To mix or not to mix? : that’s the political question : a comparative study to the role of land ownership in facilitating mixed-income housing developments in the different planning systems of London and Amsterdam

To mix or not to mix? That’s the political question

Author: Patrick van Son Student number: 10365230

Programme: Urban Studies (Research Master) Programme group: Urban Planning

Thesis Supervisor: Prof. dr. Tuna Tasan-Kok Second Reader: Dr. Jochem de Vries

Institute: Universiteit van Amsterdam Amsterdam, 22 June 2018

1

Preface

The cover picture is a cut out of the famous painting Victory Boogie Woogie by Dutch painter Piet Mondriaan. A housing programme consists out of a number of basic elements, the housing segments, which resemble the limited amount of colours used by Mondriaan. Some segments typically occur more often in the mix than others, through the influence of national housing policy on shaping these housing segments. Also some colours occur more often than others in the painting. Although the painting seems to be a very simple mix of colours and squares, the painting process was highly complex and Mondriaan constantly altered the composition of the colours. This complexity resembles the development process of mixed-income housing as investigated through this research. Attaining a mix of housing segments in urban developments is certainly desired by policy makers, but the mix can vary to a large extent as the political context influences the way planning instruments and land policy are used to facilitate a mix of housing segments.

This thesis is a result of a one and a half year study to a question I personally wanted to investigate. Why is the housing developed in cities what it is? During my years of living in Amsterdam and keeping track of new housing and area developments in this city, I wondered why some new developments consist of merely market homes, why other developments consist of intermediate rent, and why other developments provide a mix of housing segments. I wanted to find out what regulations and other factors influenced the decisions about what housing is developed in the city. At the moment of writing, I can tell that throughout the past one and a half year, I found the answer to that. Particularly in Amsterdam, all planning instruments are available to influence the mix of housing developments, especially when the land is publicly owned. A political commitment towards affordable housing development is however needed to use these instruments in planning practice. This commitment was absent for most of the past decade, but since 2017 the Municipal policy has changed in favour of affordable housing segments, and thus in favour of mixed-income housing. In London, this trend is similar, but as the research will explain, the nature of the British planning system and lack of public land will make it difficult for the Greater London Authority to initiate change and facilitate more affordable housing as part of the housing mix.

I am pleased with the large amount of freedom given by my study programme and supervisor to shape this thesis research towards my personal interests. I would also like to thank my supervisor and other academic staff of the University of Amsterdam and Bartlett School of Planning for providing useful feedback on my research. Finally, a special word of thanks to people involved with the studied cases and external experts who have been interviewed for this thesis. Without their help and input this thesis would not have been possible.

2

Index

Preface ... 1

1. Introduction: mixed-income housing, a remedy towards urban affordability problems, but a challenge to develop ... 4

2. A theoretical framework on why socio-economically mixed neighbourhoods should be developed and how it can be achieved ... 9

