middle class induced by desegregation policies is supposed to (i) confer a greater degree of internal diversity to the poor individuals’ social networks; (ii) strengthen the community’s capacity for informal social control; (iii) promote a higher exposition to middle class peer groups, because of the shared use of local infrastructure and (iv) improve their access to high-quality local urban services, such as health care and education (Galster; Killen, 1995).

Unlike the US American context, Latin American cities have always hosted spatial settings characterized by the proximity between affluent neighborhoods and shantytowns, which can be found throughout the central regions and the post-war middle and upper class vectors of expansion (Borsdorf; Hidalgo; Vidal-Koppmann, 2015). In many cases, the spatial proximity to affluent neighborhoods had entailed locational advantages for the poor populations’ economic integration, such as the possibility to work as maids, housekeepers, and watchmen (Caldeira, 2000). However, the shantytowns' territorial stigmatization in conjunction with the highly segmented access to urban services created obstacles to non-employment related cross-class interactions (Bayón; Saraví, 2013; Ribeiro et al., 2010).

Convergent trends in urban development show that the recent construction of wealthy gated communities in peripheral areas predominantly inhabited by the lower echelons tend to reproduce these spatial settings (Caldeira, 2000; Janoschka; Sequera, 2016; Prévôt-Schapira; Pineda, 2008). Based on the concept of gentrification, scholars have increasingly investigated whether these developments might dilute the structures of segregation following the center-periphery antagonism and therefore widen the opportunities for the lower echelons’ socio-economic integration (Janoschka; Sequera, 2016; Ruiz-Tagle, 2016; Sabatini; Salcedo, 2007).

In this research, we shall inquire into how far (non)employment-related relationships permeating socially distant groups living in spatial proximity differ when comparing those spatial settings where gated communities were recently built next to poor neighborhoods with those more centrally located settings of vicinity, where the affluent and poor have been living in geographical closeness over decades. Based on semi-structured interviews, we aim at investigating the impact of neighborhood effects on the living conditions¹ of the populations of four shantytowns embedded in affluent areas of Salvador, Brazil's fourth biggest city. In particular, we seek to examine under what conditions the proximity to affluent neighborhoods mitigates the impact of structural disadvantages concentrated at neighborhood scale. In this sense, we aim at contributing to the still incipient debate on neighborhood effects in Latin America and to the ongoing discussion on gentrification. Urban scholarship has analyzed the impact of gentrification on the metropolises’ socio-spatial organization (Borsdorf et al., 2015; Janoschka; Sequera, 2016; Prévôt-Schapira; Pineda, 2008) and has shown what happens inside these upper-class enclaves (Blakely; Snyder, 1997; Caldeira, 2000; Low, 2004), but hasn’t dedicated the same efforts to examine the impacts on their surrounding poor populations’ employment opportunities, access to urban services, patterns of sociability and perception on segregation.

The article is composed of four sections, excluding this introduction and the conclusion. The first section delineates the main developments made in the discussion on neighborhood effects, in dialogue with research on gentrification in Latin America. The scope will be limited to the review of US American and Brazilian literature on neighborhood effects. The second section introduces the study’s methodological framework, whereas the third section exposes the results obtained within the qualitative fieldwork, which will be discussed in the fourth section.

Current state of neighborhood effects research

In the US context, the debate about neighborhood effects was instigated by Wilson’s structural approach to poverty (Sampson, 2012). Alternatively, Massey and Denton (1993) investigated the secular mechanisms of racial residential segregation reproduced within the private real estate market and public housing policies and elucidated the relationship between the spatial concentration of poor black people and place-based racial discrimination.² According to Wilson (1987), the poor Afro-American ghetto populations’ confinement to the local social context hampered their access to information on job vacancies. By the same token, the omnipresent contact to peer groups affiliated to the local drug-trafficking was accounted to give false incentives to adhere to informal pathways of economic integration, while devaluating school as an opportunity for upward social mobility (Wilson, 1987).

On the basis of an extensive literature review, Small and Newman (2001) suggested two sets of models designed to explain how neighborhood poverty affects life chances: (i) socialization mechanisms (comprising the epidemic model, collective socialization model, institutional model, linguistic isolation model, relative deprivation model, oppositional culture model), broadly addressing the question of how neighborhoods socialize young residents; and (ii) instrumental mechanisms (network isolation model, resource model and limitation of political alliances model), which focus on how individual agency is limited by neighborhood environment, mostly in the case of adults.

Scholarship examining the link between neighborhood ecological characteristics and crime rates has emerged as a prominent research strand within the discussion about neighborhood effects: Sampson (2012) showed that the impact of neighborhood effects can be attenuated by the community's capacity for collective efficacy, a concept defined as the combination of, on the one hand, working trust and social cohesion and, on the other hand, shared expectations of informal social control. Accordingly, collective efficacy might reduce the incidence of crime and therefore mitigate neighborhood effects; vice versa, crime and violence undermine the community's collective efficacy.

Largely influenced by the debate about neighborhood effects, different desegregation and social mixing policies have been implemented in US metropolises (Small; Feldman, 2012). However, scholarship has demonstrated that these desegregation policies may also magnify the tensions between the socially distant groups living in the same or in adjacent neighborhoods, given the higher visibility of the socio-economic cleavages and the segmented access to health care and education (Clampet-Lundquist; Massey, 2008). Small (2004) showed that the poor inhabitants of Villa Victoria, an ethnic enclave located next to Boston’s upper-middle class South-End, maintain only few social contacts with their affluent neighbors. Moreover, the author argues that the local availability of social, commercial, and cultural infrastructure fosters their social isolation, because of the absence of any necessity to access extra-local (non)material resources, except for the workplace.

Research carried out in different peripheral neighborhoods of Latin American metropolises has corroborated a negative impact of neighborhood effects on the poor individuals’ income levels, the access to labor markets and the type of employment relationships (Roberts; Wilson, 2009; Ribeiro; Lago, 1999). Kaztman and Retamoso (2006) point to the failing transmission of middle-class role models and to the declining possibilities of cross-class relationships at school, because of the bifurcation of the educational system into public and private schools. Major divergencies can be attested when turning to the question of whether the spatial proximity to affluent neighborhoods widens the poor individuals’ chances of socio-economic integration. In Latin America, the emergence of these sociospatial settings has been addressed by the literature on gentrification, a process which in this context does not only imply the eviction of poorer households living in the decaying inner-city areas, but which also refers to the construction of affluent gated communities at the urban fringes (Borsdorf et al., 2015). Janoschka and Sequera (2016) distinguish three research strands that synthesize the debate on gentrification in Latin America:

1. scholarship analyzing the impact of the middle and upper classes’ patterns of migration on the spatial organization of contemporary cities. These developments are critically examined in the context of an increasing privatization of urban space, which, in conjunction with strong police controls, create obstacles to the (non)commercial use of public space by the informal proletariat;

2. research centered on the implementation of urban restructuring programs financed by national real estate companies or foreign investors, which explore the rent gap of declining central districts or semi-peripheral areas;

3. scholarship analyzing the impact of pacification policies implemented in poor neighborhoods located in highly valorized areas, which, in conjunction with public-private investments in urban services and the regularization of homeownership, contribute to the rise of living costs for the local populations and therefore trigger processes of expulsion.

