CRECIMIENTO URBANO Y SEGREGACIÓN SOCIOESPACIAL EN VALDIVIA

This work was carried out within the framework of the Fondecyt Initiation Project N°11170019 “Local integrated development strategies in the cities of southern Chile, growth, vulnerability, crisis and resilience?”. Geógrafa Universidad de Concepción, Concepción Chile Ayudante de investigación en proyecto externo https://orcid.org/0000-0001-8097-4267 maguilad@udec.cl Doctor en Geografía Universidad de Concepción, Concepción, Chile Profesor asociado de la Facultad de Geografía y Arquitectura https://orcid.org/0000-0002-4071-1195 jprada@udec.cl CRECIMIENTO URBANO Y SEGREGACIÓN SOCIOESPACIAL EN VALDIVIA


I. INTRODUCTION
Cities today constitute the main spaces of social and economic reproduction, generating a strong dynamism, but also problems that are still awaiting a response. This is the case of socio-spatial segregation. Some authors have outlined the impact of the real-estate market on urban logics (Hidalgo & Janoschka, 2014;De Mattos, 2016) and the continuing processes of unequal access to cities, that are reflected in a persistent socio-spatial segregation (Clichevsky, 2000;Sabatini, 2000), though generally starting from the study of metropolitan spaces (Ziccardi, 2008;Ruiz-Tagle & López-Morales, 2014). In this sense, this work questions, starting from the case of Valdivia, about what the way, dimension and orientation of urban growth has been in the last five-years considering the influence of the real-estate sector on this. Its second goal is making an analysis and interpretation of the urban segregation there is within the city with a double perspective: spatial (neighborhoods perceived as more or less segregated) and sectorial (elements perceived as those that contribute in a greater or lesser degree to said segregation). Thus, the question this article focuses on is whether recent real-estate growth in Valdivia has been heterogeneous and would be related to the perceived sociospatial segregation there is. A methodology that combines qualitative and quantitative techniques has been used for this, after a revision of the secondary sources.
Valdivia is located in the south of Chile, between the Calle-Calle and Cruces rivers, which have set its historic boundaries, ones exceeded for several decades now ( Figure 1). Maturana, Peña-Cortés, Ramírez & Telias (2019), mention that the situation of Valdivia, set away from the 5 Sur Highway, complicates its effective connection with the rest of the country, this being one of the factors behind its low demographic growth (166,000 inhabitants according to INE, 2017). To put the case study in context, it is necessary to highlight several milestones that have marked its history and set up its dynamism. In this sense, Espinoza & Zumelzu (2016) differentiate four phases: a pre-industrial period from 1850; a second period characterized by industrialization linked to the arrival of German immigrants who had a cultural and economic influence on the area; a third period, marked by the 1960 earthquake and post-earthquake period, which put the brakes on urban development and reconfigured urban growth and expansion patterns towards the southeast sector, where the Housing Corporation and private entities would gradually expand the city and; finally, from 2007 when Valdivia was named as a regional capital, a period where it began to generate more appeal for investors and the population in general, having been considered up until recently as a "lethargic" city within the Chilean urban system (Borsdorf, Sánchez & Marchant, 2009).
Within the framework of the research by Fuentes, Link & Valenzuela (2017), Valdivia appears both in 1992 and 2011 as characterized within average third-tier cities, with a low level of industrialization and a concentration of activities around the service sector, with a growth that would not be based on the exploitation of natural resources or on tertiary specialized and complex activities. In addition, a lack of efficient territorial organization plans is identified, along with an updated land use regularization by the public institutions in charge of this. In this sense, the municipality has had its Communal Regulation Plan and Communal Development Plan for at least a decade, with just private entities taking on urban expansion and construction, without any clear guidelines or plan. The results are presented below, after a brief review of the literature and presentation of the methodology.

Spatial segregation and the real-estate market
There are several definitions of the concept of spatial segregation. For Clichevsky (2000) this implies the differentiation and distancing of part of the population, with exclusion processes being essential, with negative effects like the increase of poverty in the most vulnerable sectors (Prada-Trigo, 2018), inequalities, urban fragmentation and social polarization; which can be analyzed both from a socioeconomic or sociocultural perspective, with the former being the one of interest for this research. According to Garín, Salvo y Bravo (2009), this would refer to the spatial separation between the residential areas of high-and lowincome groups. Torres (2013) highlights that, in the modern city, land that can be modified as the owners so wish has had a profitable value, increasing the added value and the cost of urban life, leading to the abandonment of those excluded and vulnerable sectors that cannot get access to purchase desired dwellings, with the house price being a segregating element.
