The (great?) transformation of the rented market in the Brussels capital region and Amsterdam

(1)

The (Great?) Transformation of the Rented Market

in the Brussels Capital Region and Amsterdam

Liselotte van Vliet Student ID: 6314082

Graduate School of Social Sciences University of Amsterdam Amsterdam, The Netherlands

Thesis Supervisor: Prof. Dr. S. Musterd

(2)

Introduction

Over the last decades, the scholarly interest in understanding housing pro-vision regimes – and its transformation – has grown considerably. Here, one may argue that across the industrialised world, there appears to be a trend towards a greater reliance on market mechanisms in housing allocation, to-gether with a desire for a less dominant central government intervention. As a consequence, the publicly provided rented sector continues to play a significant role in the housing policy. Interestingly, however, there appears to be a tendency to prefer moving to a smaller social rented sector; that is more strictly targeted to low-income households. These transformations in the housing policy can be regarded as such major transformations, that one may consider them to be ‘regime changes’. That is, the transformations are so substantial, that the core principles of the housing policy are altered. Such a transformation of the regime is considered to be a driving force in shaping the housing patterns the urban housing market (Forrest and Murie, 2014). A shift towards liberalisation of the housing policy is suggested to lead to a sharper difference in the socio-economic profile of tenants, both in the private rented sector and in the social rented sector (Murie and Musterd, 1996).

The main challenge in assessing the effects of a housing policy on hous-ing outcomes, is that houshous-ing outcomes are impacted by a wide variety of factors outside the reach of the housing policy. That is, the economic, politi-cal and demographic developments influence the demand for housing within the urban rented market. At the same time, these economic developments and political transformations influence the housing policy decisions as well (Harloe, 1994; Wilensky et al., 1987; Murie and Musterd, 1996; Kemeny, 2006; Boelhouwer and van der Heijen, 1992). In the context of targeting low-income households, this raises the interesting question: is the housing regime an independent factor in influencing low-income housing opportuni-ties? Consequently, the problem statement of the thesis reads as follows.

(5)

4

Problem statement: How does the housing regime influence the access of low-income households to affordable rental housing? In investigating this problem statement, we select two relevant cases, namely Brussels and Amsterdam. As such, the contribution of the present thesis is a comparative case analysis of these two urban rental markets. The two cases in this study have similarities in terms of the wider economic, political and demographic context, and have pursued different housing poli-cies in the past. For instance, Amsterdam has a large social rented sector (Hoekstra, 2003), whereas Brussels has a large private rented market and a small social rented sector (van den Berg et al., 2007; Decker, 2008).

Specifically, as a guideline to answer the problem statement, the present thesis investigates the following three research questions (RQs).

Research question 1: What is the housing regime in Brussels and Amsterdam? How has the regime developed during since 2002?

Research question 2: What is the provision of affordable rental housing in Brussels and Amsterdam, and what are the changes in the housing supply since 2002?

Research question 3: How has the access to affordable rental housing for low-income households in Brussels and Amsterdam developed since 2002?

The methodology to answering these research questions is discussed in Chapter 3. Table 1.1 gives a concise overview of the precise sections in which each research question will be addressed.

The outline of the remainder of the thesis is as follows. Chapter 2 pro-vides an overview of literature relevant to the topic of the thesis. Chapter 3 gives an overview of the methodology employed in the thesis. Chapter 4 presents the case study analysis for Amsterdam, and Chapter 5 presents the case study analysis for Brussels. Chapter 6 gives a detailed comparison of the findings of both the cases. Finally, Chapter 7, answers the research questions, translates the answers to an answer to the problem statement, and gives recommendations for future research.

Ch. 2 Ch. 3 Ch. 4 Ch. 5 Ch. 6 Ch. 7 Research Question 1 2.1 4.1 5.1 6.1 7.1 Research Question 2 2.2 4.2 5.2 6.2 7.2 Research Question 3 2.3 4.3 5.3 6.3 7.3

Problem Statement 7.5

Table 1.1: Problem statement and research questions addressed by the different chapter sections.

(6)

Chapter 2

Literature

This literature reviews the key concepts and causal narratives related to the housing policy regime and its relationship to socio-teneurial stratifica-tion. Housing patterns are influenced by a wide variety of processes and developments outside the reach of housing policy. The economic develop-ments, demographic arrangements and the welfare state are often pointed out as important factors shaping the socio-teneurial stratification. This socio-economic context is discussed first (2.1). Subsequently, the concept of a regime and its applications in housing policy is described (2.2). The two sections thereafter translate the housing regime concept into character-istics of the rented market (2.3), and discuss the concept of socio-teneurial polarisation (2.4). The chapter is concluded by a summary of the findings (2.5).

2.1 The socio-economic context

The aim of this section is to map some of the theories that describe the key causes of patterns in housing outcomes that are discussed in the litera-ture. The study of housing outcomes is divided into ‘convergence’ theories and ‘divergence’ theories. The convergence theories stress the importance of global trends in shaping the housing outcomes and policies. Economic developments, such as economic restructuring and labour market develop-ments, are key to understanding why housing outcomes and housing policy agenda’s change. The assumption is that most housing systems are influ-enced by these developments in more or less similar ways (Harloe, 1995; Lawson, 2003).

The other approach are the divergence theories. These emphasise path dependence and the singularity of a specific housing context. These diver-gence theories are concerned with explaining the differences between regions. The housing regime is a sub-category of the divergence approach: it aims to explain the differences between countries or regions.

(7)

2.1 | The socio-economic context 6

Most empirical studies assume that there is a combination of conver-gent, global economic, and diverconver-gent, situational and path dependent, and perhaps coincidental forces influencing the housing outcomes in a specific region.

2.1.1 Economic developments

Spatial patterns are often considered to be the ‘mirror of processes at work in society’. Income inequality is a significant factor in explaining housing out-comes across groups. The level of income inequality seems to have increased in the past years (Vanneste et al., 2008, p. 173).

Sassen (1991) observes an increasing labour market polarisation in the World Cities. The World Cities are becoming nodes in the global economy. This attracts the headquarters of the multinational corporations, financial services such as banks, and political centres of decision making. Connected to these institution’s is an increasing offer of well-paid jobs for the bankers, lawyers, lobbyists and decision makers in these cities. Meanwhile there is also an increasing demand for low-paid labour in the service sector. The growth of high-paid jobs in cities increases the housing demand for expensive housing in these cities, according to (Sassen, 1991).

This causes a competition for housing at the expense of low-income fam-ilies. Gentrification, evictions, conversions and mortgage foreclosures neg-atively impact the housing security and housing availability of low income households. More people are pushed into marginal housing positions. Mi-norities are more often negatively affected by economic restructuring. There is an expansion of the informal economy, particularly among new immigrant groups (Sassen, 1991).