2.1. The importance of socio-economic mixing in urban areas ... 9

2.1.1. Mixing income groups in urban developments ... 9

2.1.2. Mixing in a neoliberal context ... 10

2.1.3. Conceptualising mixed-income housing ... 11

2.2. Strategies to promote housing mix ... 13

2.2.1. Affordable housing provision embedded in planning and housing systems... 13

2.2.2. Facilitating affordable housing within the housing mix of new developments ... 18

3. An institutional framework of housing development in London and Amsterdam ... 20

3.1. Two different planning systems ... 20

3.1.1. London embedded in the British development-led system ... 20

3.1.2. Amsterdam as powerful municipality in the Dutch plan-led system ... 22

3.2. Housing institutions: who decides what about the housing system? ... 24

3.2.1. United Kingdom ... 24

3.2.2. The Netherlands ... 25

3.3. Housing segments: what type of housing can be developed in each city? ... 27

3.3.1. The United Kingdom and London ... 27

3.3.2. The Netherlands & Amsterdam ... 29

4. Research design and methodology ... 31

4.1. Research questions... 31

4.2. Comparative research design ... 32

4.3. Introducing the cases ... 33

4.3.1. Royal Albert Wharf – London ... 33

4.3.2. Cruquius – Amsterdam ... 35

4.4. Research methods ... 37

5. Empirical findings ... 40

5.1. Mixed-income housing embedded in London and Amsterdam ... 40

5.1.1. London ... 40

5.1.2. Amsterdam ... 44

5.2. Investigating the development processes of mixed-income housing ... 47

3

5.2.2. Cruquius – Amsterdam ... 52

5.3. The value of public land policy for mixed-income housing ... 56

5.3.1. Royal Albert Wharf & London ... 56

5.3.2. Cruquius & Amsterdam ... 58

6. Conclusion ... 61 6.1. Main conclusion ... 65 6.2. Recommendations... 67 6.3. Reflection... 69 References ... 71 Appendixes ... 79

Appendix 1: Anonymised Interview List ... 79

Appendix 2: Information letter (English) ... 80

4

1. Introduction: mixed-income housing, a remedy towards urban

affordability problems, but a challenge to develop

Many main cities in Western Europe cope with challenges on their housing markets in terms of accessibility and affordability (Boterman et al., 2013; Dewilde, 2017). Finding a place to live in these cities becomes increasingly difficult for large groups of residents, since there is often not enough supply of suitable housing or the available housing is too expensive. New residential developments can make a contribution to providing access to housing for potential new residents for a city, but also for people already living in the city looking for a home that better suits their housing need. In terms of equity and inclusivity, the provision of new housing should ideally be focused towards a broad range of citizens (Rowlands et al., 2016). Despite many city governments deliberately promote inclusive, socially mixed communities in new urban area developments, it appears to be something difficult to succeed in practice as new urban developments often tend to consist of merely market homes (Raco et al., 2007; Majoor, 2011; Boland et al., 2016). To explore and understand why, this paper aims to investigate the development process of mixed-income residential developments in two cities with a large pressure on the housing market, London and Amsterdam. This study will hereby offer a comparative perspective on these two cities and the differences in the planning systems in both countries, with a specific focus on the role of land ownership.

Land is a crucial factor for the development of new housing, particularly in cities where land is scarce and land values are high. Particularly public land policy is a crucial factor that can influence the outcome of spatial developments in terms of income mix (Montgomery, 1987; Evans, 2004). Land is needed to build new developments, and without land a developer cannot build any developments. The party that owns the land decides how the land will be used and developed, but within the limits of the law and planning system (Buitelaar et al., 2008). Particularly in urban areas, land prices are generally high, which makes it financially challenging to include cheaper housing segments within a development (Evans, 2004). As through land policy, the price of land can be regulated, particularly public land ownership is seen as a crucial factor for influencing the development outcome (Buitelaar et al., 2008; Cheshire, 2009). In Amsterdam, for instance, public land development through public ownership has traditionally been central in planning to achieve social goals, particularly the development of social housing (van der Krabben & Jacobs, 2013; Fainstein, 2010). In London, the land market is however much less controlled and the costs of acquiring land generally make up the highest costs of a development (Adonis, 2015). As the mentioned literature mentions, and as particularly Amsterdam illustrates, public land policy and planning instruments can influentially steer on the outcome of developments. However, the precise relation between the role of land on the one hand, and planning instruments on the other hand remains unclear. To what extent does it matter whether land is publicly or privately owned and does public land ownership really enable the use of stronger planning instruments to steer on the inclusion of affordable segments in a housing mix?

The question this study aims to answer is:

“How does the interaction between land ownership and planning instruments influence mixed-income urban residential developments within different planning systems?”

5 To understand the interaction of the planning instruments available to facilitate mixed-income housing and the public land policies to facilitate mixed-income housing, two ongoing area developments will be investigated in London and Amsterdam. In London, the Royal Albert Wharf will be investigated, a residential area development in the Royal Docks on public land. In Amsterdam, Cruquius will be investigated, a similar development in the Eastern Docklands on predominantly public land, but the Municipality adopted a strategy to apply minimal guidelines to this land. Both cases are typical for the current trends in planning in London and Amsterdam. Where in London, there is an increased focus on using public land to facilitate affordable housing within developments, in Amsterdam and the Netherlands the focus in planning shifts towards a less active and more facilitating public sector (GLA, 2017:b; Dixon, 2009; Janssen-Jansen & Woltjer, 2010). The cases are however atypical in relation to the traditional development models of both cities. This makes the cases useful to study the planning instruments in both different planning systems separately from the role of land (Gerring, 2009; Schofield, 2002). As in London there is normally no control over land prices and the development-led planning system makes it difficult to secure more affordable housing segments in a development scheme, public land might facilitate this better (Martin, 2017; Wainwright, 2014; Janssen-Jansen & Woltjer, 2010). In Amsterdam, the Municipality, through its land ownership, traditionally adopts a very active role in plan making and facilitates social housing within developments. Investigating Cruquius can nicely explain to what extent the Amsterdam planning instruments are effective in facilitating mixed-income housing without the use of public land policy.