Studies focusing on the living conditions of the gated community dwellers have uncovered similar tendencies of social avoidance as already observed in the US American urban context (Blakely; Snyder, 1997; Low, 2004): gated-communities foster local intra-class interactions, weaken non-employment cross-class interactions and contribute to their residents’ withdrawal from public space (Bayón; Saraví, 2013; Caldeira, 2000). However, qualitative oriented research exploring (non-)employment-related cross-class relationships in these spatial settings has challenged these assumptions. Sabatini and Salcedo (2007) demonstrate that the construction of gated communities in the urban peripheries in Santiago, Chile, has widened the lower echelons’ opportunities of functional and symbolic integration.

According to Salcedo and Torres (2004), positive externalities encompass the improvement of local living conditions, increasing job opportunities, alleviation of the social stigma of living in a poor neighborhood, and the increase in the value of their land. On the other hand, Ruiz-Tagle (2016) reveals that in the peripheral neighborhood of La Florida, Santiago de Chile, both employment- and non-employment-related cross-class relationships are constrained by the territorial discrimination of the poor population and the segmented access to urban services such as (pre)schools and leisure activities. The gated community employees are primarily recruited from other more distant and apparently safer neighborhoods, a fact that annihilates the benefits of spatial proximity.

In Brazil, the focus of research has been placed on consolidated patterns of spatial proximity between socially distant groups, which can be encountered both in central regions and within the vectors of middle- and upper-class expansion (Ribeiro, 2016). Urban scholarship points, on the one hand, at the widened structures of opportunities of economic integration due to the better employment options offered in the gated communities, and at the financial engagement of the surrounding affluent gated community dwellers in local cultural, educational and vocational programs (Almeida; D’Andrea, 2004; Hita; Gledhill, 2009). On the other hand, Andrade and Silveira (2013) highlight the functional character of cross-class interactions between affluent community dwellers and their surrounding poor populations in Belo Horizonte, and emphasize the persistence of social hierarchies within the employment opportunities, given the subalternate character of the personal service jobs. Studies conducted in similar spatial settings of Rio de Janeiro corroborate the absence of cross-class interactions, except for the employment relationships. This is due to the segmented access to public transport, the educational system, hospitals, and recreational space, and especially because of the shantytowns’ territorial stigmatization (Ribeiro et al., 2010).

Marques (2010), assessing the impact of poverty and residential segregation on the economic achievements of individuals in São Paulo, posits that a greater proportion of primary ties in personal networks interferes negatively in the individual’s income level, whereas the embeddedness in institutional spheres of sociability facilitates labor market integration. Though corroborating the scarcity of cross-class social interactions, he argues that segregation and poverty do not exert an independent effect on their inhabitants’ economic integration. Ribeiro (2008) asserts that the geographic proximity between the favelas and the affluent Zona Sul neighborhoods in Rio de Janeiro doesn’t eliminate the engrained relationships of social domination: as a result, cross-class interactions seldom take place, given the maintenance of the social hierarchies reified in the unequal distribution of economic, social and symbolic capital.

For the purpose of our research, we follow a holistic approach to neighborhood effects based on a relative conception of poverty, by integrating a wide set of intervening factors into our methodological framework, which, thus far, have been investigated separately, such as: economic integration, patterns of sociability and self-perception on segregation.

Methodological framework and studied areas

In our research we recur to Häußermann’s three-dimensional methodological scheme in order to assess the impact of neighborhood effects on the four neighborhoods. The material dimension refers to the locational (dis)advantages arising from the neighborhood’s distance to labor markets and to the deficiencies in terms of urban services. The social dimension comprises the analysis of the individuals’ mobilization of social networks to obtain (non-) material resources and the influence of peer groups on processes of socialization. We further examine the communities’ capacity of collective efficacy based on Sampson’s (2012) methodological framework, a component not integrated in Häußermann’s original proposal, but which becomes particularly relevant for our study given the high incidence of crime in Salvador. Finally, the symbolic dimension assesses the negative impact of the neighborhood’s territorial stigmatization on the individuals’ living conditions. The social dimension deserves our particular attention, given the fact that social processes are commonly regarded as one of the main mechanisms of neighborhood effects (Sampson, 2012).

(Non-)employment-related interactions between socially distant groups living in spatial proximity will be scrutinized on the basis of two indicators. In order to address the “degree of the neighborhoods functional autonomy”, we shall recur to Hannerz’ (1980) distinction between five functions that shape the individuals’ social reproduction: (i) family, (ii) the provision of material, social and cultural resources, (iii) leisure activities, (iv) neighboring and (v) transport/mobility. We assume that the “degree of functional autonomy” varies between, on the one hand, the availability of all five functions in a given neighborhood, providing less incentives to access extra-local resources – like already observed by Small (2004); on the other hand, the absence of these functions – except for family and neighboring, considered as intrinsically local functions. This latter situation predicts a greater orientation towards the surrounding neighborhoods within the individuals’ daily trajectories and patterns of sociability, provided that these localities facilitate the access to high quality urban services and employment opportunities. In this sense, we assume that the “degree of functional embracement”, understood as the possibilities of the individuals’ socio-economic participation in their affluent surrounding, declines in those spatial settings characterized by the absence of public spaces capable of fostering social encounters and the selling of non-durable goods.

As shown in Figure 1, Salvador can be divided into four macro-regions, following socioeconomic parameters: the middle class Centro, the middle and upper class Orla Atlântica – both concentrating the majority of employment opportunities and high-quality urban services – the lower class Subúrbio Ferroviário/Ilhas and the lower(middle) class Miolo Urbano.

Figure 1
Spatial organization and vectors of middle and upper classes expansion in Salvador

Figure 1 also highlights three vectors of middle and upper classes expansion: the first phase of expansion (1940-1960) corresponds to their migration from the central areas towards the adjacent south-western coastline; the second phase (1960-2000) shows an occupation of the urban space between Avenida Paralela and Avenida Octávio Mangabeira; the third phase (2000 until now) is characterized by the construction of gated communities on both sides of Avenida Paralela in the direction of the international airport.

Figure 2 shows the spatial distribution of household income levels disaggregated at census district level and points at a strong central-periphery antagonism, where the affluent populations are concentrated along the Centro and the Orla Atlântica regions.

Figure 2
Spatial distribution of household income in Salvador

Figure 2 also indicates that, when narrowing the scale to neighborhood level, the Centro and Orla Atlântica areas are interspersed with poor enclaves, among others, our four research sites. The unequal spatial distribution of household income coalesces along strong disparities regarding the proportion of household heads holding a university diploma, as shown in Figure 3:

Figure 3
Spatial distribution of household heads holding a university diploma in Salvador

We can observe the same patterns of proximity between socially distant groups, concerning educational levels, within the Orla Atlântica and Centro areas.

Figure 4 illustrates the spatial distribution of the number of homicides along police districts (Áreas Integradas de Segurança Pública - AISPs) drawn from the Department of Public Security of the State of Bahia for the year 2018.