Wealth is becoming more unequally distributed in society, this affects housing access. The 2008 financial crisis has, unlike other economic crises, not reduced wealth inequality; the wealth inequality continued to increase after the crisis. The wealth inequality also plays a role in re-producing inequality in a different access to the housing market. Family assets may play a role in peoples ability to purchase a home. Such ‘upgrading’ of the wealthy middle class, may reduce the access of low-income households, as home-ownership becomes further out of reach for those who (1) cannot take on a high mortgage, or (2) do not have access to other sources of wealth (Velthuis, 2011)

2.1.2 Demography

Family formation and expectations thereof have undergone a drastic and structural change since the 1970s. “Demographic and social changes such as ageing populations, increased marriage breakdown, and single parenthood have also swelled the numbers of poor” (Harloe, 1995, p. 37). There are more

(8)

2.1 | The socio-economic context 7

divorces, more single-person households, more single parents and people marry at a later age. In the post-war history, most advanced industrialised countries experienced a relative decline of the traditional nuclear family. The causes are demographic, sociological and economic: the greying of society, changing values towards cohabitation and increasing affluence and individual choice, according to Paris (1995).

Other types of household structures have become more prevalent: aged singles or dual person households, co-habiting childless couples, students, singles, divorced and remarried couples. The consequence is that the aver-age household size decreases. This also changes the demand for the housing types. The non-traditional family is associated with urban areas. Ethnic diversity increased in most European countries as they moved from an e-migration country to an in-e-migration country.

These demographic changes in household formation and household struc-ture are, according to Paris (1995), autonomous variables, analytically sep-arable from other dimensions of social change that impact the socio-tenurial polarisation. The changing demographic structure increases the pressure on the housing market, especially in urban areas, which leads to a higher com-petition for adequate housing. Especially the poor are disadvantaged by a lack of choice, when even the residual private market becomes increasingly unaffordable. The call for measures to secure decent housing for low-income households increase as their condition on the housing market becomes in-creasingly untenable (Paris, 1995).

2.1.3 Welfare state

The importance of economic trends is widely accepted as significant to un-derstanding patterns of low-income households. However, there is also a wide body of literature that stresses the importance of welfare state ar-rangements (Musterd and Ostendorf, 1997). The welfare state can have a dampening effect on income and wealth inequality and as a consequence dampen the effect on housing market inequality.

Hamnet (1996) argues that there is not a global uniform development of income polarisation as Sassen (1991) observed. Rather than polarisation, he observed that there was an upgrading of the top incomes in London. The incomes at the bottom stagnated in London. The explanation for the difference in income polarisation in London and New York is, according to Hamnet, that the British welfare state provides more income securities to low-income households than the American welfare state (Hamnet, 1996).

Malpass (1983) believes that the transformation in the housing policy regime is impacted by discourses of welfare entitlement. Liberal welfare states tend to stigmatise towards welfare recipients. These negative dis-courses towards welfare provision in itself creates a negative attitude towards the socially rented sector according to Malpass (1983). According to Murie

(9)

2.2 | The regime approach 8

(1996), the changes in the housing policy are part of a wider ideological shift in welfare provision. The ideas about welfare entitlement and the legitimacy of state interventions have shifted towards a preference for a stronger market and a less involved state.

Michael Harloe (1994) traces the causes of the housing policy regime change back to the economic context. The specific economic situation after the Second World War created a very specific context which contributed to the growth of the social rented sector. The housing shortage after the Second World War was ‘unpolarised’, it affected an ‘extensive spectrum of income and socio-economic groupings’ (Harloe, 1994, p. 38). This is key to understanding why many governments opted for mass housing provision. “Housing was targeted to sections of the population who were in housing need but also had, temporarily, some political leverage” (Harloe, 1994, p. 42). “From the mid-1970s, of the economic, political, and ideological systems which sustained the post-war welfare states has put pressure on the public resources for social rented housing. The process has weakened the political support for this sector and strengthened the position of those who oppose a collectively owned, semi-socialised service” (Harloe, 1994, p. 38).

Wealth- and income inequality influences the agenda for housing pol-icy. Inequality undermines the perceived community of interest between the majority and minority (Kleinman, 1996). As more people exit the mass provision, the public service provision becomes increasingly a privilege for the poor. The result is a bifurcation (split) of social policy: a separation of policy concerns for the masses and the disadvantaged. The general hous-ing policies support the houshous-ing conditions of the majority, through policies of output creation, access to home-ownership, stimulating asset acquisition, ensuring economic stability, providing environmental quality and stressing individual choice. Cheaper housing is not in the interest of home-owners and is seen as a risk to economic stability. The policies for the poor housed minority focus on the conditions on the housing estates, homelessness, so-cial isolation and chronic economic insecurity. There is increasingly a policy focus that stresses local solutions for urban problems and social exclusion, whilst not addressing the causes of social exclusion rooted in the extra local context: labour market security and integration, the macro-economic policy and the devolution of responsibilities from the state to local communities (Kleinman, 1996).

2.2 The regime approach

The purpose of this section is to clarify the concept of a housing regime and how it is used to compare housing policy. The regime presupposes that there are underlying ‘principles’ of a housing policy.

(10)

2.2.1 What is a regime?

The regime is aimed at explaining the differences between complex systems. The concept of a regime looks at the patterns and principles underlying policy choices. The concept of a welfare-state regime is clarified by Esping-Andersen (1990) in the Three Worlds of Welfare Capitalism.

The concept of a regime in public policy literature is clarified by May and Joachim: “In broad terms, regime perspectives provide a lens for considering the interplay of ideas, interests, and institutional arrangements. As a de-scriptive lens, regime perspectives enable a backward mapping of governing arrangements for a given policy problem identifying the contours of a policy regime.”(May and Jochim, 2013, p. 427). “In terms of governing arrange-ments, ideas are important because they provide direction for governing and serve as organising principles” (May and Jochim, 2013, p. 435).

The concept is based on theories of provision by Karl Polanyi and Roger Titmuss and is used as a tool to compare welfare state systems. The three regime-clusters are based on Karl Polanyi’s Great Transformation (1944) and R. Titmuss (1974) ‘What is social policy’ Polanyi (1957) lays out three essential forms of social provision: the market, the state and reciprocity. (Titmuss) reformulates these three models of redistribution: (1) the resid-ual model, (2) the industrial-achievement model and (3) the institutional redistributive model of social policy. In the residual model, there is only a state intervention when both the market and family are unable to meet individual needs. Merit, work performance and productivity are the ba-sis of access in to social welfare in the industrial-achievement model. The institutional-redistributive model provides universalistic services outside the market provision on the basis of need Titmuss (1974, p. 154).

