doi:10.5477/cis/reis.194.121-138
Housing as a Driver of Social Exclusion
in the European Union
La vivienda como factor de exclusión social en la Unión Europea
Almudena Martínez del Olmo
|
Key words Welfare
|
Abstract This article seeks to understand the extent and manner by which housing influences social exclusion levels in the European Union. A multivariate analysis was used along with data from the EU-SILC, to examine the impact of key variables such as tenure and overcrowding on the AROPE indicator. The results suggest a significant association between housing and social exclusion. Countries with more market-oriented housing systems tend to display higher levels of social exclusion as compared to those relying on more redistributive systems. These findings challenge the effectiveness of housing policies tied to an asset-based welfare approach in reducing inequality. Furthermore, they highlight the need to strengthen the measurement of housing in analyses of social exclusion in order to develop more accurate diagnoses. |
|
Palabras clave Bienestar
|
Resumen Este artículo busca comprender la intensidad y el modo en el que la vivienda influye en los niveles de exclusión social en la Unión Europea. Mediante un análisis multivariado, empleando datos de la EU-SILC, se explora el impacto de variables clave, como la tenencia o el hacinamiento, sobre el indicador AROPE. Los resultados muestran una significativa asociación entre la vivienda y la exclusión social. Países con sistemas de vivienda más mercantilizados tienden a presentar mayores niveles de exclusión social frente a aquellos con sistemas más redistributivos. Estos hallazgos cuestionan la efectividad de políticas de vivienda ligadas a una noción del bienestar basada en activos para reducir la desigualdad. Además, resaltan la necesidad de fortalecer la medición de la vivienda en los análisis de exclusión social para elaborar diagnósticos más precisos. |
Citation
Martínez del Olmo, Almudena (2026). “Housing as a Driver of Social Exclusion in the European Union”. Revista Española de Investigaciones Sociológicas, 194: 121-138. (doi: 10.5477/cis/reis.194.121-138)
Almudena Martínez del Olmo: Universidad Rey Juan Carlos | almudena.martinezo@urjc.es
Far more than a mere consumer good, housing is considered an essential pillar of human and social development. Its accessibility, quality, and adequacy to people’s needs significantly influence their well-being, opportunities, and level of integration in society (Cortés, 1997). Therefore, it is a need as well as a fundamental right. However, large sectors of the population are deprived of decent residential conditions, undermining their life possibilities (Leal, 2017). While the analysis of the relationship between housing and social inequality is not new, it has been the subject of growing interest over recent decades, both by academics and public policymakers, given the intensification of housing-related problems and their social repercussions. The delayed emancipation of young people, the growing processes of social inequality in space, and the residential barriers preventing the social inclusion of immigrants are some of the most frequently studied manifestations. In this regard, and beyond the influence of economic, labor, and legislative factors, the literature has highlighted the importance of housing systems in shaping social inequalities. It has identified different housing models, each with its own characteristics and consequences for residential and social well-being (Kemeny, 1991; Norris and Winston, 2011; Olsen, 2018). These systems vary depending on the role and means of intervention of the different agents responsible for their provision (the state, the market, civil society and the family). This study examines the complex relationship between housing systems and social exclusion processes in the European Union (EU), in an attempt to shed light on one of the deepest manifestations of social inequality in societies. The main objective is to understand how different housing system configurations (which are characterized by the diversity of tenure models, levels of state intervention and real estate market dynamics) may influence the likelihood of individuals and social groups being in situations of exclusion. The relevance of this research lies in the growing evidence linking housing inequalities to the deepening of social gaps. The 2008 economic crisis highlighted the housing vulnerability of large segments of the population, which, in turn, exacerbated problems such as poverty, inequality, and social exclusion. Furthermore, in line with the 2030 Agenda, the EU has declared that reducing poverty and social exclusion is one of its main objectives. Therefore, understanding the role of housing as a risk factor is fundamental to designing more effective and equitable public policies.
To achieve the main objective, it is necessary to determine the degree to which the characteristics of housing systems explain the variability in levels of social exclusion in the EU countries. To this end, this work will explore the dimensions and mechanisms through which housing influences social exclusion, considering variables such as tenure, housing affordability, and the physical and social conditions of housing. Furthermore, the aim is to identify groups of countries with similar patterns in terms of housing and levels of social exclusion. This will offer a better understanding of national dynamics and serve to encourage the design of more effective housing policies aimed at reducing social exclusion and promoting social cohesion.
To achieve these objectives, a multivariate quantitative analysis will be conducted using comparative data from the different countries of the so-called “European Union of 27”. First, a multiple regression model will be applied to assess the impact of the most relevant variables related to housing systems (e.g., homeownership rate, proportion of social housing, rent affordability rate) on the AROPE (At risk of poverty and/or exclusion) indicator. This indicator refers to the percentage of the population that is at risk of poverty or social exclusion. Second, a cluster analysis will be performed to identify groups of countries having similar characteristics in terms of housing and levels of social exclusion. To achieve this, the EU-SILC (European Survey on Income and Living Conditions) will be used as the main source of information. Its annual periodicity and consideration of basic residential variables and the AROPE indicator provide significant, comparable and up-to-date data for the proposed objectives.
The research will be structured as follows: First, a literature review will be carried out on the relationship between housing and social exclusion. It is followed by an explanation of the various relevant scientific contributions on the configuration of housing systems and their connection with different expressions of social inequality. Then, the research methodology and results are detailed, providing a description of the data sources, the variables used, and the statistical methods employed with their respective results. Finally, the main findings are presented and discussed, offering recommendations for future research.