Figure 4
Spatial distribution of homicides in Salvador

Figure 4 shows a strong tendency of spatial concentration of homicides in both peripheral regions of Subúrbio Ferroviário and Miolo Urbano, a fact particularly relevant for the assessment of neighborhood effects in Salvador's peripheral areas.

The research was conducted in the shantytowns Calabar, Vale das Pedrinhas, Bate Facho and Vila Verde, extended across the municipality of Salvador. These neighborhoods paradigmatically represent the spatial settings of geographic proximity between socially distant groups in a central, a semi-central, a semi-peripheral and a peripheral setting (cf. Figure 1).

Except for Vila Verde, all shantytowns share a common trajectory: their demographic growth stems from the massive influx of families coming from Bahia's impoverished hinterland between 1940 and 1970; the process of informal land occupation mostly occurring in Salvador’s less accessible valleys, thalwegs and hillsides was further accelerated by the increasing demand by the surrounding affluent neighborhoods for cheap workforce, particularly in the construction business and in the personal services sector. Located at the fringes of Miolo Urbano, the recently built Alphaville II gated community reflects a more recent trend in Brazilian suburbanization: large scale gated communities are now increasingly being constructed in poor urban peripheries, where cheap lands are still available and where real estate companies can achieve higher profits.

Calabar’s origins may be traced back to the 1950s, when migrants arriving from Bahia’s impoverished hinterland settled in the empty valleys of the neighborhood and the hillsides belonging to the Catholic Church. Vale das Pedrinhas and Bate Facho closely followed the expansion of middle and upper classes neighborhoods in the late 1970s towards Salvador's shoreline (Vale das Pedrinhas) and Av. Paralela (Bate Facho). A considerable portion of the privately-owned housing stock, mostly composed of two or three-story autoconstructed dwellings, is located in areas of high environmental risk of flooding and landslides, which occur periodically during the rainy seasons, particularly, in the cases of the hill and valley-shaped Calabar and Bate Facho. Given their strategic location amidst Salvador’s middle and upper-classes neighborhoods, the populations of Calabar, Vale das Pedrinhas and Bate Facho were forced to fight against attempts by large real estate companies to take over the land. In tandem with the city-wide rise in crime rates and increasing police controls, these land conflicts have exacerbated the tensions between the rich and the poor living in geographic closeness.

Though living in spatial proximity, physical boundaries maintain the social distance between the poor and the affluent in all four settings: gated and fenced entrances, watchtowers and private security officers patrolling the adjacent streets of the gated communities impede the access of non-residents. Often, the built environment, such as the major thoroughfares Avenida Centenário (Calabar) and Av. Juracy Magalhães Júnior (Vale das Pedrinhas), sewer channels (Vale das Pedrinhas) and natural barriers (Vila Verde), serves as spatial distance markers. The highest degree of spatial isolation can be encountered in the case of the highly protected Alphaville II, separated by a tropical forest and by a large entrance road only suitable for cars. Yet, in the case of the shantytowns, these boundaries acquire a more symbolic character, such as a bar, a bus station or an abandoned building, markers that are easily recognized by their residents. However, despite the absence of physical barriers or armed gang-members controlling the shantytowns’ entrance, non-residents only rarely venture into their neighborhood, according to the interviewees.

Based on the 2010 decennial statistics published by the United Nations Development Programme, Table 1 draws the economic and socio-demographic profile of the census districts (Unidades de Desenvolvimento Humano - UDH), to which the four neighborhoods belong:

Table 1
Economic and socio-demographic profile of the studied neighborhoods

STUDIED AREAS	UDH Calabar / Alto das Pombas	UDH Vale das Pedrinhas/ Nordeste de Amaralina	UDH Bate Facho	UDH Vila Verde
Socio-demographic/economic profile
Spatial setting	central	semi-central	semi-peripheral	peripheral
Surrounding middle/ upper class neighborhoods	Barra, Ondina, Graça	Horto Florestal, Candeal	Imbui	Alphaville II
Population size	10,152 hab. (Calabar: 6,484)	32,825 hab. (Vale das Pedrinhas: 5,162)	1,399 hab. (~5,000 estimated)	8,823 hab.
Average per capita income	R$ 492,16	R$ 780,70	R$ 330,00	R$ 365,26
Enrolled in higher education	4.88%	22.78%	1.40%	0.98%
Social vulnerability index
Child mortality estimation per 1,000 live-births	19.05	12.7	30.2	27. 6
Proportion of individuals vulnerable to poverty	32.30%	24.86%	54.72%	50.97%
18+ person without elementary school degree and employed in the informal labor market	25.40%	21.58%	42.61%	41.89%
Proportion of children aged 0-5 not at school	40.66%	48.06%	49.13%	56.21%
Proportion of individuals aged between 15 and 24 which neither study nor work	9.89%	7.16%	17.08%	41.89%
Proportion of mothers without primary education with minor children, among the total number of female-headed families	35.86%	27.27%	50.29%	61.88%
Proportion of extremely poor children	7.07%	5.15%	16.11%	11.51%
Homicide rate in AISP 2018
	AISP 14 Barra: 3	AISP 15 Nordeste: 27	AISP 09 Boca do Rio: 31	AISP 12 Itapuã: 120

According to Table 1, the peripheral neighborhoods of Bate Facho and Vila Verde show considerably higher homicide and social vulnerability levels than their more central counterparts. In these locations, we therefore hypothesize to find a stronger impact of neighborhood effects on the populations’ living conditions. For our study, we carried out semi-structured interviews in each neighborhood between November 2018 and May 2019.

After the conduction of a first test sample following an open questionnaire structure, we established a limit of 20 persons per neighborhood. This procedure allowed us to further develop the key-aspects of the final questionnaire³, which was then applied in the case of the 80 interviewees. We sought to select an equanimous number of interviewees representing three life cycles: below 18 years old, between 18 and 65, and above 65 years old – in order to attend to neighborhood effect heterogeneity and to facilitate the comparison between the four neighborhoods. The individual-level attributes synthesized by Table 2 served as control variables, particularly relevant for network analysis.

Table 2
The interviewees' socio-demographic profile

Neighborhood	Calabar	Vale das Pedrinhas	Bate Facho	Vila Verde
Gender
Masculine	40%	45%	45%	40%
Feminine	60%	55%	55%	60%
Age and Income
under 18 years old	50%	45%	60%	45%
between 18 and 65 years old	30%	35%	30%	45%
above 65 years old	20%	20%	10%	10%
income under 1405 US$	20%	35%	80%	70%
income above 1405US$	80%	65%	20%	30%
Unemployment rate
	20%	30%	55%	35%
Educational level
with high school degree	40%	30%	10%	35%
without high school degree	60%	70%	90%	65%

The lower unemployment rates observed in the Calabar and Vale das Pedrinhas sample, paired with higher income and educational levels, stand in a marked contrast with the peripheral neighborhoods Bate Facho and Vila Verde. As a common trait of all four shantytowns, a large part of the interviewees work in the informal labor market, catering to both the local and extra-local demand in (non)durable goods, such as fruits, vegetables, drinks and snacks; on average, only 18% declared to be formally employed.