Commodification is when something is treated as a market good. The access to this good depends on ones financial means. Decommodification occurs when the policy treats the good as a social entitlement. The access to this good is determined by need or citizenship. He theorises that there is a continuum of service provision between the pure form of market provision and pure state-allocation. “But, whereas the pure forms, as ideal types, are useful analytically, the position adopted in this article is that of conceptu-alising greater or lesser tendencies along a scale with the polar opposites of commodification and decommodification” (Doling, 1999, p. 156). In a residual welfare model, there is a value for individual choice and initiative: individuals are free to choose their schools, their health care provision, and pay a market price for this. There is government provision, but ideally this is restricted to those who are not able to access essential services without government support. This leads to a commodification of a good: the good is treated as a market good rather than an entitlement upon citizenship. In universal welfare models, these services are provided to all citizens on the basis of a cost price or reduced rate. The goods are considered to be an

(11)

Figure 2.1: Model of the trichotomy of welfare states in Esping Andersen Esping-Andersen (1990). The ‘civil society’ may also be labelled as ‘family’ or ‘Gemeinschaft’.

entitlement rather than a market good. This creates a decommodification of the good. Esping-Andersen writes: “The criterion for social rights must be the degree to which they permit people to make their living standards independent of pure market forces” (Esping-Andersen, 1990, p. 3).

Esping-Andersen used these theories to classify welfare-states into three ideal-type welfare-state regimes: the liberal welfare state-regime, the corpo-ratist welfare state-regime and the social democratic welfare state regime. In the liberal welfare state, there is an emphasis on market provision. Wel-fare is provided as a residual service. The United States and the United Kingdom are examples of countries with liberal welfare state regimes. The social-democratic welfare state regime has universal welfare provision: ac-cess is granted as an entitlement to all citizens. Sweden is an example of a social-democratic welfare state. Corporatist welfare states have a high level of service provision, but there is a tendency to reproduce rather than reduce existing social inequalities. The corporatist welfare state place an importance to family life. Germany is an example of a corporatist welfare state (Esping-Andersen, 1990).

A defining criteria of liberal or residual welfare provision is that the welfare provision is restricted to low-income households. This restriction is ensured via means-testing. The means-test is an income or savings test in which it is determined if the person would be able to afford market provision. The access to the service is denied if the person can access to the service at a market rate. Such means tests aim to activate people for participation in the labour market. The means-testing makes a distinction between poverty caused by in-activeness and poverty caused by diseases. Katz (1989) calls this the distinction between the able bodied and the unable bodied. Those who could be able to work and provide for themselves should not rely on welfare provision. Those who are unable to work, due to age, sickness or disability, are can legitimately access welfare provision. These criteria for accessing a service are highly centred around notions of distributive

(12)

jus-2.2 | The regime approach 11

tice: should the state intervene to ensure a minimum quality, or refrain from intervention? “To a social democrat, reliance on the market for the basic means of welfare is problematic because it fails to provide inalienable rights and because it is inequitable. To a laissez-faire liberal, reliance on the welfare state is dangerous because it cripples freedom and efficiency” (Esping-Andersen, 1990, p. 4).

2.2.2 Housing as a field of policy

Some argue (Kemeny, 2006; Lennartz, 2011; Harloe, 1995; Boelhouwer and van der Heijen, 1992) that housing provision has some essential differences compared to the other domains of welfare provision. Harloe (1995) argues, what essentially sets apart housing from other domains, is the double func-tion of housing as both a capital good and an essential good (Harloe, 1995). The function of housing as an essential good means that in normal times, alike with food, even the least off will have some form of housing provision in the absence of government allocation. This unlike education, health-care or pensions, which cannot be profitably provided to the less off without government support. The second feature as a capital good, makes housing decommodification more controversial than the decommodification of ser-vices such as education, pensions or health-care, because private property is core to the values of capitalist societies (Harloe, 1995).

Bengtsson (2001) equally sees the capital intensiveness as a distinct fea-ture of housing provision and argues, that in addition, a core difference between housing and other welfare domains is that housing is much more embedded in the wider structure of society than education or health-care. Housing is not only impacted by the wider structure of society, it is also an factor of influence to the wider functioning of the welfare state.

Tenure is a concept that describes the ownership structure of a dwelling. Tenure is categorised into three major formal types: owner occupied, so-cial rented and private rented. However, there are also variations in rules and regulations within a tenure type (Doling, 1999). Doling (1999) em-phasises that a single tenure can have different meanings depending on the arrangements towards the tenure: a tenure type can be more commodified or decommodified depending on the rules and regulations regarding access and exit of this tenure. Therefore, it is not sufficient to look at the crude numbers of housing per type of housing provider to understand the specific nature of a housing regime; it is also important to compare the qualities of the tenure that provided (Doling, 1999).

Boelhouwer and van der Heijen (1992) emphasise the outcomes of hous-ing provision and consumption is impacted by factors outside the reach of housing policies. This is much less the case with education or health care. A consequence is that housing policy alone is not sufficient to understand the nature of housing provision (Boelhouwer and van der Heijen, 1992).

(13)

2.3 | Typologies of housing regimes 12

2.3 Typologies of housing regimes

Several scholars have come with contributions to transfer the analysis of ’welfare provision regimes’ to the domain of housing policy. This section gives an overview of three main contributions in this field. Each of the au-thors splits the concept of provision in two dual categories, whereby the one category treats housing as a commodified good and the other as a decom-modified good. The actual housing policy is in a spectrum between pure commodification and pure decommodification of housing. The first contri-bution is from Lundqvist (1986), he provides an analysis of the aims and scope of housing intervention in relation to decommodification. The second contribution is by Harloe (1995), the criterion to measure decommodifica-tion for him is the growth and decline of public and private housing provision and allocation in relation. The third contribution is from Bengtsson (2001), he looks at the narrative of the entitled in relation to decommodification. The fourth contribution is from Doling (1999), he looks at the level of de-commodification within each tenure. Kemeny (2006) analyses the concept of decommodification in relation to the competition between public and private housing providers.

The central feature in the classification of Lundqvist is the desired role of the government in the housing provision. He distinguishes between a comprehensive and a supplementary role for the government. The supple-mentary role of the government is limited in scope: ‘its operations are geared towards meeting particular needs and solving specific problems’ (Lundqvist, 1986, p. 12). The supplementary intervention considers the housing essen-tially as a commodity: the market is considered the main mechanism for housing distribution, and housing should be bought and sold as freely as possible, the level of housing welfare that can be enjoyed is dependent on the household’s ability to pay. A consequence is that people who are able to pay also get the best available housing. The allocation of housing is seen as most efficient, but its defence is also ideologically in nature (Lundqvist, 1986, p. 14). Most supplementary approaches do allow interventions when the market is not functioning properly. The interventions are aimed at ‘helping households in real need’ to reach some politically defined ‘decent’ standard of housing, and interventions are seen as ‘temporary’ (Lundqvist, 1986, p. 14). A comprehensive policy is on the other hand “not confined to regulating, supplementing or restraining the operations of an independent and normally well-functioning housing market” (Lundqvist, 1986, p. 12). The aim of a comprehensive policy is to maximise the welfare of the entire population. The government exercises significant control over the volume, timing and location of residential building for all income levels and all types of buildings. The governmental policy works alongside market allocation. Housing is considered to be a social right for all. The need is a central feature of welfare entitlement; This need is defined by household size, health, age,

(14)

and income. Lundqvist concludes that the debate on conversions is essen-tially an ideological debate about the role of the state in economic and social affairs (Lundqvist, 1986). The defence for comprehensive housing policies is also ideological in nature (Lundqvist, 1986).