Theoretical framework
and state of the art
Housing and social exclusion:
an increasingly close link
Unlike poverty, which is more closely linked to a lack of economic resources, the concept of social exclusion originated in France in the 1970s to designate a population that did not receive social benefits. Acceptance of this term has gradually progressed and expanded to refer to how, through the interaction of various factors of disadvantage or social vulnerability, certain individuals or groups are marginalized or excluded from the mechanisms needed to feel and develop fully as human beings (Subirats, 2004). Thus, beyond socioeconomic imbalances or disparities within a social scale, the concept of social exclusion refers to a state of being outside of society (Giddens, 1998; Alguacil, 2006). This, far from being a static and independent condition, it is a gradual and dynamic process that makes sense when considering its opposite: social integration (Tezanos, 2001). The difficulty of identifying and operationalizing the positions that designate exclusion conflicts (given the numerous potential states and itineraries), has led authors such as Robert Castel (1995) to distinguish intermediate stages or zones of vulnerability or precariousness within the integration-exclusion axis. Thus, the positioning of an individual or group in each of the defined zones depends on the configuration and interdependence of a multitude of variables linked to different social dimensions (Orum and Silver, 2019). These include housing occupying a fundamental position given its large impact on the development and fulfillment of the individual. Health, employment status, safety, mood, autonomy, and privacy are all contingent upon living in a suitable and adequate habitat that meets one’s needs. Therefore, if this habitat has limitations or basic deficiencies, other vital areas will also be compromised, and as a consequence, the door to social exclusion opens (Lowe, 2004). Engels’ The Condition of the Working Class in England, Engels (1845) clearly documented how the terrible housing conditions off 19th century workers had a devastating effect on their health and wellbeing. The decay and unsanitary conditions of their homes were breeding grounds for disease, violence, alcoholism, and other social problems, which significantly shortened their life expectancy. In contemporary society, authors as relevant as Sennett (2018), Sassen (1991) and Bourdieu (1979) have confirmed that housing continues to act as a powerful mechanism for reproducing social inequalities and generating processes of social uprooting, segregation and isolation. These processes may lead to vulnerability and social exclusion.
Thus, the great importance of housing for human well-being suggests that it be considered a primary need. This results in its recognition as a fundamental right (Pisarello, 2009). This appeared as far back as the 1945 Universal Declaration of Human Rights (article 25.1) and the 1976 International Covenant on Economic, Social and Cultural Rights (article 11.1). In Europe, the Revised European Social Charter (1965), the Charter of Fundamental Rights of the European Union (2000) and the constitutions of various countries have also reaffirmed this right and undertake to guarantee it. However, despite this widespread recognition, housing continues to be one of the most fragile or wobbly pillars of the welfare state (Togersten, 1987). Without a doubt, the passage of time has revealed how the development of welfare states, rights, and health and technological advances have led to a progressive improvement in residential standards in Western societies, so that the extreme forms of residential insecurity, typical of the industrial era, have become a much more limited reality. However, residential inequalities continue to persist over time, taking on new forms. In addition to the inability to eliminate certain pockets of extreme residential insecurity, significant barriers exist to housing access, as well as processes of gentrification/peripheralization/segregation, a lack of residential stability, and the inadequacy of housing to meet the residential needs of much of the population. All of this inevitably compromises the life opportunities of these individuals (Murie, 2000; Domínguez, Leal and Barañano, 2021). Such is the case for the foreign-born population, among whom overcrowding, substandard housing, discrimination in access to housing, and residence in deteriorated environments are commonplace and reinforce their vulnerability in other areas such as employment and healthcare. Young people constitute another social group for whom housing has become a major obstacle to the development of a life plan demanding the ability to become independent and live in adequate conditions. Housing problems are also endemic among a large proportion of low-income households, single-parent families, and people in situations of dependency.
Currently, a scenario also exists in which housing problems, traditionally found in the most disadvantaged social groups, appear to be worsening and spreading to more affluent social strata. This is causing a major global housing crisis that is undermining the well-being and cohesion of European societies (Fields, 2017; Potts, 2020; Saiz, 2023). Given this situation, factors such as labor market dynamics, health of the economy, education, legal regulations, and even individual preferences have been frequently cited as explanatory mechanisms for the precarious housing situations that affect may groups of society. This precariousness diminishes their opportunities and quality of life (Gallent, 2019; Parkinson, Wood and Campbell, 2024). However, the formation, scope, and exacerbation of housing problems ultimately depend on understanding how housing is structured in each society. This structure determines not only the extent to which it transmits other social inequalities to the residential sphere, but also its capacity to generate or exacerbate them (Murie, 2012a). From this perspective, over recent decades, scientific literature has increasingly highlighted the link between housing configuration under an asset-based welfare model and more pronounced levels of social inequality (Stephens, Lux and Sunega, 2015). Based on neoliberal ideology, this paradigm assumes that poverty reduction and the promotion of social mobility require individuals to manage their own well-being by accumulating assets rather than relying solely on their income or social transfers from the State (Sherraden, 2003, 2016). In the housing sector, this logic translates primarily into the promotion and idealization of homeownership, understood as a possession that fosters stability, equality, and economic security in the face of an uncertain future. It also permits the accumulation of wealth and the aspiration for greater social well-being (Doling and Ronald, 2010). However, persistent and growing difficulties in accessing homeownership, the concentration of homeownership among the wealthiest segments of the population and an increased residential insecurity stemming from price volatility and financial conditions are some of the factors that have led to questioning if this model is in fact a guarantor of equality and well-being. It has even been labelled a false promise (Arundel and Ronald, 2021). Given this debate, it is crucial to delve deeper and determine how housing is conceived and implemented in each society. This will make it possible to assess its connection with processes of social exclusion and, ultimately, develop effective public policies that fight against poverty and social exclusion.