The interviews conducted in the four shantytowns were transcribed and then submitted to discourse analysis. In this case, we extracted valuable information from the interviews, which helped us to further develop our empirical analysis and to deepen our understanding on the impact of neighborhood effects in the four socio-spatial settings. The semi-open structure of the interviews allowed us, on the one hand, to cover a broad range of topics, showing a significant variability according to the interviewees’ life-cycle, and, on the other, to identify common structures within their social construction of reality (Luckmann; Berger, 1991). After the interviews’ transcription, these topics were re-aligned to the sample questionnaire in order to allow a direct comparison between the four neighborhoods.

We further conducted fieldwork observations, including participation in several meetings held by the neighborhoods’ counselors, which provided important insights into the neighborhoods’ socio-institutional organization.

Assessing neighborhood effects in four shantytowns of Salvador

Material dimension

The study reveals that in all four neighborhoods deficiencies prevail in the local access to schools, child-care centers, health care units and recreational facilities. Likewise, we attested the scarcity of employment opportunities, except for Vale das Pedrinhas, which hosts a wide range of employments, albeit in the informal labor market. Given these constraints, virtually all interviewees declared that they had to leave their neighborhood for work, studies, and leisure activities. A strong variability could be confirmed concerning the populations’ “degree of functional embracement”. In Calabar, employment opportunities arise from the wealthy neighbors’ demand for personal services. Furthermore, the respondents enumerated the vast opportunities emerging from the informal labor market, which mostly consist in informally selling merchandises in the streets along strategic points of nearby beaches and at the entrances of the apartment towers.

Contrasting with this stable demand-supply relationship, the majority of Vale das Pedrinhas interviewees stated that the “degree of functional embracement” vis-à-vis the close-by affluent neighborhoods has decreased in recent years, owing to (i) a decline in the demand for personal services as a result of the financial crisis; (ii) the option for job recruitment strategies based on the premise “neighborhood’s safety and external image first” at the expense of the factor “geographic proximity”; (iii) changes in the employees’ contracts which leads to a rise in employers’ salary expenses, discouraging formal hiring of housekeepers.

The Vale das Pedrinhas case also exemplifies that, unlike Calabar, the proximity to the affluent neighborhoods does not automatically entail locational advantages for the informal street trade, given the prevalence of the residential-only land use of Horto Florestal and Candeal and the mechanisms of vigilance and security isolating the apartment towers. The decline of the “degree of functional embracement” becomes particularly manifest when examining the revitalization process of the former Ceasinha do Rio Vermelho – an open market located amidst Vale das Pedrinhas and Horto Florestal – where the shantytown’s inhabitants used to sell their regional products. Public-private joint investments have converted this place, renamed to Mercado do Rio Vermelho, into a high-end complex of restaurants, bars, and luxury grocery stores, which excluded Vale das Pedrinhas dwellers from socio-economic participation. According to the interviewees, the former Ceasinha assumed a pivotal role in the intermediation to jobs along informal conversations with the affluent neighbors, which no longer take place at the Mercado do Rio Vermelho.

In Bate Facho, the interviewees reported great difficulties to encounter a gainful job, even in the informal labor market, since the neighborhood lacks a commercial infrastructure capable of employing the local residents. Furthermore, the interviewees complained that the affluent Imbui inhabitants exploit them as ready-available and cheap workforce, alleging that their overall precarious living conditions forces them to accept any job offer. This exploitative relationship often coalesces with practices of discrimination, combined with strong racial bias, as reported by those working in the gated communities as housekeepers and maids:

Yes, I worked at one of these Imbui condominios. But I soon was surprised about the low salaries they paid; I even make more money selling snacks at the bus stops. But what really made me leave is that they always suspected me, a Black woman, of theft and that I supposedly had a bad influence on their children

(Rose, 25 years old, street vendor).

Rose, who had arrived in Bate Facho together with her family when she was six, further stated that her interactions with her employer, a 60 year-old lawyer, had always been tense, given the constant disagreements about her salary and non-paid extra-hours. But confidence further deteriorated with the spread of drug-trafficking in her neighborhood. As rumors of theft spread throughout Imbui, nobody wanted to hire her anymore. According to Rose, the chances of a direct contact with the residents of the close-by gated community of Imbui outside employment relationships are very low now: working as a street vendor means to spend most of the time at the bus and metro stations dispersed along the Av. Parallela, two means of transport rarely used by the middle and upper classes.

Similar claims about the exploitation of cheap workforce surfaced in the Vila Verde case, where the interviewees often complained about low wages paid to clerks, watchmen and maids at the close-by hotels, the Hangar Business Park and the Norte Shopping, which tend to recruit their workforce from the neighborhood because of their “punctuality” and “readiness” to accept low salaries, as Marco, a 47 years old construction worker observes. Most noticeable, and in sharp contrast with Sabatini and Salcedo’s (2007) findings, is the fact that none of the interviewees declared to be currently working in the nearby gated community of Alphaville II. When applying for personal services jobs, most interviewees faced the territorial stigmatization of their neighborhood, which impeded them from taking the job. According to Márcio, the absence of social contact also stems from the fear of crime, which must be attributed to the fact that the gated community is embedded in a peripheral area notorious for gang-related drug trafficking:

Look where they built this gated community! I mean, most of them settle down in those already upper-class areas, Patamares, Pituba, but this one is completely isolated, no way to get into contact with them. When we once showed up in front of their entrance, they got really scary, they'd thought we were invading their space

(Márcio, 45 years old bus driver).

Márcio, who formerly played an important role as a community leader in Vila Verde, narrates his frustrated expectations when applying for a job as security guard in the gated community Alphaville II. In his opinion, this is far more isolated from its geographic surroundings than any other more central upper-class neighborhood in Salvador. Fifteen years ago he lived in Santa Cruz, an impoverished shantytown located in-between Vale das Pedrinhas and the upper-class Rio Vermelho, where he had no difficulty to find a job as a watchman and where one day he found himself sitting right next to his bosses’ son in a bar of the bohemian Rio Vermelho neighborhood.

In the four analyzed samples, non-employment related cross-class relationships were either scarce or completely absent, as demonstrated in the Vila Verde sample. Strong segmentation tendencies could also be observed in the shared use of the playground facilities located in the Avenida Centenário. Though accessible to both social groups, the interviewees stated that the gated community dwellers didn’t want their children to play with Calabar’s children. Despite the absence of physical barriers of the built environment and the rather open pattern of Imbui’s housing structure, we could not find any substantial interaction via the informal street trade during our visits to Bate Facho. This lack of cross-class interactions coincides with the absence of public spaces susceptible to foster the physical contact between the Imbui and Bate Facho dwellers, except for Praça do Imbui, which hosts a wide range of bars and restaurants. Discriminatory practices could be observed in the shared use of recreational facilities for kids located within the Imbui boundaries. This place has become the object of constant conflicts caused by the Imbui inhabitants’ attempts to impede the access of Bate Facho children by means of walling.