Harloe (1995) distinguishes between two models of housing provision: the residual model and the mass model of housing provision. Mass housing provision is a form of universal provision, because it is not specifically tar-geted to the poor: “The abnormal form of provision has been mass social rented housing”. Typically this has involved large-scale programs of social rented housing, much less closely targeted to the poor. Such housing has frequently been managed in a rather less controlling or paternalistic way. Little or no means-tested subsidies have played a small part in such pro-grams. Instead, they have been assisted with ‘indiscriminate’ bricks and mortar subsidies (Harloe, 1995, p. 523). “The poor were not a strategically significant group in relation to mass social housing. Mass social rented hous-ing was at best a very flawed form of universalistic provision, since it tended, by one means or another, to exclude many of those who were poorest and in the greatest housing need” (Harloe, 1995, p. 523). The residual model of social housing provision involves small-scale building programmes, targeted to the poor. Historically, they were closely linked to slum clearance. More recently they have served the ‘new urban poor’, many of whom are outside the labour market and excluded from private market provision. Residual so-cial housing tends to be a more or less stigmatised form of provision which accommodates politically, socially and economically marginalised groups. It therefore shares many characteristics attributed to the residual welfare states (Harloe, 1995, p. 523).

Doling believes that the size of the public and private rented market is not sufficient to understand whether housing is treated as a commodified or decommodified good. A single tenure can be more or less (de)commodified as a consequence of the rules and regulations of allocation and exit Dol-ing (1999). A distinction with (Harloe, 1995), is that DolDol-ing’s concept of decommodification is not structured around the tenure-type, but the spe-cific rules and conditions tied to each tenure. Doling (1999) distinguishes between tenure types and tenure forms. The three tenure types are the owner-occupied type, the social rented type and the private rented tenure type. These are general and invariable. Tenure forms are specific forms that this type takes on a specific context, these are variable across contexts (Doling, 1999, p. 158). The rules of access and the rules of exit determine the degree of security and access to a tenure type.

The essence of decommodification is that welfare is available as a right founded in citizenship rather than a right founded in wealth or income (Dol-ing, 1999, p. 157). Within a tenure-type there can be more or less decom-modification. Access criteria to an accommodation type can be financial and non-financial. The rules of access can be regulated via means-testing:

(15)

Concept Commodification Decommodification Aim and scope Supplementary: market for

the masses, public rented housing for low-income households

Comprehensive: interven-tion should benefit all households

Size of the public rented sector

Stable, small Large-scale construction

Housing subsidy Subject subsidies (housing allowances)

Object subsidies (bricks and mortar subsidies)

Allocation Means-testing Accessible to broad groups Mass Open to median income

groups

–

Table 2.1: Concise overview of housing-topology concepts, as related to commod-ification, and decommodification.

the applicant tenant needs to show that the income is not too high or too low to be admitted into the rented sector. An example of rule of exit is for example when the household needs to move out when the income has surpassed a certain level.

The main feature determining the level of commodification in Kemeny’s (2006) analysis is whether or not the public housing providers are allowed to compete with private housing providers. Kemeny (2006) argues that a commodification in the housing system does not need to imply that the government withdraws from housing intervention, but actively intervenes in the housing allocation. He classifies the regimes into dual, unitary and inte-grative rental system. The defining feature is the extent to which the social rented housing sector may and can compete with the private rented housing. The competition is based on the idea of a mixed economy: there are free market mechanisms, but the government is one of the providers competing with the private sector. A dual rental market is a housing market in which the private and public provision of housing is segmented. There is no com-petition between the private and public housing; the social rented sector is reserved for those with a demonstrable need and who cannot satisfy their housing needs via the market provision. Means-testing is a mechanism to keep the markets dual. The state intervention serves to ‘keep the rental mar-ket as a sheltered preserve for profit providers, protected from competition by non-profit providers’ (Kemeny, Kersloot, and Thalmann, 2005, p. 858). In unitary rental markets, the barriers for the non-profit sector to compete with the profit sector are removed; in integrated rental markets there is an effective capacity to compete. The market-state nexus is not defined by the level of intervention of the state of the housing market, but by the influence

(16)

2.4 | Residualisation 15

that the non-profit sector is allowed and is capable of having on the housing provision.

The types of housing provision regimes are described by the above au-thors with a differing emphasis. But a red-wire in each of the accounts is that the defining features for classifying a housing policy are: (a) the state/market mix in housing supply, (b) the rules for housing allocation, and (c) the scope and purpose of housing intervention. The significance of the method of housing allocation are mentioned in all housing regime classi-fications. Means-testing and targeted allocation in the public rented sector are associated with a commodification of housing. A residual housing regime does not necessarily mean that the central government has withdrawn from intervention in housing provision (Kemeny, 2006). The liberalisation can also consist of a changing scope and purpose of government intervention within the provision system. The purpose of housing intervention is in a mass or comprehensive regime often defined in terms of the wider economic and social functioning of housing in society.

2.4 Residualisation

“Increasingly, social housing accommodates a narrower and narrower section of society the poor and those social groups excluded from access to main-stream private market provision” (Harloe, 1994, p. 411). Saunders (1990) observes the same trend: “the trend, therefore, seems to be that tenure divisions are coming to reinforce and express existing social divisions. As the haves move increasingly into home ownership, the have-nots become ge-ographically concentrated in public renting”. And within the sector there is also a difference: “the poorer and least popular council estates (where few tenants want to buy and even fewer can afford) will increasingly come to be seen as the dumping ground where the problems and causalities of our society can be concentrated, controlled and forgotten” (317). Malpass (1983 p. 44) argues “as the low-paid, the unemployed, the elderly, single parents, the disabled and others who were so disadvantaged in the housing market that they were unable to obtain adequate accommodation privately” (p. 240). The residual population is described in “selling the welfare state” (Forrest and Murie, 1988) as the political powerless, economic marginality and social undesirability (p. 83). This group is less able to mobilise itself collectively to bargain for better housing conditions.

2.4.1 Residualisation

Residualisation is, according to Malpass (1983), a process in which the so-cial rented sector is “largely, if not completely, connected to those amongst the low paid, the unemployed, the elderly, single parents, the disabled and

(17)

2.4 | Residualisation 16

Decommodification Residualisation

Expansion social rented sector Contraction social rented sector Good technical quality dwellings Declining technical quality dwellings

Socially mixed Concentration low-income households Open to broadly defined groups Means-tested access to social rented sector

Table 2.2: Concepts in decommodification as related to residualisation.

others, who were so disadvantaged in the housing market that they were unable to obtain adequate accommodation privately”.