An approach to the processes of social exclusion according to housing system configuration
Despite fulfilling a primary social function and therefore being a necessity, housing also operates as a commodity. Therefore, its access and adequacy are conditioned by market logic (Madden and Marcuse, 2018). This duality implies that, instead of a universal concept, housing acquires specific meanings and roles in each society, depending on how it is integrated into its welfare systems. Within these, the weight assumed by the different agents in charge of its provision – specifically the State, the family, the market and civil society – and the means by which they do so will be key in explaining how housing (through its configuration) conditions social inequalities. This has been confirmed by comparative housing studies. Inspired by the classification of welfare regimes conducted by Esping Andersen in 1990 (liberal, conservative and social democratic), they demonstrate the coexistence of different housing systems linked to differentiated processes of inequality. Based on this perspective, and despite its polarized and reductionist view, Kemeny’s (1995) contribution is essential. His analysis of the form taken by the tenure regime led to his identification of two novel housing systems. Here, the link between the promotion of homeownership and a greater incidence of inequality was already considered. It defines dualist rental systems, in which the State provides social housing exclusively to the most disadvantaged. Social housing is isolated from the private rental market which is expensive, minority and unattractive compared to homeownership, which receives strong political support and is the predominant form of tenure. This structure (characteristic of Anglo-Saxon countries) concentrates disadvantaged groups in the rental market. Thus, those who do not have access to the limited public rental market must resort to the expensive and insecure private market, which affects their residential well-being and increases their risk of poverty and exclusion. Conversely, it describes unitary rental system, in which housing policy encourages competition between the social and private rental sectors. Here, homeownership takes on a secondary role due to the efficiency and attractiveness of the rental market. When the differences between the for-profit (private) rental sector and the non-profit (social or public) rental sector are relatively small in quality and price, that which Kemeny referred to as an “integrated rental market” is consolidated. In these systems, social equality is fostered by the balance and competition between private and social rental, which is the majority tenure system. In this model, the Netherlands appears, having the most developed social rental sector. It is followed by Sweden and Denmark, which have a balance between social and private rental, and finally Germany and Switzerland, where private rental predominates.
While Kemeny’s theory has limitations in that is does not encompass the diversity of national contexts (Lux, Kährik and Sunega, 2012; Hick, Pomati and Stephens, 2022) and shows a certain weariness with the evolution and diversification of housing policies (Stephens, 2020; Blessing, 2015), its analysis continues to offer two crucial insights for understanding inequality. First, it highlights the importance of the tenure system as a determining factor in the creation of inequalities. Secondly, it reveals how a residential welfare model based on privatization and a welfare state harms low-income groups. This contrasts with the greater equality promoted by collectivist models, showing the different social implications of a universal housing policy versus a selective one (Bengtsson, 2001).
In addition to the crucial role played by the tenure regime, Barlow and Duncan (1994) include the role of residential provision structures, identifying four housing systems having distinct impacts on social inequality. Their study reveals a clear trend: the more commodified the housing provision system, the greater the inequality it generates. This thesis is evidenced by contrasting the liberal system (England) with the social democratic one (Sweden). In the liberal model, the promotion of speculative housing, dominated by large companies, and the management of land by private developers, along with limited state intervention focused on subsidies for vulnerable groups, lead to clear inequality. Housing is viewed as an asset, which encourages property accumulation among the wealthy and stigmatizes the disadvantaged. In stark contrast, the Swedish social democratic model views housing as a social necessity. It promotes de-commodified production with strong restrictions on private profit and state management of land to prevent speculation. Extensive state intervention seeks equality of access and quality, prioritizing the satisfaction of residential needs and social integration over the accumulation of private assets. This approach shows how a less market-driven provision system with strong state intervention can significantly mitigate inequality. The corporatist system (France) falls in an intermediate position, given that, despite the high presence of private developers, the management and costs of land and housing are regulated. This limits private profit, even though the notion of asset prevails. State intervention is more significant than in the liberal model, although it is temporary and seeks to mitigate residential problems without altering existing social differentiation. Finally, the rudimentary system, based on small-scale private development and self-promotion with speculative land management, prioritizes housing as an asset and, given its lack of regulation, tends to generate greater inequality.
Although Barlow and Duncan do not specify it, the rudimentary model can include the countries of southern Europe, which are excluded from the influential interpretive schemes of Esping Andersen and Kemeny but are essential for presenting a broad picture of welfare and housing in Europe. In accordance with Allen et al. (2004), a later and less stable socioeconomic and residential development (with respect to northern Europe) has consolidated a distinct housing system in southern Europe, including Spain. The following are its main characteristics: high rates of homeownership at the expense of social rental housing, high percentages of second homes, the fundamental role of the family in housing provision, and self-promotion and self-construction as mechanisms to access housing (except in the case of Spain, where housing provision is regulated through the work of large companies in conjunction with a more developed urban planning system). The inclusion of this, the so-called family-based model, provides a more complete view of how housing provision is configured. In it, the family has traditionally proven to be an essential institution to ensure residential well-being, in the absence of state intervention. The limited social housing available for rent and promoted by the State in favor of homeownership as a strategy for social integration and economic development makes the family the institution responsible for developing strategies to favor access to housing for its members. Thus, the high percentage of empty and second homes in southern Europe is based not only on speculative or tourist investment, but also, and fundamentally, on the creation of assets. These assets seek to facilitate access to housing for different household members. In this way, this avoids the social risks that may arise from a clientelist political system and a segmented and highly informal labor market (Pareja-Eastway and Sánchez-Martínez, 2015). This context creates significant housing difficulties for those without family support networks or sufficient economic resources, since their access to housing depends largely on a limited, expensive, and precarious free rental market. This places them in a clear position of social disadvantage.
Finally, the evolution from a more universalist model to a liberal one leads the Eastern European countries to form another distinct typology, since their unique history gives them particular singularities. This has led to what Clapham (1995) refers to as “transitional housing systems”.
These contributions to comparative housing studies in Europe have led to the identification of five residential systems that are linked to different notions of well-being and differentiated processes of inequality (Martínez del Olmo, 2020). First, there are the so-called liberal systems, which prioritize private property in a speculative market with minimal state intervention, generating high inequality and exclusion. In contrast, social democratic systems, which conceive of housing as a social necessity, are characterized by a predominance of affordable social housing and de-commoditized production. On the other hand, corporatist systems, which give greater prominence to renting compared to the liberal model, implement strong regulation of land and housing costs to facilitate affordability. Family-based systems, which rely on homeownership and the profitable management of housing, also place the responsibility for ensuring residential well-being on the family. This has negative consequences for the most vulnerable population. Finally, transitional systems promote homeownership after the privatization of social housing, causing serious access problems.