Social dimension

Within Calabar sample, we observed a strong tendency of extra-local orientation among interviewees towards the close-by affluent neighborhoods. This dynamic ascribes a higher degree of diversity and territorial dispersion to the interviewees’ social ties, therefore increasing their likelihood to obtain information about job opportunities. The access to social leverage networks could be predominantly confirmed within the sample of interviewees whose daily commutes between, either home and workplace, or home and school, widened their interaction radius. The access to information about job opportunities is also conditioned by the interviewees’ contact with local keypersons already employed in the gated communities, in particular janitors and watchmen. Assuming a pivotal function for job referrals in Calabar, they were frequently informed by their employers when there was a demand for housekeepers, electricians and gardeners. Concerning the interviewees’ mobilization of social networks to obtain (non-)material resources, a certain shift within the primary structures of reciprocity and support was attested. Accordingly, this fact stems from a lack of confidence in their neighbors and the all-encompassing impact of the local drug traffic on former between-neighbors’ relationships.

This decline in between-neighbors reliability had been partially compensated by the affiliation to dense associative networks within the Protestant churches and neighborhood councils, which promote a wide array of (non-)material resources, like food supply, parental counseling, small loans and emotional support. In Vale das Pedrinhas, we detected a predominant affiliation to the primary spheres of sociability, family and friendship, even when controlling for income and educational levels, occupational status and age. As Pedro comments, a major reason for this withdrawal towards more intimate and trustworthy networks stems from the constraints arising from the dispute between rival drug gangs for the territorial control of Vale das Pedrinhas.

Pedro, a 52 years old carpenter, nostalgically recalls former times when he built his house together with his also recently arrived neighbors and with the help of his cousins. As far back as he can remember, the poorest families always received some support from the neighborhood council and the Catholic church. Once everybody had finished to build one’s home and got a job, mostly in the construction business, his neighbors stopped helping each other, so that “now, everybody is on one’s own”, as he laments. When, in 2014, one of the city’s notorious drug-lord decided to move to his neighborhood trying to seize the control from the rival gang, two of his cousins got involved and soon started to steal money to buy drugs:

Pedro is not alone in his complaints about weakening trust in kinship ties: half of Vale das Pedrinhas interviewees stated that the spread of drug trafficking had forced them to break up with friends, neighbors and even parents involved in illicit activities, mostly given the fear of retaliation from the local drug lords. Given the prevalence of local kinship and friendship networks, the interviewees commented to face serious difficulties to access information about extra-local job opportunities, particularly in the formal labor market. The social network structure of Bate Facho interviewees reveals the highest degree of localism within the four analyzed neighborhoods, corroborating a strong correlation between, on the one hand, high levels of crime and social vulnerability, and, on the other, a truncated social network structure, which hampers their access to information about job vacancies (Wilson, 1987). In contrast to Calabar, the lack of local employment opportunities, schools and leisure activities does not yield a stronger extra-local orientation in the interviewees’ social networks, given the reduced possibilities of socio-economic participation in the surrounding residential-only neighborhood of Imbui.

We observed a clear predominance of primary ties promoting social cohesion and the stabilization of the economic situation in a short run. However, in this case, primary ties do not act as leverage networks for the economic integration, because of the lack of local job opportunities, the high rates of unemployment – provoking the individuals’ dissociation from work-place based networks – and the lack of material resources. Still, virtually all the respondents confirmed receiving and providing support in food supply and in situations of emergency. This becomes particularly evident in the case of the flooding of 2008, when the whole population helped each other, as observed by Michelle:

At least we have a strong support network here. When we had these flooding everybody went to the main street and offered help to his neighbor, who lost his house or furniture. I think that’s one of the good things here, since we are a poor community, everybody helps out

(Michelle, 26 years old, unemployed).

Michelle, a divorced mother of three, draws many resources from her long-lasting relationships with her neighbors, such as child-caring, food, a ride to the supermarket and emotional help, and extends her helping hand whenever she can. Attracted by the vacancy as a housekeeper in the close-by gated community of Imbui, she settled down in Bate Facho in a precarious self-built dwelling, when she was 15, and still compares her social life to her early adolescence in Elisio Medrado, a poor town located in Bahia’s hinterland where she was born. For her, Bate Facho – despite being a poor community – has preserved much of its rural character, especially given its relative spatial isolation provided by the tropical forests of the Pituaçu Park. According to Michelle, the local residents behave much less “urban” and “self-interested” than her friends and relatives living in the more central parts of Salvador, where she misses the “open door mentality” of her neighbors and the everyday chats with her friends.

This observation, confirmed by many other interviewees commenting on the 2008 flooding, illustrates that in a poor community such as Bate Facho, where everybody is struggling to get along with few resources, strong support networks between neighbors still play an important role when compared to the other physically more consolidated shantytowns, where the mutual support between neighbors has decreased once the community gained access to basic infrastructure and urban services.

The absence of non-profit organizations or philanthropic associations capable of promoting positive social references, together with the lack of opportunities in formal work relations and low educational levels, are considered enticements for involvement in illicit activities in these neighborhoods. As already observed by Sampson (2012), the high incidence of criminal activities and homicides tend to weaken the mechanisms of informal social control, like monitoring and punishing the adolescents’ truancy and idleness, due to the fear of victimization and mistrust vis-a-vis unknown co-residents.

This becomes particularly evident in the case of Vale das Pedrinhas, where gang-related drug trafficking has created parallel structures of social and institutional organization. According to the interviewees, their daily routines were seriously disrupted by violent disputes over the territorial control of the local drug market, thus causing temporary shut-down of local commerce and paralysis of public transport and schools. Like in Vale das Pedrinhas, in Vila Verde, we observed a significant destabilization of the community social organization due to increasing violence and crime, which, in turn, reduces the interviewees’ probability to intervene on behalf of the community’s well-being. These difficulties to confront criminal groups contributed to a more aggressive intervention of the local police. The weak capacity of collective efficacy found in Vale das Pedrinhas and Vila Verde sharply contrasts with Calabar, where the respondents stressed the importance of the crime-preventing interventions conducted by, on the one hand, the BCS, the Protestant churches and the community library, and, on the other, community leaders.

Symbolic dimension

In Calabar, strong evidence was found for enduring processes of the population’s territorial stigmatization, despite the improvements made in terms of public security and community's engagement to improve the neighborhood's deteriorated public image. Most of the interviewees recur to different counterstrategies to disassociate from the image of a crime-ridden shantytown. These strategies mainly consist in distinguishing between what they see as, on the one hand, the group of hard-working and ambitious inhabitants and, on the other, the group of inhabitants pursuing illegal activities or relying on cash-transfer programmes for their daily living. This becomes particularly manifest in the following interview:

I think in our days you have to separate between those struggling and willingly to work and those lazy ones who receive the Bolsa Familia. I think it's particularly them who deal drugs, don't go to school and, well, have no ambitions. Me and my family, we were always hard workers

(Mário, a 37 years old, construction worker).