Arbaci (2007) considers residualisation as a concept that can be applied to both private and social rented housing; while other authors (Forrest and Murie, 2014) only refer to the developments in the social rented sector when using the concept of residualisation. The authors of ‘selling the welfare state’ (Forrest and Murie, 2014) divide the concept residualisation in five potential dimensions: (1) the size of the social rented sector; (2) the nature and quality of the housing stock; (3) the characteristics of the tenants and the level of social mix within a tenure type; (4) the nature of the policy intervention and (5) the selectivity of the housing allocation (Forrest and Murie, 1988). Residualisation is a process whereby the social rented sector changes from a mixed tenure to a low-income tenure. Indications of residualisation are: contraction/expansion building output, technical quality buildings, level of social mix and concentration low income households, changes in the entry and exit requirements (Forrest and Murie, 2014).

2.4.2 Filtering

The tightness of the housing market influences the competition for a certain housing type or quality. This can lead to neighbourhood filtering: people with a low income move to the areas they can still afford, these areas are often deteriorating as a consequence of disinvestment.

An inferior quality of the housing stock is a characteristic of a residual housing sector (Meurs and Kempen, 1995). The marginal role of social hous-ing is underscored with disinvestment in the buildhous-ing stock. The financial position of social housing can be negatively impacted by budget cuts, vacan-cies and rent arrears. A lack of investments in both maintenance and repairs undermines the quality of the building stock in the long run. The process of disinvestment has been carefully theorised by W.C. Grigsby (Megbolugbe, Hoek-Smit, and Linneman, 1996). His analysis focused on neighbourhood disinvestment, filtering and decline. Grigsby saw neighbourhood decline as the effect of the choice to no longer invest in the area (Megbolugbe, Hoek-Smit, and Linneman, 1996).

(18)

2.5 | Chapter conclusions 17

The laissez-faire approach to housing will ultimately result in stronger cycles of investment and disinvestment. The location and dwelling types are important components to the commodity-value of the home. Neighbour-hood disinvestment and a deteriorating housing stock will reduce the value of all involved dwellings at that location of the metropolitan housing market. The tightness of the sub-market impacts the competition for a housing type, and therefore also influences the price-dynamics in the housing market. A market-driven housing context is characterised by a predominance of new housing in higher quality, private sector housing sub-markets. Consequen-tially, the supply of high-quality housing exceeds the demand. This leads to a chain system of housing succession whereby the higher-income house-holds move to the newly built high-quality sub-markets and the lower-income households move to the areas abandoned by the high- and middle income households. This results in neighbourhood filtering. The lowest-income groups have no choice but to live in the housing they can afford. Municipal zoning and luxury sub-markets exclude low-income households, thus concen-trated causing a further restriction in their residential choices. This leads to areas of spatially poverty, which are facing even more disinvestment, as both the residents themselves are unable to invest in the upgrading of the area, and the owners fear that their investment in redevelopment will be obsolete if the other neighbours do not follow suit (Galster, 1996). Grigsby believed that enduring investment in the quality of the area and building stock could smoothen the cycles of investment and disinvestment.

2.5 Chapter conclusions

The wider processes of welfare state restructuring, labour market inequality and labour market upgrading and changing household patterns influence the social stratification in society which is reflected in the tenure stratification. The housing regime is also influenced by this context. “Demography is not uni-vocally accepted as a cause for the increasing pressure on the low-rent private rented sector. Others stress that there was also, simultaneously, a reduction of options in the low-rent sector, as a consequence of slum clear-ance, lack of new investment, urban renewal, and rent policies” (Murie and Musterd, 1996, pp. 510 - 511).

The above mentioned literature focuses mostly on global processes, but the local institutional political, economic and social setting are of importance to understanding how context, housing regime and tenure stratification re-late to another. In addition, a persons individual housing context is influ-enced by many coincidental or individual factors other than income. Hous-ing preferences, individual networks, the individual housHous-ing needs, other household characteristics, and coincidence and chance are of influence to a person’s housing context.

(19)

Chapter 3

Methodology

This section contains an overview of the methodology used for this master thesis. The topics discussed are: the research design, the subjects and sam-ple, the unit of analysis, the research questions and hypothesis, the variables and data sources, the validity and the reliability, and the limitations. The research design is a comparative case study.1

The theoretical population of this research is urban housing markets in advanced welfare states. The period of the investigation is from the year 2002 until the year 2012. The unit of analysis are the social rented sector and the private rented sector.

The selected cases are Amsterdam and the Brussels Capital Region. It is ‘diverse case selection’: they are extreme cases when it comes to government intervention in the housing provision. In the past, the Netherlands has invested large sums in the development and expansion of the social rented sector, particularly in the big cities (Hoekstra, 2003). Belgium has shifted its focus on the owner-occupied sector and middle class suburban housing, rather than the promotion of affordable urban rental housing (Decker, 2008). In Chapter 1 we raised the following question: is the housing regime an independent factor in influencing low-income housing opportunities? Con-sequently, derived from this question, we defined the problem statement of the thesis as follows.

Problem statement: How does the housing regime influence the access of low-income households to affordable rental housing?

1 _{Comparative research differs from cross-country research in which individual cases are}

treated. Cross country research, such as those commissioned by the OECD and EU, do not provide in-debt case descriptions of the individual components of the variables. Another approach is the in-debt single case study, in which the developments and qualities of a single case are described. This type of research looks at the changes and institutional developments in a single context. The comparative research operates through the contrasting of contexts. The aim is to understand the mechanisms of similarity and difference between cases (Oxley, 2001).

(20)

19

RQ 1 RQ 2 RQ 3

Housing regime Housing provision Low-income pat-terns

Variables Allocation social housing,

Housing stock by tenure,

Income profile by tenure,

Scope and subsidy, Housing supply by tenure,

Affordability by tenure,

Aim of housing pol-icy.

Rental housing by price.

Waiting lists.

Data sources Secondary litera-ture. Housing Survey (Ams.), Housing Survey (Bxl.), BGHM statistics (Bxl.), BGHM statistics (Bxl.), Observatory Rental prices (Bxl.). Observatory Rental prices (Bxl.)

Table 3.1: Variables and data sources investigated per research question (RQ).

The central hypothesis is that the housing regime influences the oppor-tunities of low-income households on the urban rental market through the provision of affordable housing. Specifically, as a guideline to answer the problem statement, the present thesis investigates the following three re-search questions (RQs).

Research question 1: What is the housing regime in Brussels and Amsterdam? How has the regime developed since 2002? Research question 2: What is the provision of affordable rental housing in Brussels and Amsterdam, and what are the changes in the housing supply since 2002?

Research question 3: How has the access to affordable rental housing for low-income households in Brussels and Amsterdam developed since 2002?

Each of the three research questions is investigated with respect to three variables. A concise overview of the variables and investigated data sources is given in Table 3.1.

With regard to validity and reliability, it is acknowledged that case study research is often critiqued for scoring low on its external validity. In fact, the comparative case study method is keen on pointing out the difficulty of transferring the knowledge of the one context to another. Housing outcomes are highly embedded in the cultural, demographic, economic and political context of a city. Nevertheless, case study analysis that extend beyond the individual case by evaluating on existing theories or developing new theories.