Although the classification of different European countries under housing system typologies is subject to constant evolution, it has served as a theoretical reference for numerous studies which, from a more applied and concrete perspective, have continued to delve deeper into the social repercussions of different housing systems.
Among the studies of social vulnerability in ethnic minorities, Arbaci’s approach (2019) is especially noteworthy. It analyzes how European cities that are integrated under corporate housing systems display the lowest levels of residential segregation due to their mixed forms of production and a unitary tenure system that favors social mixing. In contrast, liberal cities would experience the highest levels of segregation as a result of large-scale speculative housing production and a dualistic tenure system. Meanwhile, cities with social democratic systems display intermediate levels of segregation due to large-scale housing production that shapes their unitary tenure system. Finally, cities with family-oriented systems, while exhibiting low segregation due to their fragmented housing production, have greater residential inequality due to a dual tenure system that prioritizes housing as an asset. This marginalizes low-income, ethnic minorities in a free market of expensive, precarious, and informal rentals.
On the other hand, considering the age variable, it has been seen that the precarious housing situation characterizing young people is differentially conditioned by the different housing systems. As Bosh (2017) pointed out, emancipation rates tend to be higher in countries of the social-democratic and liberal models, since in these countries, social housing and aid intended for this group are more developed, in contrast to countries integrated into the transitional and Mediterranean model.
Housing studies occupy a prominent place in the scientific literature related to social class. The concept of housing classes, created by Rex and Moore (1967), was already emerging as an indicator and reproducing mechanism of social inequalities, establishing a dichotomy between those who have resources and can access homeownership versus those who are in a precarious economic and employment position and are confined to renting public housing. The changes taking place in housing systems since the 1970s, going from a more redistributive model to a liberal one focused on promoting housing as an asset, have been identified as triggering factors for the growing isolation, segregation, and social stigmatization suffered by lower-income groups residing in social housing. This social housing has been subject to a process of disinvestment, abandonment, and privatization (Forrest, Murie and Williams, 1990; Murie, 2007). This has been the case in England, as well as certain Eastern European countries where social housing has become a trap that reinforces social exclusion. Elsinga and Hoekstra (2015) observed that housing policies promoting homeownership, as part of asset-based residential welfare, increase inequality between types of homeowners, but especially between homeowners and renters. This, in turn, leads to unacceptable levels of inequality and exclusion. Dewilde (2017, 2022) demonstrated that the financialization of housing in asset-based systems (liberal and family-oriented) is related to an increase in residential precarity, both among the renting population and among low-income homeowners. Conversely, greater state intervention in social democratic and corporatist systems is linked to less inequality, as better residential conditions are recorded among the low-income population and those living in rented accommodations.
Methodology, hypothesis
and information sources
In addition to focusing on and confirming the coexistence of different housing systems at the European level, numerous studies have linked these systems to residential behaviors and processes of social inequality that vary in form and intensity. The first objective of this work is to use multiple linear regression analysis to determine the degree and type of relationship existing between housing configuration and the degree of social exclusion. In this way, its influence on social inequality in its most extreme form is considered, offering a novel approach. This methodology is crucial since it allows us to determine the specific influence of various residential variables on levels of social exclusion, while also considering their simultaneous interaction. The initial hypothesis assumes a close link between housing type and social exclusion.
For this analysis, the EU SILC (European Union Statistics of Income and Life Conditions) survey is used as the primary source of information. It contains up-to-date and comparable data at the European level on the variables needed to analyze housing configurations. In addition to the indicators provided by EU SILC, other basic indicators from Eurostat and the Housing Europe Observatory are also included.
For the operationalization of the concept of social exclusion, the AROPE indicator (At Risk Of Poverty and/or Exclusion), also included in the EU SILC, is taken as a reference. Its multidimensional nature permits the identification of those at risk of both poverty and social exclusion based on the interaction between three dimensions: risk of poverty, severe material deprivation and low work intensity. Its formulation responds to the social integration objectives set by the European 2030 agenda. It permits a comparative analysis between European countries, while also addressing and refining the policies, strategies and measures that can be put into practice in the fight against poverty and social exclusion from a residential perspective.
The multiple linear regression analysis uses the AROPE indicator as the dependent variable and the variables related to the different dimensions that express how housing is structured as independent or predictor variables. Thus, it references the most influential theoretical and empirical analyses on housing configuration (Kemeny, 1991; Cortés, 1997; 2008; Feantsa, 2017; Oliván and Martínez, 2023). First and foremost, housing affordability is considered a fundamental dimension, encompassing not only the measurement of available aid to cover housing costs and the relationship between these and household income, but also the importance of different forms of housing tenure. This is a variable that has proven to be fundamental in processes of residential inequality, as well as in problems related to housing affordability (Kemeny, 1995; Murie, 2012b). According to the broad notion of the meaning of housing, its articulation also includes its physical attributes and how they respond to the residential needs of households. This leads to an examination of the dimensions related to habitability and the adequacy of the housing. While residential stability, referring to the degree of security of a given residential situation, constitutes another key dimension, its incorporation is not feasible given the lack of reliable and comparable quantitative data on key issues such as the incidence of evictions or subletting (Feantsa, 2022). Similarly, and for the same reason, it is impossible to include homeless and unoccupied individuals. Therefore, the analysis of the aforementioned dimensions refers to the population residing in dwellings. Thus, considering the relevance of the residential dimensions described and in accordance with the available quantitative information. The inclusion of the variables contained in Table 1 was proposed a priori.
Although the regression analysis included all of the proposed residential variables to offer a comprehensive view, the need to ensure the statistical validity of the model resulted in the exclusion of those variables having low explanatory power or those that did not meet the basic assumptions of the linear regression model. Therefore, the following variables were selected: the burden of homeownership, the percentage of social housing, the burden on the cost of housing in the market rental market, the percentage of the population living in overcrowded conditions, and the percentage of the population suffering from one or more substandard housing problems. These last two variables, belonging to the dimension of adequacy and habitability, were merged using the composite indicator called the “rate of severe housing deprivation”1, since its inclusion further improves the model’s significance.