For Mário, Calabar was always associated with drug-dealing, a fact that formerly caused great indignation among the “hard-working families”. When he was in his early twenties, he participated in several public demonstrations against police violence in Calabar, shutting down public transport on the Av. Centenário and prompting the mayor to invest in local infrastructure, health care and education. However, according to him, “the cash transfer programmes have made people less ambitious”, and also less committed to common goals within the community. With strong Christian beliefs, he “stopped fooling around with his old school mates and wasting money on the beach”, when his third son was born and he seriously started to worry about money. For him, “to get a better life, a father has to work hard and to be responsible for his kids, get them a good education, be a role model, but here many apparently have other beliefs…”.

In Vale das Pedrinhas, the interviewees pointed at different forms of discrimination in the job-searching processes, but also in the biased treatment by schoolteachers and policemen. This observation also holds for the Bate Facho case, where the territorial stigmatization creates obstacles for access to certain services – taxi drivers categorically avoid the neighborhood because of the supposedly high homicide levels. Strategies to circumvent the neighborhood's stigmatization consist in indicating a different residential address, like Joana confirms:

I put Imbui on the application form when I am trying to get a job. Most of the people don't even know Bate Facho, but when they know it, they say: oh that's where the murderers live, oh you live there? So, it really helps because the ZIP code here is the same, I put Imbui on it

(Joana, a 24 years old, maid).

Also in the case of Vila Verde, we could attest the existence of a series of stigma, which affects the local population economic integration and which relates to the negative image of the surrounding crime-ridden neighborhoods of Mussurunga and São Cristovão. Additionally, routine police controls have increased since the affluent Alphaville II inhabitants’ arrival and are perceived as a serious threat, since the policemen don't differentiate between who “looks like a thief” and who is a “hard working guy”, as Eliane, a 54 years old housekeeper laments.

As far as the interviewees' subjective perception of segregation is concerned, their sense of “belonging to” or being “excluded” from Salvador’s affluent areas is directly correlated to their chances of socio-economic integration and, owing to the declining demand for personal services by their affluent neighbors, to their possibilities to use public space for (non)commercial activities. Subsequently, we encountered the highest degree of territorial identification in the case of Calabar, where “nobody moves out without a good reason” (Ana, a 18 years old maid), given its strategical proximity to the city's formal/informal labor markets. This perception of widened opportunity structures sharply declines in the less centrally located shantytowns embedded in residential-only wealthy environments, where the inhabitants have been gradually excluded from using the public space for commercial activities.

In all four neighborhoods, physical and symbolic boundaries have been internalized by a similar discourse of the interviewees, which distinguishes different life worlds and class-hierarchized opportunities for upward socio-economic mobility, particularly regarding the access to schools and public security. Yet, in Calabar, the absence of ostensive security controls might explain the non-hostile perception of affluent gated-community dwellers. It needs to be emphasized that most of the Calabar interviewees have to cross the affluent neighborhoods in order to reach other parts of the city and to access (non-)material resources, thus promoting a frequent physical, albeit superficial, contact between the socially distant groups.

In Vale das Pedrinhas, a strong spatial separation is provided by the main roadway Avenida Juracy Magalhães Júnior and none of the interviewees declared to enter the surrounding wealthy neighborhoods for any purpose, except for working. Their gradual expulsion from the Mercado do Rio Vermelho is largely perceived as a hostile land take-over by the upper-class, a dynamic which has reinforced inter-neighborhood tensions and might be made accountable for the rise in assaults and thefts. For some interviewees, the transformation of former shared public spaces into high-end entertainment zones is perceived as a more general threat of being expelled from Vale das Pedrinhas and being forced to relocate to more peripheral neighborhoods, where they will become isolated from the labor markets and where they will be exposed to even higher crime rates.

In this context, the recent concession of property titles by the local government, has increased the local populations’ living costs because of the rise in monthly expenses for basic urban services and housing taxes. Some interviewees of Bate Facho and Calabar reported that, given this financial burden and the possibility to sell the house for an above-average price to large real estate companies, some of their relatives living right next door to the affluent apartment towers have decided to migrate to more peripheral neighborhoods.

According to the Vila Verde interviewees, the construction of Alphaville II had heightened the local population's expectations for better job opportunities in the personal service sector, as it had already occurred in the close-by Bairro da Paz. In this case, the poor population had greatly benefitted from the construction of the gated communities of Patamares, as documented by Hita and Gledhill (2009). Since these expectations were not met, Alphaville II became the object of constant complaints, regarding the biased police controls and the rise of drug trafficking. Even though the gated community is geographically separated by a tropical forest, its luxurious mansions make socioeconomic cleavages more visible and disseminate anger and jealousy among the poor population:

Before we only saw these high class condominios when going to the city, now they are right here and I get really angry because I know that they didn't really work for that, there are many corrupt businessmen who moved in. I also think that our whole area one day may turn to their hands, look at all the new road built over there, they are certainly not for us

(Augusto, a 27 years old watchman).

Augusto, who came to Vila Verde with his family when he was a kid, worries about the continuously rising prices for groceries, water and electricity. Only five years ago, he built his own house on a huge vacant lot located close to the entrance road of Alphaville II, which then “belonged to nobody”. Within the last two years, most of the remaining vacant lots has been used by the city’s public authorities and private investors to build new roads, large-scale supermarkets and the Salvador Norte Shopping. As rental prices skyrocketed, some of his friends who had planned to build a house too were forced to leave Vila Verde and settled down in the far-distant Boca da Mata, “where unemployment soars and crime is on the rise”, as he explains.

Sharing Augusto’s worries, the arrival of affluent families is mostly perceived as a hostile attempt of land take-over fueled by real estate speculation, which reflects the local governments plans to revitalize the neighborhoods located at the margins of the Avenida Paralela by virtue of public investments in infrastructure – such as the new metro station Bairro da Paz and the recent implementation of the Avenida 29 de Março.

Discussion

Inferring from our results, we can conclude that neighborhood effects show a direct correlation with the socio-economic profile of the neighborhoods’ populations (cf. Table 1) and therefore might exert a stronger influence in the peripheral neighborhoods Bate Facho and Vila Verde, especially given their considerable distance to Salvador’s central labor markets. However, further qualitative fieldwork needs to be carried out in order to verify whether neighborhood effects are directly related to the population's socio-economic profile, as initially hypothesized. It has to be emphasized that in all four neighborhoods, low educational levels and underperforming public schools turn out to be the limiting factor for the populations' long-term economic integration into the formal labor market; this impasse is exacerbated by the declining demand in the personal service sector, which formerly had guaranteed a stable income without requiring any further professional or educational qualifications. Our study shows that three intervening mechanisms might explain the variability of neighborhood effects in the four shantytowns.