(21)

20

Comparative case study analysis has the room for an in-debt investigation of the traits and qualities of its constructs. This greater detail provides a window of opportunity to add to the internal validity of the research. This thesis makes use of existing statistical data from the Dutch housing survey and the Brussels housing observatory. Both these data sources are based on sample interviews. The Dutch housing survey publishes the 95 per cent intervals of reliability with the results, but the Brussels housing observatory does not. As such, it should be assumed the measuring might contain some inaccuracy.

A limitation of the study is that many low-income housing patterns are ‘out of sight’ from the regular housing census. The poorest are not part of this survey: the homeless, the illegal. The study is only confined with the measurement of official forms of housing, while it is known that in both cases there is a substantial ‘grey’ housing market. The process of ‘displacement’ to municipalities outside the studied area are not included in the research. Also, the ‘housing regime’ is a rough classification of the national housing policy and does not encompass the full complexity of all the instruments that are available. There are many aspects that affect housing policy or that have an influence on housing outcomes that are outside the reach of this study, that are nevertheless potentially interesting. Individual housing outcomes are impacted by a wide variety of factors. Isolating a single causal variable is often not possible as is provided in an experimental setting.

The geographic limitations is that this study does not take into account some of the regional dimensions of the housing market. The boundary of the case Amsterdam is set as the municipality Amsterdam. The housing cor-porations in the agglomeration of Amsterdam cooperate in the allocation of social rented housing. This agglomeration contains both municipalities that are within the urbanized area such as Diemen, commuting cities such as Almere and small villages such as Beemster. Some of the municipalities around Amsterdam are on average wealthier, such as Amstelveen, and other areas are on average less wealthy than Amsterdam, such as Zaandam. The geographic boundary of the Brussels case is the Brussels Capital Region. It is recognised that the urbanisation of Brussels extends beyond its admin-istrative boundaries into both the Flemish and Walloon regions (Vanneste et al., 2008, p. 175). The inner city of Brussels is traditionally poorer than its suburban commuting areas around the Region. The gentrification in Brussels, the rising property prices and rising rental prices have led to concerns that people may move out of the city in the search for affordable ac-commodation. Households that decide to move outside the city’s boundary become out of sight of this analysis.

(22)

Chapter 4

Case 1: Amsterdam

The purpose of this chapter is to gain insight in the relationship between the Dutch housing regime and its significance to the composition of the private and social rented market in Amsterdam. The case study is subdivided into three sections. The first section describes the features of the Dutch housing policy regime: the aim and scope of housing intervention, the allocation of social rented housing and the type of housing subsidy (4.1). The second section highlights the characteristics of the rented market: the housing stock, the characteristics of the rented stock and the rental prices (4.2). The third section describes the demographic profile of tenants in Amsterdam (4.3). The fourth section describes the housing patterns in Amsterdam (4.4). The chapter closes with a discussion and summary of the results (4.5).

4.1 Characteristics of the case

The Dutch Housing survey provides three different classifications for low-income households: a low low-income, the EU-target group and eligibility for housing allowance. The EU target-group are all households with an income below a fixed threshold. It was decided to be 33.000 but there are debates to increase the income ceiling. Low-income households have a household that falls in the lowest 40 per cent of the national income strata. A stable 52 per cent of households in Amsterdam have a low income. The eligibility for housing allowance is based on a calculation of the income relative to the household size. The amount of people entitled to housing allowance has increased. There were in 2002 119 thousand people who were entitled to housing allowance. In 2012 this were 148 thousand households. This is mostly the consequence of a greater entitlement. Most of these target-group households are single-person households. Colloquially speaking, low income households are often single-person households, single parents, unemployed or underemployed retired, ethnic minorities, etc.

(23)

4.2 | Housing policy regimes 22

4.2 Housing policy regimes

This section aims to describe the characteristics of the Dutch housing pol-icy regime and its implications for Amsterdam. The elements that will be discussed are: first the aim of the housing policy intervention (4.2.1), sec-ond the allocation criteria for social rented housing (4.2.2), thirdly the type of housing subsidies (4.2.3), and the section closes with a discussion of the summary of the findings (4.2.4).

4.2.1 Aim of the Dutch housing policy

The aim and scope of the Dutch housing policy can be organised into four main phases: the pre-war housing policy was strongly rooted in civil soci-ety, the post-war Dutch housing policy can be classified as ideal-typically comprehensive, the first steps away from a comprehensive housing policy were set in the 1970s and since the 1990s there was a desire for a more supplementary role of the Dutch state in housing provision.

The first phase started before the First World War, when the first step-ping stones of a Dutch housing policy were set with the first housing act of 1901 which specified the minimum dwelling quality. The early housing asso-ciations were still organised within the Dutch pillarized civil society (Salet, 1999). The civil society organisations were marked by a religious or political affiliation (Dielemans, 1994).

The initial decades after the Second World War marked the second phase, where the role of the central government increased. The central government became more involved in the price setting, quality standards and building programs in the rented sector (Salet, 1999). The housing policy was tied to an economic agenda of international competitiveness: the rents were kept low as to maintain the wages low (Salet, 1999). By the 1970s and 1980s there was a desire to reduce the government spending on housing subsidies. The individual housing allowances were implemented in the 1970s and the rents were increased every year to bring them closer to the market level in the 1980s (Dielemans, 1994).

A new phase, and a key turning point in the housing policy of the Nether-lands was the 1989 Housing Memorandum by Heerma (Salet, 1999). This memorandum expressed a desire for more market functioning in the housing market. The target group became more clearly as being low-income groups. A projection implied in 1995 a greater degree of financial autonomy for the housing corporations (Salet, 1999). The 1989 white paper redefined the po-sition of housing associations, and by implication, the role of social rented housing (Schutjens et al., 2002, p. 649). The problem of skewness was for-mulated in the white paper. There was a desire to house a larger share of the population in the owner occupied sector, but to maintain a social rented

(24)

sector with affordable homes for low income households (Marrising and van Kempen, 2004).

The desire for more market functioning and less government involvement remained a central theme in the Dutch housing policies that followed. The Mensen Wensen Wonen memorandum of 2000 expressed a desire to further increase home-ownership, particularly in the larger cities. In specific there was also a desire for more high-quality dwellings. The social rented sector still played a role, particularly for those with a low income or few oppor-tunities in society. This vulnerable group was assisted through individual housing allowances and the supply of affordable rental dwellings (Ministerie van VROM, 2000).

4.2.2 Housing allocation in Amsterdam

The target group for the social rented sector has traditionally not been limited to low-income and marginal households, since 2009 it is restricted though. Throughout the 20th century there have been political debates about the legitimate target-group for the social rented sector in the Nether-lands. In 1968, it had been decided that the Dutch social rented sector would be open to both low-income and middle income groups (Dielemans, 1994).

The mortgage deduction and individual housing allowance did make a step into making renting relatively less attractive for the middle income group and more attractive for low-income households. The first step to reformulating the legitimate target group of the social rented sector was set with the Nota Heerma of 1989. The problem of skewness was first mentioned in this policy memorandum. Skewness is a situation where higher or middle income households occupy the social rented market or whereby low-income households are not able to access affordable dwellings. The Heerma policy memorandum analysed the broad nature of the Dutch social rented sector (Dielemans, 1994).