Secondly, after ensuring the validity of the regression model, the aim was to classify and characterize how different housing systems were associated with varying degrees of social exclusion. For this purpose, a cluster analysis based on the K-means procedure was applied, since it grouped countries into clusters based on the residential variables extracted from the regression model, thereby evaluating their relationship with the AROPE indicator. This approach improves the understanding of how housing systems are linked to social exclusion, permitting the creation of a “map” of specific types of housing systems and social exclusion in the European Union. It was assumed that countries having more commercialized housing systems and governed by asset-based welfare have higher levels of social exclusion.
Defining the role of housing as a factor of social exclusion
The results of the regression model, reflected in the coefficients of determination (R²), confirm the close relationship between housing type and social exclusion. Specifically, the R² reveals that 63.2 % of the variation in AROPE is explained by the residential variables included in the model, a figure which, according to the adjusted R², is reduced to 56.5 %. This still manages to explain more than half of the variation in the AROPE indicator. Furthermore, the value 2.381 resulting from the Durbin-Watson statistic indicates that there are no autocorrelation problems.
The F value (9.459) and the significance level (0.000), as reflected in the ANOVA table, reaffirm that the model as a whole is highly significant.
Regarding the degree of contribution of each independent residential variable in the prediction of Y and the direction in which they affect (positive or negative correlation) the dependent variable (AROPE), the results obtained in the standardized beta coefficients reveal that the variable “severe housing shortage” is the one that most contributes to the explanation of the AROPE indicator (0.669), since its p-value is highly significant (0.001). The correlation between both variables is positive, indicating that when residential conditions are especially precarious, the probability of suffering poverty or social exclusion increases considerably. This means that the higher the percentage of the population experiencing severe housing deprivation, the higher the percentage of the population in situations of poverty or social exclusion (AROPE). Secondly, homeownership is a significant factor. The negative coefficient for this variable (-0.475) suggests that homeowners have a lower risk of poverty or social exclusion as compared to those who do not own their homes. The percentage of social housing would also be one of the most representative variables influencing the prediction of Y. In this case, the correlation with the AROPE indicator is also negative. The negative coefficient (-0.416) indicates that, keeping the other variables constant, the presence of social housing tends to decrease the risk of poverty or social exclusion as compared to low levels of social housing. Furthermore, the p-value of 0.013 suggests that this relationship is statistically and highly significant. Regarding the burden of market rental costs, the positive coefficient (0.391) indicates that a higher incidence of this situation among the population is strongly associated with a greater risk of poverty or social exclusion, since spending a large portion of one’s income on rent limits the ability to cover other basic needs.
The behavior of housing systems in the face of social exclusion
The confirmation of a significant relationship between housing configuration and the risk of poverty or social exclusion rate, measured through the AROPE indicator, allows us to classify and characterize the residential behavior of different European countries and their AROPE rates. Then, a cluster analysis was applied, and four distinct clusters were identified:
Cluster 1: primarily consisting of Eastern European countries (Slovakia, Romania, Hungary, and Bulgaria), as well as Baltic countries (Lithuania and Estonia) and, as an exception in Southern Europe, Greece. They represent highly commodified residential models, with a clear predominance of homeownership and minimal levels of social housing, which has been associated with serious problems of access to housing, as indicated by the higher incidence of inflated rental costs. Lower housing affordability combined with more precarious levels of its habitability and adequacy (severe housing shortage) leads to a trend towards higher AROPE rates, ranging from 17.6 % in the case of Slovakia to 32 % in the case of Romania.
Cluster 2: the largest and most heterogeneous conglomerate, including the countries of Southern Europe (Spain, Italy, Portugal, Cyprus, and Malta), some from Eastern Europe (Slovenia, Poland, and the Czech Republic), as well as Latvia, Ireland, and Luxembourg. Homeownership also plays a significant role in this cluster, but there is also a certain degree of social housing. This contributes to a more balanced tenure system, resulting in lower rates of oversupply in the market rental market compared to conglomerate 1. Levels of housing insecurity are also more moderate. The form and intensity of these housing dimensions are related to AROPE rates that do not reach the high levels seen in the previous cluster. The maximum rate here is 26.5 % (Spain). Within this cluster, there are certain exceptional cases such as Slovenia and the Czech Republic, whose AROPE indices are quite moderate (13.7 % and 12 %, respectively). This indicates that, despite the significant correlation between housing configuration and the AROPE index, variables linked to other dimensions could have a mitigating effect on social equality in the residential model in these countries. Thus, according to the figures provided by Eurostat in 2023, there may be a potential influence of the high employment rates in Slovenia and the Czech Republic (81.4 % and 75.1 %) or their moderate GINI coefficients (23 and 24.4), which are below the EU average (29.6).
Cluster 3: made up of Central European countries (Germany, France and Austria) and Nordic countries (Sweden and Finland), which represent less commodified housing systems. In this case, this implies mixed forms of tenure with homeownership registering the lowest levels in favor of a market and social rental, reaching levels that are much more significant than in the previous clusters. This greater balance in tenure corresponds to lower levels of market rent overburden. Furthermore, these countries have a lower incidence of severe housing shortage, resulting in AROPE rates that range from 15.8 % to 21.3 %. This is significantly lower than the peak reached in the aforementioned clusters.
Cluster 4: consisting of the Netherlands and Denmark. This cluster features highly de-commodified housing systems and, therefore, very high levels of social housing, combined with very low rates of extreme housing insecurity. In line with these residential patterns, their AROPE levels are below 20 % (17 % in the Netherlands and 17.9 % in Denmark). However, the potentially negative impact of the increasing processes of housing residualization and privatization on AROPE levels should not be ignored. This is especially relevant considering the overload in the free rental market, which registers the highest figures and homeownership levels, exceeding those of cluster 1.