The first mechanism uncovers shifts within the primary systems of social support, resulting from broader societal transformations, such as alterations in the family composition and a generalized decline of collective mobilization on behalf of common goals, but also from essentially spatial determinants, such as the impact of crime, high levels of poverty and unemployment. Showing major convergences with the US urban context of highly segregated ghettos (Sampson, 2012), the cases of Vale das Pedrinhas and particularly Bate Facho illustrate that the absence of institutional resources makes it particularly difficult for the local populations to participate in decision-making processes in dialogue with the local governments, for example, when requesting investments in urban services. On the one hand, this institutional vacuum left by the state was filled by evangelical churches, which promote a wide array of (non)material resources to their members; however, they fail to strengthen the community's capacity of collective efficacy and social cohesion since they engage in a strong competition with each other and disseminate a rather individual-centered model of upward social mobility. On the other hand, drug-trafficking related rival gangs exert a substantial influence on the community's social organization, while disseminating mistrust and fear among neighbors.

The second mechanism relates to the impact of crime on the individuals’ chances of economic integration, on their patterns of social interactions, on their self-perception, and on the neighborhood’s territorial image. Even though, we could not attest the diffusion of a crime-related ghetto-specific behavior, as documented by US American scholars (Sampson, 2012; Small; Feldman, 2012), virtually all interviewees, except for Calabar, declared to suffer from the all-encompassing influence of crime on their daily livings. The third mechanism relates to the populations’ territorial stigmatization. As already observed by Wacquant (2008), the strong association between social and spatial attributes in the assessment of the “other” may foster different forms of place-based discrimination. Virtually all interviewees reported having been victims of (race-based) discrimination either when applying for jobs, approaching the surroundings’ affluent apartment towers, or interacting with non-local residents.

As far as the locational advantages arising from the poor populations’ proximity to affluent neighbors is concerned, economic, social and symbolic interactions between socially distant groups living in spatial proximity have been theoretically framed by the concept of social integration by Chilenean scholars (Ruiz-Tagle, 2016; Sabatini; Salcedo, 2007). This approach can’t be uncritically transposed to the Brazilian context, given the strong physical separation of the affluent gated communities from their poor surroundings and the maintenance of engrained social and racial hierarchies by means of segmented access to urban services, both creating obstacles to the lower classes socio-economic integration. We therefore suggested to examine (non)employment cross-class relationships on the basis of the two indicators “degree of functional autonomy” and “degree of functional embracement”, which do not presuppose any a priori relationships of social integration. Inferring from our results, we might conclude that, in the case of the four shantytowns, the relationship between spatial proximity and socio-economic integration is conditioned by the capacity of the public space to promote (non)employment related cross-class interactions, the impact of crime, and the gated communities’ degree of securitization.

On the one hand, the Calabar case illustrates that the proximity to affluent neighbors provides large opportunities of socio-economic participation in those neighborhoods characterized by a mixed use of space – comprising residential, commercial and public functions – and by a lower degree of securitization and spatial isolation. In this context, inter-neighborhood relationships have grown based on superficial social encounters or employment ties during decades of vicinity. On the other hand, the factor spatial proximity loses its relevance for the lower echelons’ socio-economic integration into exclusively residential surroundings, like Alphaville II, characterized by a high degree of securitization and spatial isolation. Contrary to what Salcedo and Torres (2004) observed, the gated community’s implementation did not trigger positive externalities for the Vila Verde population – such as an increase in cross-class interactions, the transmission of middle-class role models or the valorization of the shantytown’s public image – nor did it entail positive trade-offs in terms of economic integration, as already observed by Caldeira (2000) in upper-class gated communities of São Paulo. Situated between these two extremes, the possibilities of socio-economic participation in Vale das Pedrinhas and Bate Facho depend on the permeability of class-boundaries at intermediary areas, which have been targeted by projects of revitalization reflecting the privatization of urban space adjacent to the gated communities. Furthermore, the chances of economic integration for the low-wage servant class have declined during the last years, owing to the increase in crime and the financial crisis.

Final considerations

Our research aimed at exploring the impact of neighborhood effects on the individuals’ living conditions in four shantytowns of Salvador, addressing the question of under what conditions the spatial proximity to affluent gated communities widens their opportunities of socio-economic integration. The study identified three mechanisms operating at neighborhood level, whose common structures, similar causal pathways but also internal variability might give important impulses for comparative research. The hypothesis of opportunity-enriching environments can only be corroborated in the case of Calabar, whereas the interviewees of the less central neighborhoods draw substantially less benefits from their geographical closeness to affluent surroundings, given the lack of public space capable of fostering social encounters and the informal street trade, the stronger impact of crime and the higher degree of securitization of the gated communities.

The case of Alphaville II illustrates that the construction of gated communities close to poor peripheral and highly segregated neighborhoods can leverage the expansion of the middle and upper class vector – in this case, along the Avenida Paralela – by draining large investments in infrastructure and urban services and by attracting private companies catering to its affluent residents. In Vale das Pedrinhas and Bate Facho, the combination of, on the one hand, public investments and urban revitalization programs targeting the shantytowns adjacent areas, and, on the other, the concession of property titles to their inhabitants has increased the poor populations’ living costs and contributed to their expulsion from public spaces, as largely documented by scholars focusing on gentrification (Janoschka; Sequera 2016).

In Brazil, despite mounting empirical evidence that both extreme wealth and poverty have become even more spatially concentrated in large metropolises since the neo-liberal shift of national economies in the 1990s, (qualitative) research on neighborhood effects still remains a desideratum (Roberts; Wilson 2009). Likewise, the debate about the implications of concentrated disadvantages in highly segregated neighborhoods on urban planning is still incipient and denotes a missing dialogue between empirical scholarship and public policies.

However, in order to elaborate a more sustainable solution to the concentration of structural disadvantages in poor neighborhoods it is crucial to go beyond place-based urban interventions, aiming at widening the opportunities at neighborhood level via investments in public services, and people-based interventions, by means of desegregation programs. Therefore, urban policies should follow a more holistic approach to social inequalities, comprising socially more integrative labor and housing policies, particularly considering that the individual and family costs of concentrated disadvantages are borne not exclusively by the lower echelons but by members of all classes.

• 1
  Neighborhood effects might constrain the individuals’ living conditions in innumerous ways, ranging from lower school performance and difficulties to find a job, over disproportionally higher out-of-wedlock and teenage birthrates, to the erosion of the community’s capacity for informal social control. These constraints are found to be particularly detrimental in the case of children and adolescents, whose daily routines predominantly take place within the boundaries of their own neighborhood (Small; Feldman, 2012; Sampson, 2012). Following Häußermann’s (2003) approach, in our study, the impact of neighborhood effects on the individuals’ living conditions will be addressed at the material dimension (income and educational levels, occupational status and physical health conditions), the social dimension (patterns of sociability and processes of socialization), and the symbolic dimension (perception of being integrated/excluded from broader society). In this sense, we seek to stress the causal interrelatedness of the three aforementioned analytical dimensions. However, given space limitations, our research mainly focusses on the social and institutional mechanisms that might account for neighborhood effects. Convergent trends encountered in all four research sites show that future studies should pay greater attention to the negative impact of territorial discrimination, i.e. the symbolic dimension, on the individuals’ chances of socio-economic integration and on cross-class interactions.
• 2
  In the Brazilian context, the emergence of black ghettos as a result of racially discriminatory practices institutionalized in the housing market can’t be confirmed. Nevertheless, socio-economic disparities inscribed into the cities’ spatial organization strongly coalesce along racial/ethnical lines, as already attested by Ribeiro and Telles (2000) for the case of Rio de Janeiro. Particularly in Salvador, with a strong legacy of slavery, the causal interrelatedness between these two factors becomes evident, taking into consideration that the city’s black and brown population still earns less than half (730,22 reais) of the white population’s per capita average income (2,012 reais).
• 3
  A sample questionnaire can be found at the annex of this paper.