Considering this, among the significant variables that can be presented between one class and the other, leaving clear the socioeconomic differences that divide the city, there are components related to the materiality and quality of the dwelling, just as Sorribes (2012) mentions; elements related to environmental risk; limited access to services and facilities (Agostini, Brown & Gongora, 2008;Ziccardi, 2008;MacDonald, 2011); polarization of the city into homogeneous neighborhoods, which reduces the possibilities of social mobility (Saraví, 2008); or labor precarity and informality (Winchester, 2008). The differences between these components are what define one sector of the city from others, with the most vulnerable sectors being those with the worst indicators in these components.
For Garín et al. (2009) andAzócar, Henríquez, Valenzuela y Romero (2008), the social housing policy has also had an impact on residential segregation, through a system of subsidies, as it meant locating its beneficiaries in lower cost areas, where there were already poor settlements, minimizing the possibilities of social integration, something that Sabatini (2000) confirms. For Azócar et al (2008) a stigmatization would have been produced in these neighborhoods, accentuating the factors or conditions of social risk that lead to poverty. Thus, although the living conditions of the poorer population may have improved in Chile, their spatial distribution would not have been fundamentally changed.
In this regard, one of the most significant debates regarding socio-spatial segregation refers to whether there would have been a greater integration in recent years (Sabatini, Wormald, Sierralta y Peters, 2009) or, on the contrary, there would have been an increase of the segregation (López-Morales, 2015). The latter questions that some of the works on socio-spatial segregation are not conceptually or methodologically coherent with social reality or with the scale of this phenomenon (Ruiz-Tagle & López-Morales, 2014). In this sense, most studies turn to methodologies that, through statistical indexes and indicators, try to approach the concept of socio-spatial segregation, generally at a district scale (Garín et al., 2009;Azócar et al., 2008). However, as Espino (2008) states, there are also a series of symbols, statues and imaginaries that are limiting factors when attempting class integration, a perspective that this work takes on to try to approach the perception of socio-spatial segregation in Valdivia and its relationship with its recent growth reaching the components and causes of this (Ruiz-Tagle & López-Morales, 2014). To address this perspective, a mixed methodology is used which is explained in the following point and that is articulated starting from the use of interviews and statistical information.
Added to this, authors like Daher (2013) andDe Mattos (2016), state that the real-estate sector has become fundamental for the new economic model developed since the 1970s (Hidalgo & Janoschka, 2014). The overaccumulation of capital and the need for investment, added to urban and population expansion, found in this, a strategic niche, working together with other sectors of the economy. In this way, the real-estate sector takes on a key role in the development and growth of cities, since it moves between the financial sector and the real economy, connecting to investments that are in pension funds, insurance, credit, etc. (Cattaneo, 2011;Gasic, 2018). The market in this case acts as a modeler of the city's production, through an urban business model (Harvey, 2007), where economic efficiency decisions prevail over other aspects like socio-spatial integration or equal access to goods and services. The result is a "com-fuse" (compact and diffuse at the same time) city model that advances in a segmented way over rural spaces (Abramo, 2012), along the main transportation routes and a densification of the city center from a selective verticalization, where the aforementioned processes of urban segregation would take place (Fuentes & Pezoa, 2017). From this perspective, the strong real-estate growth and existing deregulation of this, would act as stimuli towards this socio-spatial segregation.

II. METHODOLOGY
This work applies a methodology that combines qualitative and quantitative techniques, triangulating data of different natures to obtain more reliable results (Yeung, 2003), which allows addressing the study problem from different angles. This goal is based on the value of applying mixed methodologies over others that focus only on qualitative or quantitative data. In this sense, statistical data of building permits and final inspection certificates were used, where the ArcGis software helped to spatialize the growth of Valdivia, and to understand the distribution of infrastructure. In addition, in-depth interviews were made. Fuster (2016), regarding these, mentions that these give the person interviewed freedom to express themselves, but at the same time there is control over the information, with the advantage of being able to adapt to the informant and subtly guide dialog, a very valuable complement for quantitative methods. This work also proposes, in a novel manner, integrating a restriction matrix from the research of Barrenechea, Rodríguez y Troncoso (2014). This was adapted to the research needs, generating specific dimensions and aspects, starting from the segregation dimensions identified in the revision of the literature, to try to attain a spatial systematization of this, making reference to the spatial distancing produced by the people's purchasing power (Toro & Orozco, 2018). For this, the components that have an impact on the separation of classes were chosen, considering aspects that according to the literature have an influence on this process, using indicators like the house price and its quality or the number of green areas, along with other insights: security, accessibility, among others. To complete this, the interviewees were asked to give a rating for each preestablished sector of Valdivia, using a progressive scale regarding the districts and dimensions that were defined previously (Table 1). In this sense, those variables that do not present major obstacles were considered as "0" and those that were as "3", with these being identified by the interviewees as the main problems to overcome urban segregation. The data was counted in the matrix's columns and rows, generating a desegregation, in this way, of the socio-spatial vulnerability by urban sectors and components thereof.