The analyses performed demonstrate, on the one hand, how levels of poverty or social exclusion are closely linked to how housing is configured in each context. The most representative variables are those related to tenure (specifically homeownership and social housing) as well as excessive rent costs in the open market and the impact of severe housing insecurity. Severe housing shortages and excessive rent costs emerge as factors associated with a higher risk of AROPE (positive correlation), while homeownership and social housing are associated with a lower probability of being in a situation of poverty or social exclusion (negative correlation). However, the spatial representation of these variables from cluster analysis confirms that countries following more market-driven provision models which are focused on homeownership and in which problems of affordability, habitability and adequacy are greater, tend to display higher AROPE levels as compared to those countries with more redistributive residential models in which tenure is more balanced and residential well-being is greater (in terms of affordability, habitability and adequacy). This reinforces the theses presented by Kemeny (1995) and Barlow and Duncan (1994), which link commodified residential systems, based on the notion of housing as an asset, with greater social inequality. In this case, this inequality is expressed through the weight acquired by the AROPE index. This dynamic is especially detrimental to vulnerable groups such as ethnic minorities and young people. These groups often face greater difficulties in accessing housing due to factors such as discrimination or greater job insecurity. As a result, these groups are disproportionately represented in the private rental market, where high costs and insecurity exacerbate their risk of social exclusion and poverty. This is particularly relevant considering that most of the European countries analyzed lean towards housing models based on its conception as an asset (conglomerates 1 and 2).
At the same time, the research findings highlight the importance of designing more redistributive housing policies that promote affordability, habitability, and unitary rental systems to reduce inequality and improve living conditions. They also suggest the need to place greater weight on housing when measuring and analyzing social exclusion processes. This may be done by monitoring existing variables and incorporating others that help establish a more comprehensive and accurate diagnosis.
Allen, Judith; Barlow, James; Leal, Jesús; Maloutas, Thomas and Padovani, Liliana (2004). Housing and Welfare in Southern Europe. Oxford: Blackwell Publishing Ltd.
Alguacil, Julio (2006). Barrios desfavorecidos: Diagnóstico de la situación española. In: Vidal Fernández, F. (ed.). V Informe FUHEM de políticas sociales: La exclusión social y el estado del bienestar en España. Madrid: FUHEM.
Arbaci, Sonia (2019). Paradoxes of Segregation: Housing Systems, Welfare Regimes and Ethnic Residential Change in Southern European Cities. Oxford: John Wiley and Sons. Crossref.
Arundel, Rowan and Ronald, Richard (2021). “The False Promise of Homeownership: Homeowner Societies in an Era of Declining Access and Rising Inequality”. Urban Studies, 58(6): 1120-1140. doi: 10.1177/0042098019895227
Bengtsson, Bo (2001). “Housing as a Social Right: Implications for Welfare State Theory”. Scandinavian political studies, 24(4): 255-275. doi: 10.1111/1467-9477.00056
Blessing, Anita (2015). “Repackaging the Poor? Conceptualising Neoliberal Reforms of Social Rental Housing”. Housing Studies, 31(2): 149-172. doi: 10.1080/02673037.2015.1070799
Bosh, Jordi (2017). “La relación entre política de vivienda y emancipación residencial de la juventud europea”. Papers, 102(1). doi: 10.5565/rev/papers.2238
Bourdieu, Pierre (1979). La distinción: Criterio y bases sociales del gusto. Barcelona: Taurus.
Clapham, David (1995). “Privatisation and the East European Housing Model”. Urban Studies, 32(4-5): 679-694. doi: 10.1080/00420989550012834
Cortés, Luis (1997). Hablando sobre exclusión residencial. Cáritas Española Editores.
Cortés, Luis (2005). “La crisis de la vivienda”. Documentación Social, 138: 81-100. Available at: https://www.caritas.es/main-files/uploads/2005/01/DS100138-VIVIENDA-Y-ALOJAMIENTO.pdf, access October 13, 2025.
Cortés, Luis (2008). Nuevos y viejos problemas residenciales: Vivienda y exclusión. In: Bellet, C.; Ganau, J. and Llop, J. (eds.). Vivienda y sociedad. Nuevas demandas, nuevosInstrumentos. Lleida: Editorial Milenio.
Dewilde, Caroline (2017). “Do Housing Regimes Matter? Assessing the Concept of Housing Regimes Through Configurations of Housing Outcomes”. International Journal of Social Welfare, 26(4): 384-404. doi: 10.1111/j.1468-2397.2011.00811.x
Dewilde, Caroline (2022). “How Housing Affects the Association Between Low Income and Living Conditions-deprivation across Europe”. Socio-Economic Review, 20(1): 373-400. doi: 10.1093/ser/mwab003
Domínguez-Pérez, Marta; Leal-Maldonado, Jesús and Barañano-Cid, Margarita (2021). “Vivienda, transformaciones urbanas y desigualdad socioespacial en las grandes ciudades españolas”. Ciudad Y Territorio Estudios Territoriales, 53(M): 5-12. doi: 10.37230/CyTET.2021.M21.00
Elsinga, Marja and Hoekstra, Joris (2015). “The Janus Face of Homeownership-Based Welfare”. Critical Housing Analysis, 2(1): 42-51. doi: 10.13060/23362839.2015.2.1.174
Engels, Friedrich (1965) [1845]. La situación de la clase obrera en Inglaterra. Buenos Aires: Editorial Futuro.
Feantsa (2017). European Typology of Homelessness and Social Exclusion (ETHOS). Available at: https://www.feantsa.org/en/toolkit/2005/04/01/ethos-typology-on-homelessness-and-housing-exclusion, access October 13, 2025.
Feantsa (2022). Seventh overview of housing exclusion in Europe. Available at: https://www.feantsa.org/public/user/Resources/reports/2022/Rapport_Europe_GB_2022_V3_Planches_Corrected.pdf, access October 13, 2025.
Fields, Desiree J. and Hodkinson, Stuart N. (2017). “Housing Policy in Crisis: An International Perspective”. Housing Policy Debate, 28(1): 1-5. doi: 10.1080/10511482.2018.1395988
Forrest, Ray; Murie, Allan and Williams, Peter (1990). Home Ownership: differentiation and fragmentation. London: Unwin Hyman.
Gallent, Nick (2019). Whose housing crisis? In: Gallent, N. (ed.). Whose Housing Crisis? Bristol: Policy Press. doi: 10.51952/9781447346067.ch005
Giddens, Anthony (1998). The Third Way and Its Critics. Cambridge: Polity.