Up until recently, there were no ‘exit’ criteria. If the tenants income rose after accepting the dwelling, the tenant could stay in the dwelling. There were only income tests when the tenant wanted to move to a differ-ent dwelling, the housing corporations could then either reject the request or assign a more expensive dwelling. This policy was ruled to be against the principles of the EU competition policy in 2009. The aid to the Dutch housing corporations were seen as illegal state aid because it also funded commercial activities employed by the housing corporations. In order to preserve the status for housing corporations as ‘service of general economic interest’ the Dutch government decided to define the target group for the so-cial rented sector more explicitly and narrowly to low-income groups and to check whether the housing corporations enacted this new regulation (Gruis and Priemus, 2008).

(25)

Since this EU decision, there were additional rent increases for current tenants with an income above 33.000 euros (Lawson, 2011). The decision of the European Commission of 2009 pushed for a significant change in the criteria for access to the social rented sector. The commission had ruled that ‘the provision of social housing may qualify as services of general economic interest only if the provision is restricted to a target group of disadvantaged citizens or socially less advantaged groups’ (Sol and Vos, 2012, p. 7). Otherwise the support would have been illegal state aid.

The Dutch government decided that at least 90 per cent of all dwellings in the social rented sector should be allocated to households with a maximum income of 33.000 euros (Sol and Vos, 2012). The accommodation model in the Netherlands is an advert-model. The housing corporations publish the vacant housing in a newspaper or on the Internet. The dwelling characteris-tics and the qualifying conditions for the tenants are indicated on the advert. Assignment to a dwelling requires an active responses. The dwellings are allocated to the respondent who meets the qualifying conditions on the basis of sequence on the waiting list (Lawson, 2011).

Some households may apply for priority allocation. The types of prior-ity are: (1) priorprior-ity as a consequence of urban renewal. These households need to move as a consequence of the demolition or renewal of their current dwelling; (2) households who received permanent residence and are allocated a dwelling on a national basis; (3) urgency on social or medical grounds; (4) urgency after leaving an centre for homelessness; (5) urgency on the basis of professional necessity for the region: teachers, policemen and medical work-ers; (6) in case of neighbourhood problems or livability problems; (7) other types of priority (AFWC, 2014).

4.2.3 Scope and type of housing subsidy

The post-war expansion of the social rented stock has been aided by exten-sive bricks and mortar subsidies. During the initial decades after the Second World War the central government gained an increasing role (Salet, 1999). The central government became more involved in the price setting, quality standards and building programs in the rented sector.

The government assisted the mass building programs for the social rented sector through lending conditions at favourable rates in exchange for a stronger involvement in conduct of the housing corporations (Salet, 1999). The Dutch social housing construction programs were more extensive than in most other countries (Schutjens et al., 2002). Most of the new housing construction for the non-profit sector took place in the major cities (Diele-mans, 1994).

By the 1970s and 1980s there was a desire to reduce the housing subsidies. The individual housing allowances were implemented in the 1970s and the

(26)

rents were increased every year to bring them closer to the market level in the 1980s (Dielemans, 1994).

In 1995 there was a grossing operation which implied a greater degree of financial autonomy for the housing corporations (Salet, 1999). Their gov-ernments subsidy obligations stopped and the debts were cancelled. The housing associations gained more freedom in setting their rent, determin-ing their builddetermin-ing programs and were also allowed to build more luxurious homes (Schutjens et al., 2002). The housing corporations became social en-trepreneurs. These activities on the commercial market enabled the housing corporations to make revenue on the sale of private sector dwellings and use the revenues for the construction of the otherwise unprofitable social rented dwellings (Kempen, 1994).

There were still object subsidies. The housing associations were able purchase construction land in exchange for a say of the local government in the development of these new areas. These newly developed areas became mixed in tenure, a part of it was social housing. As a consequence, the hous-ing associations became active in the construction of owner occupied and free-rent dwellings. This way of financing the construction of social rented housing via the sale of owner-occupied housing was pro-cyclical. The new housing construction for rental dwellings came to a low point during around 2002 and 2003. The costs for new construction increased as a consequence of a heated real-estate market (Schuiling and van der Veer).

The new construction output did not meet the intended targets (RIGO, 2011). The housing corporations still play an important role in the construc-tion of new owner occupied dwellings, as their building program consists of a mix of social rented dwellings, luxurious rental dwellings and housing for owner-occupation. The rental prices in the social rented sector is income dependent.

4.2.4 Summary findings

The Dutch central government has since the 1970’s displayed an expres-sive desire to allow more market functioning to the supply of housing. The legitimate target group for access to the social rented sector has been a sub-ject of debate, which has surpassed the national borders and has become a subject of debate in the European commission. Since 2009, the target group has become more strictly defined to households with an income below a certain threshold. Since this policy has become legislation, the step to a targeted housing allocation scheme is official. There are not only strict criteria for entering the social rented sector, but also for monitoring of cur-rent social tenants. While there is a desire for a more supplementary role of the government in housing provision, it cannot yet be said that the more comprehensive aims of the housing policy have been abandoned. There is a

(27)

4.3 | The rental market in Amsterdam 26

greater reliance on subject subsidies, in the form of housing allowance and rent control, and a lesser commitment to construction subsidies.

Aim Target-group Subsidy 1950s and 1960s Economic

com-petitiveness.

Low and modal in-come groups.

Bricks and mortar subsidies.

1970s and 1980s Ibid. Ibid. Start housing al-lowances.

1990s Market function-ing.

Low-income. Cancelling of loans and object-subsidies.

Table 4.1: Aims, target-groups, and subsidies throughout the past decades.

4.3 The rental market in Amsterdam

The aim of this section is to provide an overview of the characteristics of the rental market in Amsterdam. The items that will be elaborated on are: the characteristics of the rental stock (4.3.1); the rent levels (4.3.2); and it closes with a summary of the findings (4.3.3).

4.3.1 Characteristics of the Amsterdam rental stock

The housing market in Amsterdam is dominated by the rent controlled hous-ing stock. Much of the rented stock in Amsterdam is owned by houshous-ing cor-porations. Up until the end of the 1990s, the social rented stock expanded relative to the private rented stock. The share of rented housing decreased in the last years. Three developments are important to understand the development of the size of the rental market in Amsterdam.

First, the rental market is determined by the housing construction pol-icy. The housing corporations play an important role in the construction of social rented housing and also of owner-occupied housing. The local gov-ernment and the housing corporations decide about the development of new neighbourhoods in cooperation with each other. The housing corporations are able to purchase municipal land at a reduced price from the munici-pality, in exchange, the local government receives a say in the outlook of the new area (as discussed in section 4.1.3). One of the concerns of the local government is the tight housing market in Amsterdam. Among other groups, newcomers moving to Amsterdam experience a difficulty finding an adequate housing. The housing agreement of 2001-2002 observed a greater need for a more differentiated housing stock. There is also a greater need for more owner-occupied and higher quality dwellings according to the report Amsterdam (2001).