Urban growth and evolution of the real-estate market
The spatialization of building permits between 2014 and 2019 allows visualizing their concentration and seeing where Valdivia is growing ( Figure 2). Most non-habitational permits are located in the districts of Mercedes, Camilo Henríquez, Pantano and Aguirre, these being the central spaces for inhabitants and visitors since, as can be seen, these are the ones with the highest number of commercial projects, mainly corresponding to hotels, hostels and cabins ( Figure 3). The permits related to habitational and mixed use, as well as those that are smaller than 64 units (Figure 4) are mainly in the districts of Huellehue, the southern sector of Las Ánimas, the southeastern outskirts of Las Mulatas, and the northern part of Teniente Merino, which mainly correspond to individual enterprises like cabins or extensions for rental purposes, generating a microeconomy focused on university students and tourists. These fundamentally coincide with the historic distribution of social housing in Valdivia ( Figure 2). In addition, those permits for 65 to 250 units, along with those of Figure 2, show where the city is residentially growing towards through private actors who foster the city's Southeastern growth, building private gated communities and subsidized social dwellings, located in these sectors. In general, the outskirts of Valdivia would be developing this dynamic, as was mentioned in some interviews: [ENT-1]: In general terms, an "inside-out" growth model has been generated, mainly towards the south, actually dominated by longitudinal corridors over transversal ones, so accessibility has been allowing the city to expand.
It is also relevant to mention the presence of high-rise building projects (Figure 5), which range from 4 to 11 floors in the different districts of the city, concentrated in Mercedes and Isla Teja. The case of Valdivia is particular in that the verticalization processes coincide with the area where people with the highest purchasing power are concentrated, with this type of infrastructure being a symbol of modernity and economic status in the city. In this sense, there is a particular case, where a permit for a 19-floor tower with 250 housing units appears, located in the north sector of Las Mulatas (an    area that still did not have verticalization projects). This responds to the fact that the city will eventually start to densify with high-rise projects in other districts, but without a Communal Regulatory Plan (PRC in Spanish) that regulates them, ones which considerably change the city's traditional landscape, as has been mentioned in other cities (Pérez, González, Villouta, Pagola y Ávila, 2019). Alongside this, there is an important concentration in Las Ánimas, where habitational permits and projects of residences or cabins predominate, being an aspect that, on promoting trade, tourism and new residences, could generate a greater integration of this district to the commercial dynamics of Valdivia and a greater connection of its population with the rest of the city.
In summary, it can be seen that, while most habitational and mixed permits correspond to working-class sectors and the peripheries, like Las Mulatas or Teniente Merino, those which refer to high-rise buildings or to commercial and hotel-based uses, tend to be more centrally located and in the most prestigious neighborhoods, like Isla Teja, Mercedes or Camilo Henríquez. The result is a dual model, where the most attractive areas for tourism, business or middle and upper classes receive greater investment, while the working-class sectors correspond to local enterprises, social housing or, in the best case, gatedcommunities. The result is a dual city in the sense that Abramo (2012) points out, and would respond to the neoliberal model  (Hidalgo & Janoschka, 2014). This situation will possibly have an effect on the perception of spatial segregation, and referring us, just like in other cases of Chile and Latin America, to similar spatial logics detected inside the city (Daher, 2013;Dammert, Delgadillo & Erazo, 2019).

Spatial segregation
The application of the matrix (Table 1) showed that in Valdivia, there would be, as was inferred from the statistical analysis, a segregation marked by polar opposites, that is to say, there is a voluntarily segregated group corresponding to a high socioeconomic class, who look to live in the outskirts, mainly in Isla Teja, Estancilla and the eastern part of Estación. While, the most vulnerable population would be located on the outer strip to the south of Las Mulatas, Teniente Merino and Huellehue, with some social housing and occupied areas concentrated in Las Ánimas.