References

• ALMEIDA, Ronaldo; D’ANDREA, Tiaraju. Pobreza e redes sociais em uma favela paulistana. Novos Estudos CEBRAP, v. 28, p. 94-106, 2004.
• ANDRADE, Luciana T.; SILVEIRA, Leonardo S. Efeito-território. Explorações em torno de um conceito sociológico. Civitas, v. 13, n. 2, p. 381-402, 2013.
• BAYÓN, Maria C.; SARAVI, Gonzalo A. The cultural dimensions of urban fragmentation segregation, sociability, and inequality in Mexico City. Latin American Perspectives, v. 40, n. 2, p. 35-52, 2013.
• BLAKELY, Edward J.; SNYDER, Mary G. Fortress America: gated communities in the United States. Washington D.C.: Brookings Institution, 1997.
• BORSDORF, Axel; HIDALGO, Rodrigo; VIDAL-KOPPMAN, Sonia. Social segregation and gated communities in Santiago de Chile and Buenos Aires. A comparison. Habitat International, v. 15, p. 1-10, 2015.
• CALDEIRA, Teresa P. do R. Cidade de muros Crime, segregação e cidadania em São Paulo. São Paulo: Edusp, 2000.
• CLAMPET-LUNDQUIST, Susan; MASSEY, Douglas S. Neighborhood effects on economic self-sufficiency: a reconsideration of the moving to opportunity experiment. American Journal of Sociology, v. 114, n. 1, p. 107-143, 2008.
• GALSTER, George C.; KILLEN, Sean P. The geography of metropolitan opportunity: a reconnaissance and conceptual framework. Housing Policy Debate, v. 6, n. 1, p. 7-43, 1995.
• HANNERZ, Ulf. Exploring the city: inquiries toward an urban anthropology. New York: Columbia University Press, 1980.
• HITA, Maria G.; GLEDHILL, John. Anthropologies of the urban periphery: Salvador, Bahia; Manchester: Brooks World Poverty Institute, 2009.
• HÄUßERMANN, Hartmut. Armut in der Großstadt. Die Stadtstruktur verstärkt soziale Ungleichheit. Informationen zur Raumentwicklung, v. 3/4, p. 143-157, 2003.
• JANOSCHKA, Michael; SEQUERA, Jorge. Gentrification in Latin America: addressing the politics and geographies of displacement. Urban Geography, v. 37, n. 8, p. 1175-1194, 2016.
• KAZTMAN, Ruben; RETAMOSO, Alejandro. Segregación residencial en Montevideo: desafíos para la equidad educativa. Santiago de Chile: Comisión Económica para América Latina y el Caribe, 2006.
• LOW, Setha. Behind the gate Life, security, and the pursuit of happiness in fortress America. New York: Routledge, 2004.
• LUCKMANN, Thomas; BERGER, Peter L. The social construction of reality: a treatise in the sociology of knowledge. London: Penguin Books, 1991.
• MARQUES, Eduardo C. de L. Redes sociais, segregação e pobreza São Paulo: Editora Unesp, 2010.
• MASSEY, Douglas S.; DENTON, Nancy A. American apartheid: segregation and the making of the underclass. Cambridge: Harvard Univ. Press, 1993.
• PRÉVÔT-SCHAPIRA, Marie-France; PINEDA, R. Cattaneo. Buenos Aires: la fragmentación en los intersticios de una sociedad polarizada. EURE, v. 34, n. 103, p. 73-92, 2008.
• RIBEIRO, Luis C. de Q. Proximidade territorial e distância social: reflexões sobre o efeito de lugar a partir de um enclave urbano. VeraCidade, v. 3, n. 3, p. 1-21, 2008.
• RIBEIRO, Luis C. de Q. Metamorfoses da ordem urbana da metrópole brasileira: o caso do Rio de Janeiro. Sociologias, v. 18, n. 42, p. 120-160, 2016.
• RIBEIRO, Luis C. de Q.; LAGO, Luciana de. O espaço social das grandes metrópoles brasileiras: Rio de Janeiro, São Paulo e Belo Horizonte. Cadernos Metrópole, v. 4, p. 9-32, 1999.
• RIBEIRO, Luis C. de Q.; KOSLINSKI, Mariane C.; ALVES, Fátima; LASMAR, Cristiane. Desigualdades urbanas, desigualdades escolares Rio de Janeiro: Letra Capital, 2010.
• RIBEIRO, Luis C. de Q.; TELLES, Edward 2000. Rio de Janeiro: emerging dualization in a historically unequal city. In: MARCUSE, Peter; VAN KEMPEN, Ronald (orgs.). Globalizing cities A new spatial order? London: Blackwell Publishing. p. 78-94.
• ROBERTS, Bryan R.; WILSON, Robert H. Urban segregation and governance in the Americas New York: Palgrave Macmillan, 2009.
• RUIZ-TAGLE, Javier. La persistencia de la segregación y la desigualdad en barrios socialmente diversos: un estudio de caso en la Florida. EURE, v. 42, n. 125, p. 81-108, 2016.
• SABATINI, Francisco; SALCEDO, Rodrigo. Gated communities and the poor in Santiago, Chile: functional and symbolic integration in a context of aggressive capitalist colonization of lower-class areas. Housing Policy Debate, v. 18, n. 3, p. 577-606, 2007.
• SALCEDO, Rodrigo; TORRES, Alejandro. Gated communities in Santiago: wall or frontier? International Journal of Urban and Regional Research, v. 28, n. 1, p. 27-44, 2004.
• SAMPSON, Robert J. Great American City Chicago and the enduring neighborhood effect. Chicago: University Press, 2012.
• SMALL, Mario L. Villa Victoria The transformation of social capital in a Boston barrio Chicago: University Press, 2004.
• SMALL, Mario L.; FELDMAN, Jessica 2012. Ethnographic evidence, heterogeneity, and neighborhood effects after moving to opportunity. In: VAN HAM, Maarten et al (orgs.). Neigborhood effects research: new perspectives. Dordrecht: Springer, 2012.
• SMALL, Mario L.; NEWMAN, Katherine. Urban poverty after the truly disadvantaged: the rediscovery of the family, the neighborhood, and culture. Annual Review of Sociology, v. 27, p. 23-45, 2001.
• WACQUANT, Loic. Urban outcasts: a comparative sociology of advanced marginality. Cambridge: Polity Press, 2008.
• WILSON, William J. The truly disadvantaged: the inner city, the underclass and public policy. Chicago: University Press, 1987.

Publication Dates

• Publication in this collection
  04 Dec 2020
• Date of issue
  Sep-Dec 2020

History

• Received
  31 July 2019
• Accepted
  21 July 2020