(28)

Second, there is a policy of urban restructuring and tenure mix in Ams-terdam. A part of the social rented stock has been demolished or renovated and replaced by a mix of owner-occupied, private rented and social rented dwellings. A large part of the new stock is of a better quality and middle and upper range price (Aalbers, 2004). The urban restructuring policy is aimed to attract middle income groups and families to (former) poor neigh-bourhoods by diversifying and upgrading the housing offer. 90 per cent of demolished dwellings are rented houses.

Thirdly, there are tenure conversions: a part of the social rented stock is sold to private persons. This policy never fully took off (Aalbers, 2004). A small number of social rented dwellings have been sold to private landlords and owner-occupiers. 17.000 social rented dwellings in Amsterdam have been sold to private agents in the period between 1998 and 2011.

1990 2002 2012

n % n % n %

All inhabited dwellings 344.790 369.880 384.280

Owner-occupied housing 36.070 10 67.870 18 116.700 30

Rented Housing 308.720 90 302.020 82 267.580 70

Social rented housing 216.670 59 188.570 49

Private rented housing 85.340 23 79.000 21

Table 4.2: The housing stock in Amsterdam by tenure. Source: CBS, BZK/WB - WoonOnderzoek Nederland (WoON). Data on social rented housing and private rented housing for 1990 was not gathered in these sources.

The superintendent is the owner of a rented building, this can be a hous-ing corporation, an individual landlord, the municipality or institutional investors. Social rented housing is by definition owned by housing corpora-tions. The landscape of superintendents has diversified since 2009. There are fewer institutional investors and more private or unknown landlords. This suggests that the financial crisis has influenced the housing market in Amsterdam. Institutional investors decided to sell a part of their stock, and private landlords became more numerous on the housing market. It is unlikely that there were no private landlords up until 2009, but these may have been confined to the grey market. Theoretically, these changes in the private rental market do not have a direct impact on the entitlements for individual tenants, rent control and protection, since this is not dependent on the tenure type. A distinction between private and social rented housing can be found in the access and allocation of the rented stock for tenants. The social rented housing is allocated via formal bureaucratic procedures that includes a means test, waiting lists and priority procedures. Some pri-vate landlords also use waiting lists for the access to their rented stock to prevent discrimination of potential tenants.

(29)

Figure 4.1: The distribution of the housing stock by owner of the building. Source: CBS, BZK/WB - WoonOnderzoek Nederland (WoON).

The social rented stock in Amsterdam is of a high quality. There have been continuous investments in the maintenance and quality of the build-ing stock. Aside regular maintenance, there is also a conversion policy in Amsterdam, whereby a part of the housing stock is sold.

2002 2012

Private rented Social rented Private rented Social Rented

Single family dwelling 5 12 7 9

Multiple family dwelling 95 88 93 91

Table 4.3: Type of dwelling by type of landlord CBS, BZK/WB - WoonOnderzoek Nederland (WoON).

2002 2012

Private rented Social rented Private rented Social rented

Three or less rooms 76 66 75 72

Four rooms 17 27 18 23

Five or more rooms 7 6 7 6

Table 4.4: Dwelling size type of landlord. Source: CBS, BZK/WB - WoonOnder-zoek Nederland (WoON)

4.3.2 Rent levels in Amsterdam

Both private and social rented housing is under the scrutiny of rent control. The national rent legislation lays out a measurement system which links the dwelling quality to the legal maximum rental price of the dwelling.

(30)

Rents in the social rented sector is on average slightly lower than in the private rented sector. The base rent is the rent is the net rent plus utility and service costs. The rising housing prices is - among other reasons - a consequence of a quality upgrading of the rental stock. The rent level is set on the basis of the dwelling quality, such as its size. However, the prices increased across the range, too. The luxurious section of the rental market is not rent-controlled. There is a small difference in price.

2002 2012

Average base rent 355 325 570 450

Average net rent 340 285 545 375

Table 4.5: Average rent. Source: CBS, BZK/WB - WoonOnderzoek Nederland (WoON).

The rental market is organised into four distinct price-quality categories. These price-quality categories are used for calculating a household’s entitle-ment to housing allowance. The criteria for entitleentitle-ment to housing allowance is in both tenure the same. It is based on a means test for the household and the housing characteristics. The means test for the household is based on the household income, wealth and household composition. The dwelling characteristics include the price-quality level of the housing. The maximum rent level for an individual dwelling is based on a point system. The housing receives a score based on the size, amenities and quality. The price-quality segments indicated in this table are based on these scores. These segments are used for calculating the height of the housing allowances that a household may receive. Housing of a high-quality fall in the liberalised rent segment, households cannot apply for housing allowances in this segment. The land-lord and tenant are free to negotiate the rents. The cut-off points for these rent price segments are determined every year by the central government. They increase with inflation and also the general rent increases. The table ’Cut-off points for the price-quality segments of the rental market’ shows price-range that corresponds to each of the segments of the rental market.

2002 2012

Low rent-quality segment 289,59 E 366,37 E Medium low rent segment 427,47 E 524,37 E Medium high rent segment 458,32 E 561,37 E Liberalized rent segment 541,31 E 664,66E

Table 4.6: Cut-off points for the price-quality segments of the rental market. Source: CBS, BZK/WB - WoonOnderzoek Nederland (WoON).

(31)

Figure 4.2: Rental housing in Amsterdam by price-quality segment Source: CBS, BZK/WB - WoonOnderzoek Nederland (WoON) (Own translation).

housing stock in Amsterdam according to its price and quality. The amount of dwellings within each of these four rent-quality categories has changed. Nearly half of the dwellings provided in both the social and private rented sector were low-rent housing. There are fewer dwellings in the low-rent segment and more dwellings in the liberalised rent segment. The change is the most visible in the private rented sector. This pattern indicates that there is an upgrading of the rental market: the dwellings provided are of a higher quality. This leads to higher prices. The quantity of dwellings offered that are within the price-quality range of rent control decrease, and there is an increase of the amount of dwellings offered in the more luxurious segments of the rental markets.

2002 2012

Low rent-quality segment 40.220 94.680 19.790 55.530

Medium low rent segment 25.150 90.100 21.610 85.350

Medium high rent segment 9.880 25.170 14.970 34.00

Liberalised rent segment 10.090 6.730 22.640 13.700

Total rent controlled 285.200 231.250

Total liberalised 16.820 36.340

Table 4.7: Quantity of rental dwellings in Amsterdam by price-quality segment: CBS, BZK/WB - WoonOnderzoek Nederland (WoON).

Accessibility of the rental market

The accessibility measure calculates which share of the rented market is affordable to low-income households. This is calculated by measuring the maximum rent that a household could afford, given that they do not spend more than a certain share of their income on accommodation. A rule of

The (great?) transformation of the rented market in the Brussels capital region and Amsterdam