From the results, Las Ánimas ends up being the district with the greatest restrictions and thus the most segregated of the city, with this being a historically working-class and stigmatized neighborhood. In addition, it is in this and in some sectors of Las Mulatas where recent international immigration has arrived, who in some cases work informally, increasing job insecurity and the precariousness of the district. On the other hand, eradications made from Santiago to Valdivia, have generally been located in this district. In this regard, Godoy (2019) mentions that the moves the State makes, do not always relocate vulnerable families in districts that are much different from where they came, that on occasions it ends up worsening their quality of life, assigning precarious dwellings to them in marginalized sectors that are disconnected from their former networks, increasing segregation and insecurity, which seems to be the case of Las Ánimas.
Pantano and Las Mulatas follow Las Ánimas in the ranking. The former, a historic industrial neighborhood, has a mix of heritage façades with houses of a precarious infrastructure, where extensions of the dwellings also dominate, for rooms to let. There is a certain degree of heterogeneity regarding people who live in the district, especially in terms of education levels, as recently enterprises related to creative industries have been set up, this being the reason for the educational differences. This sector's heterogeneity is left clear by the proximity of some dwellings and others, which does not mean there is no inequality and segregation between those living there, but rather this is lived in a much closer and reduced space (Godoy, 2019;Jirón & Mansilla, 2014). The neighborhood of Las Mulatas stands out by its heterogeneity and its lack of access to administrative services, being very dependent on the city center. Data shows a public disinvestment in the sector's transportation services, making the differences between those who can access private transport and those who depend on public services visible (Jirón & Mansilla, 2014). This aspect, added to the general deficiency that Valdivia has regarding the materiality of the dwelling, leads to this being a district where in some areas the quality of life is good, while in sectors mainly in the southern area, it is complicated to maintain a suitable standard, replicating on a smaller scale, the dynamics already detected by other authors (Saraví, 2008;Winchester, 2008).
The districts of Huellelhue, Teniente Merino and Estación come next in the ranking, as their restrictions are not as high and they are districts that, in general, do not have major problems beyond the backfill of marshlands and the sewerage issues of Aguas Décimas. Aguirre and Camilo Henríquez follow the same trend as Teniente Merino, with the difference being the central location of the first two. These are mixed zones with residential and commercial uses, representative of the average standard of Valdivia.
At the top end are Estancilla, Mercedes and Isla Teja, the city's privileged sectors where there are several gated real-estate projects or private plots. The first is a district that has large green areas as lotting projects prevail. In this respect, Rojo (2015) indicates that this type of housing in closed spaces generates that reality is distorted in such a way that, living there leads to ignoring other urban realities, generating segregation on a deeper scale than simply the spatial one, since the symbolic aspects and psychological aspects of status are given to these processes (Janoschka, 2016). Mercedes, on the other hand, corresponds to the city's commercial and tourism sector. This is where the greatest dynamism and flow of people can be seen, alongside Isla Teja, especially in the summer months. It is a sector that is well provided for with services and facilities. Also, the district has natural landscapes for those who live or visit it, meaning that it is expensive to live or buy in this sector, caused by the high demand there is to use land in the downtown area and all its associated benefits. This is seen in extracts from an interview: Finally, Isla Teja is the sector with the least restrictions on average, but the cost of living is really high. It is a district that has historically been a privileged sector of the city, where the first German families set up home, later donating land to the Austral University, which turned Isla Teja into a tourist attraction, with Saval Park, the Botanical Garden and the Arboretum, along with the German façades. All this attracts tourists and investors, who saw the possibility of a target audience to begin building in this sector, along with investments in businesses. Janoschka (2016) mentions that this type of urban formation is violent for those who cannot access it, making these places desirable due to the amount of capital invested, creating sectors with all the comforts for a quality urban life, being promoted both by private entities and the State, even more so in a city where the touristic spaces are those which have the best projects. Pontes et al. (2020) indicate that, in the end, it is the relationship between tourism and the real-estate market that causes certain areas to be more valued than others, thus segregation in this case is strongly influenced by the tourism approach Valdivia has, with districts having greater urban benefits than others. Figure 7 spatially represents the districts and values of the sectorial total of the restrictions matrix provided by the interviewees.
As a complement of the analysis by sectors, another can be briefly made regarding the dimension of segregation in Valdivia. In this sense, the price of housing, socioeconomic diversity, and environmental risk stand out as its main exponents (Table 1). With regard to the former, this is the greatest limiting factor to have a home in Valdivia, considering the limited supply and the high demand. Aside from this, the investment required to improve land conditions to be able to build, increases the end price, added to the existing speculation and added value.