Hick, Rod; Pomati, Marco and Stephens, Mark (2022). Housing and poverty in Europe: Examining the interconnections in the face of rising house prices. Cardiff: Cardiff University.
Kemeny, Jim (1991). Housing and Social Theory. London: Routledge eBooks. doi: 10.4324/9780203413562
Kemeny, Jim (1995). From Public Housing to Social Market. London: Routledge.
Leal, Jesús (2017). “La vivienda imposible”. Cuadernos de Relaciones Laborales, 35(1): 11-14. doi: 10.5209/crla.55001
Lowe, Stuart (2004). Housing and Social Exclusion. In: Lowe, S. (ed.). Housing Policy Analysis, (pp. 120-155). London: Palgrave. doi: 10.1007/978-1-137-09061-4_5
Lux, Martin; Kährik, Anelli and Sunega, Peter (2012). “Housing Restitution and Privatisation: Both Catalysts and Obstacles to the Formation of Private Rental Housing in the Czech Republic and Estonia”. International Journal of Housing Policy, 12(2): 137-158. doi: 10.1080/14616718.2012.681574
Madden, David and Marcuse, Peter (2018). En defensa de la vivienda. Madrid: Capitán Swing Libros.
Martínez del Olmo, Almudena (2020). “El sistema de vivienda del sur de Europa: ¿continuidad o ruptura?”. RES. Revista Española de Sociología, 29(1): 153-180. doi: 10.22325/fes/res.2020.10
Murie, Allan (1997). “The Social Rented Sector, Housing and the Welfare State in the UK”. Housing Studies, 12(4): 437-461. doi: 10.1080/02673039708720909
Murie, Allan (2000). How can we end inequalities in housing? In: Pantazis, C. and Gordon, D. (eds.). Tackling inequalities. Cambridge University Press.
Murie, Allan (2012a). “Housing in the Welfare State: Partitioning Places and People”. Local Economy, 27(5-6): 480-485. doi: 10.1177/0269094212446295
Murie, Allan (2012b). Three: Housing, the welfare state and the Coalition government. In: Kilkey, M.; Ramia, G. and Frarnsworth, K. (eds.). Social Policy Review 24. Bristol: Policy Press. doi: 10.1332/policypress/9781447304470.003.0004
Norris, Michelle and Winston, Nessa (2011). “Home-ownership, Housing Regimes and Income Inequalities in Western Europe”. International Journal of Social Welfare, 21(2): 127-138. doi: 10.1111/j.1468-2397.2011.00811.x
Oliván, Fidel and Martínez, Elena (2023). Prevención y atención de la exclusión residencial. Factores explicativos. Provivienda.
Olsen, Gregg M. (2018). Housing policy, the welfare state and social inequality. In: Greve, B. (ed.). Routledge Handbook of the Welfare State. London: Routledge. doi: 10.4324/9781315207049-34
Pareja-Eastaway, Montserrat and Sánchez-Martínez, M.ª Teresa (2015). “El sistema de vivienda en España y el papel de las políticas: ¿qué falta por resolver?”. Cuadernos económicos de ICE, 90: 149-174.
Parkinson, Sharon; Wood, Gavin A. and Campbell, Iain (2024). “Labour and Housing Market Precarity: What is the Impact of Time-related Underemployment?“. International Journal of Housing Policy, 24(1): 1-22. doi: 10.1080/19491247.2022.2089083
Pisarello, Gerardo (2009). “El derecho a la vivienda como derecho social: implicaciones constitucionales”. Revista catalana de dret públic, 38: 43-66.
Potts, Deborah (2020). Broken cities: Inside the global housing crisis. New Jersey: Wiley.
Saiz, Albert (2023). “The Global Housing Affordability Crisis: Policy Options and Strategies”. MIT Center for Real Estate Research Paper, 23(1). doi: 10.2139/ssrn.4402329
Sassen, Sassen (1991). The global city: New York, London, Tokio. Princeton University Press.
Sennett, Richard (2018). Building and Dwelling. Ethics for the City. New York: Farrar, Straus and Giroux.
Sherraden, Michael (2003). Assets and the social investment state. In: Paxton, W. (ed.). Equal shares? Building a progressive and coherent asset-based welfare policy. London: IPPR.
Sherraden, Michael (2016). Assets and the poor: New American welfare policy. Routledge.
Silver, Hilary (2024). Social Exclusion. In: Orum, A. M. (ed.). The Wiley Blackwell Encyclopedia of Urban and Regional Studies. New Jersey: John Wiley and Sons Ltd. doi: 10.1002/9781118568446.eurs0486
Stephens, Mark (2020). “How Housing Systems are Changing and Why: A Critique of Kemeny’s Theory of Housing Regimes”. Housing, Theory and Society, 37(5): 521-547. doi: 10.1080/14036096.2020.1814404
Stephens, Mark; Lux, Martin and Sunega, Peter (2015). “Housing: Asset-Based Welfare or the ‘Engine of Inequality’?”. Critical Housing Analysis, 2(1): 22-31. doi: 10.13060/23362839.2015.2.1.173
Subirats, Joan (2004). Pobreza y exclusión social. Un análisis de la realidad española y europea. Madrid: Fundación “La Caixa”.
Tezanos, José F. (2001). La sociedad dividida: estructuras de clase y desigualdades en las sociedades tecnológicas. Madrid: Biblioteca Nueva.
Torgersen, Ulf (1987). Housing: The wobbly pillar under the welfare state. In: Turner, B.; Kemeny, J. and Lundqvist, L. (eds.). Between State and Market: Housing in the Post-industrial Era. Stockholm: Almqvist and Wicksell International.
1 The severe housing deprivation rate is defined as the percentage of the population living in overcrowded conditions and exhibiting at least one of the following housing deprivation measures: leaking roofs, no bathtub/shower or indoor toilet, or housing that is considered too dark.