Although the State has prioritized access to housing as a symbol of overcoming poverty, reducing it as the years have gone by, it does not necessarily mean that the quality of dwellings and the ways of life of the working classes have been resolved. On the contrary, the way the State has given land and homes responds to urban enterprise strategies and policies to encourage the valuation of certain pieces of land, generating disparities in the allocation of resources and triggering segregation processes (Alvarado, 2019).
In the case of socioeconomic diversity, this is related to the previous concept, considering that land segmented into prices links access to housing with the capacity to pay or take on debt (Sabatini, 2000;Abramo, 2012). The result is a conformation of different neighborhoods starting from homogenous social stratums, with the resulting socio-spatial polarization. This would end up being the visible face of the market logics that operate on the land and organize the city. Regarding environmental risk, the third most restrictive dimension, it is linked to the fact that Valdivia is built on marshland, which generates an obstacle when it comes to investing in the city. There are several areas where there is greater risk, those closer to the marshes being the most sensitive ones. There are other dimensions, with a lesser weight, like security, with the districts of Las Mulatas and Las Ánimas the most dangerous. In the case of dwelling materiality and green areas, these are not so restrictive, considering that the latter correspond to a classic trait of Valdivia, although having said this, there still are neighborhoods with limiting factors in this dimension. Finally, the labor informality is not that visible. It is seen more on the peripheries or in industrial areas, where it has not been perceived as a greatly important dimension.

IV. DISCUSSION
The case study presented shows similarities with the perspective of Chichevsky (2000) or Garín et al. (2009), as spatial segregation would be formed from a physical point of view. The spread, in recent decades, of urban fragments for the population with greater resources in areas like Isla Teja or Estancilla is a reflection of this. Consequently, the increase in land value that accompanies the higher added value of these sectors would have an effect on a segregation of the population with fewer resources, just as Torres (2013) indicates. In this way, the case of Valdivia, where the price of housing is the main segregating element, would be linked to studies like those of Saraví (2008), who also highlight this element. The methodological approach made has allowed more clearly highlighting symbolic aspects of segregation that are not always evident in more quantitative approaches (Ruiz-Tagle & López-Morales, 2014).
In this sense, the case of Valdivia, which went from being a "lethargic" city within the Chilean system (Borsdorf, Sánchez & Marchant, 2009) to a regional capital with an important real-estate dynamism, reflects that when there is an urbanistic effervescence and limited or non-existent public regulation, the result is a rampant spatial segregation (Fuster, 2016). In this way, the results obtained leave Valdivia closer to the perspectives of López-Morales (2015) or Ruiz-Tagle than those of Sabatini et al. (2009). Although, in this case, the size of the city could lead to thinking about other results, which they seem to reproduce, on a smaller scale, the same logics of Santiago. In this cocktail, the subsidized housing policy, just like in other cases analyzed (Azócar et al, 2008;Garín et al, 2009) would act to accentuate differences, more than to reduce segregation. Just like in other cases (Cattaneo, 2011), the real-estate sector here takes on a central role in the development and growth of urban spaces, modeling the production of the city through an urban enterprise model (Harvey, 2007), which is replicated at a different scale in the cases with the most literature.

V. CONCLUSIONS
The case study of Valdivia shows a growth that has been occurring in a way where sectors like Estancilla, Isla Teja and Huellehue hold most of the population with greater purchasing power. On the other hand, districts like Las Ánimas or the peripheries of Las Mulatas are working class neighborhoods that house the most vulnerable population, generating focal points or segments in the urban network that begin to be stigmatized or classified by socioeconomic status and level, which are recognizable both in physical and symbolic patterns. The way Valdivia grows is mainly due to high-rise densification projects alongside lotting on the urban periphery, being the type of "compact and diffuse" city that Sanabria and Ramirez (2017) or Abramo (2012) mention, that would be taking place in Valdivia, where the downtown areas begin to be ever more densified, while those outside this perimeter and towards the periphery gradually spread. The new role of the city as the administrative capital, its tradition as a center for higher education and the growing domestic and international tourism would be the drivers behind this phenomenon.
The lack of an updated PRC, some shortcomings in terms of availability and distribution of services, added to public and private plans that are not effectively coordinated, produce a city that is organized by who can invest more and where, with the private sector having the greatest gains in this regard, regulating and setting the housing prices. The result is a difference in the distribution and typologies of building permits there are, which is later reflected in the mentioned socio-spatial segregation. Therefore, this research has allowed making new contributions about urban dynamics in intermediate cities, opening up new topics for research which could be developed by reviewing other case studies.