Table 1. Analysis dimensions and variables
|
DIMENSIONS |
VARIABLES |
DESCRIPTION |
SOURCE/YEAR |
|
AFFORDABILITY |
Tenure regime |
Percentage of owner-occupied homes. Percentage of social housing. |
EUSILC 2023/HOUSING EUROPE |
|
Direct housing aid |
Percentage of public spending on direct aid to the population for housing as a percentage of GDP. |
Eurostat 2021 |
|
|
Housing cost burden according to tenure system |
Percentage of population living in households where total net housing costs represent more than 40 % of net disposable income. |
EUSILC 2023 |
|
|
Heavy financial burden of housing |
Percentage of the population that considers housing costs, including mortgage payments (installment and interest) or rent, insurance, and service charges (sewage disposal, garbage collection, regular maintenance, repairs, and other charges), to be a heavy financial burden. |
EUSILC 2023 |
|
|
Rent affordability rate |
Percentage represented by the cost of rent with respect to the household’s disposable income. |
EUSILC 2023 |
|
|
HABITABILITY |
Substandard housing |
Percentage of population that suffers from one or more of the following problems: a) leaking roofs, damp walls/floors/foundations or rot in window frames; b) insufficient lighting in home; c) no bathtub/shower; d) no indoor toilet with water flushing for exclusive use in the home. |
EUSILC 2023 |
|
Unfavorable residential environment |
Percentage of population affected by one or more of the following problems: a) noise from neighbors or the street; b) pollution, dirt or other environmental problems; c) crime, violence or vandalism in the area. |
EUSILC 2023 |
|
|
ADEQUACY |
Overcrowding |
A person is considered to live in an overcrowded household if the home does not have a minimum number of rooms equal to: • One room per home. • One room per couple of household members. • One room per single person over the age of 18. • One room per couple of single people of the same sex between the ages of 12 and 17. • One room per person for each single individual between the ages of 12 and 17 and not included in the previous category. • One room per couple of children under the age of 12. |
EUSILC 2023 |
Source: Author’s own creation based on data from Eurostat.
Table 2. Summary of Modelb
|
Modelo |
R |
R squared |
Corrected R squared |
Standard error of the estimate |
Durbin-Watson |
|
1 |
0.795a |
0.632 |
0.565 |
3.08508 |
2.381 |
a. Predictor variables: (constant), severe housing shortage, excessive cost of market rental housing, social housing, homeownership.
b. Dependent variable: AROPE.
Source: Own elaboration based on Eurostat data.
Table 3. Anova Results
|
ANOVAa |
||||||
|
Model |
Sum of squares |
df |
Mean square |
F |
Sig. |
|
|
1 |
Regression |
360.124 |
4 |
90.031 |
9.459 |
0.000b |
|
Residual |
209.390 |
22 |
9.518 |
|||
|
Total |
569.514 |
26 |
||||
a. Dependent variable: AROPE.
b. Predictor variables: (constant), severe housing shortage, excessive cost of market rental housing, social housing, homeownership.
Source: Own elaboration based on Eurostat data.
Table 4. Standardized beta coefficients
|
Coefficientsa |
|||||||||||
|
Model |
Non-standardized coefficients |
Standardized coefficients |
t |
Sig. |
Correlations |
Collinearity statistics |
|||||
|
B |
Stand. Error |
Beta |
Zero order |
Partial |
Semi partial |
Tolerance |
VIF |
||||
|
1 |
(Constant) |
27.278 |
5.195 |
5.251 |
0.000 |
||||||
|
Property |
-0.186 |
0.076 |
-0.475 |
-2.456 |
0.022 |
0.327 |
-0.464 |
-0.317 |
0.447 |
2.236 |
|
|
Social housing |
-0.240 |
0.088 |
-0.416 |
-2.718 |
0.013 |
-0.456 |
-0.501 |
-0.351 |
0.713 |
1.402 |
|
|
Overburdened housing costs, market rent |
0.187 |
0.070 |
0.391 |
2.680 |
0.014 |
0.430 |
0.496 |
0.347 |
0.786 |
1.272 |
|
|
Severe housing shortage |
0.938 |
0.232 |
0.669 |
4.046 |
0.001 |
0.643 |
0.653 |
0.523 |
0.611 |
1.636 |
|
a. Dependent variable: AROPE.
Source: Own elaboration based on Eurostat data.
Table 5. Results of cluster analysis
|
Clusters |
FINAL CLUSTER CENTERS |
COUNTRIES |
AROPE (%) |
|||
|
Property |
Social housing |
Market rental overload |
Severe housing shortage |
|||
|
1 |
87.01 |
1.84 |
35.78 |
6.51 |
Slovakia |
17.6 |
|
Romania |
32 |
|||||
|
Hungary |
19.7 |
|||||
|
Lithuania |
24.3 |
|||||
|
Croatia |
20.7 |
|||||
|
Greece |
26.1 |
|||||
|
Estonia |
24.2 |
|||||
|
Bulgaria |
30 |
|||||
|
2 |
75.02 |
4.56 |
23.73 |
3.83 |
Slovenia |
13.7 |
|
Portugal |
20.1 |
|||||
|
Poland |
16.3 |
|||||
|
Malta |
19.8 |
|||||
|
Luxembourg |
21.4 |
|||||
|
Latvia |
25.6 |
|||||
|
Cyprus |
16.7 |
|||||
|
Italy |
22.8 |
|||||
|
Spain |
26.5 |
|||||
|
Ireland |
19.2 |
|||||
|
Czech Republic |
12 |
|||||
|
Belgium |
18.6 |
|||||
|
3 |
59.82 |
15.34 |
16.38 |
2.30 |
Finland |
15.8 |
|
Sweden |
18.4 |
|||||
|
Austria |
17.7 |
|||||
|
France |
20.4 |
|||||
|
Germany |
21.3 |
|||||
|
4 |
65.10 |
25.45 |
39.40 |
2.15 |
Netherlands |
17 |
|
Denmark |
17.9 |
|||||
Source: Author’s own creation.
RECEPTION: October 28, 2024
REVIEW: April 22, 2025
ACCEPTANCE: July 28, 2025