The underlying issue crucial to this research is: what is the value of public land ownership towards facilitating mixed-income housing? To what extent is this factor crucial and is it different for two cities with a totally different planning culture? As city governments make housing policies and make social mixing policies, it would be expected that -if they own land- they would use their land to support these policy goals. It is therefore crucial to question the argumentations behind public land policy: is a public landowner bringing in cheap land to facilitate a higher share of more affordable housing or does it play a financially driven role within the general land market (Haila, 2016; Montgomery, 1987)? This would imply an important role for local politics as they decide where the priority is. Wainwright (2014) and Besussi (2016) illustrated that in London boroughs, under austerity-driven neoliberal influences, public land is regularly used to generate money to fund social services and to stimulate gentrification. Also Heurkens & Hobma (2014) and Raco (2018) conclude that these neoliberal influences of deregulation and privatisation, have led to more market-oriented development practices in both the UK as the Netherlands. Planning and land policy would therefore be increasingly focussed on achieving economic goals, and less to the public benefit, such as affordable housing (Boland et al., 2016). Since neoliberal policy is more integrated in London and the UK, it is likely that the public land policy will be less facilitative towards affordable housing segments, than in the more socio-democratic Amsterdam. On the other hand, public land ownership maybe does not have to be a crucial factor in facilitating this housing mix. Planning instruments might provide enough regulations and stimuli for the private sector to include affordable segments within a housing mix (Rowlands et al., 2016). Private developers might follow these planning guidelines since they know they have to comply with housing targets and the fact whether land is publicly or privately owned might be only a small factor.

6

The conceptual structure of the research

As this study is a comparative case study, it is essential to look at the differences between different levels of organisation of housing and land development in different institutional contexts (Squires & Heurkens, 2016). The study will hereby largely adapt the theory on mixed-income housing provision by De Kam, Needham & Buitelaar (2014). On a more abstract level, this theory can be conceptualised by the three-level framework by Healey & Barrett (1990) on how planning decisions -that shape the planning outcome of developments- are influenced by three levels of mechanisms.

At the highest level, there are the cultural norms and values that people have about what should be developed and how this should be achieved. This level determines the way for instance state and market relations are defined (the planning culture), the welfare system and housing system. The Netherlands has a Napoleonic law and a relatively social-democratic welfare system, while the UK has a common law and is a third way country with characteristics of liberal welfare states. These kind of factors will not be directly investigated for both cases, but it is discussed as an important context that is influential in shaping the second level, politico-juridical rules. The planning processes are guided by rules that determine what actions different actors are allowed to take, and which not, to achieve a desired planning outcome, such as mixed-income housing development. The planning system (and its instruments) is therefore placed on this middle level. The politico-juridical rules determine how resources, the lowest scale, may be used in the planning process. The resources are described by Healey & Barrett (1990) as the paths through capital flows that are derived from the economy by different actors. These resources can be used in the planning and development process and hence have a large role in shaping the outcome. Land ownership is placed as the most important resource that will be investigated for the two cases.

The use of planning instruments derived from the planning system and the land policy through land ownership will shape the planning process for both cases and how mixed-income housing is generated. As the theoretical framework will illustrate, for each development this will likely been done by a combination of shaping, regulating an stimulating measures. Mixed-income urban residential developments will be investigated by covering the three different aspects mentioned in the theoretical framework: 1)the mix in the housing programme, 2)the finance of the affordable share of the mix, and 3)the duration of the housing mix.

Conceptual model

-Desire for affordable housing Norms & values (macro)

-Political context –Housing system -Welfare system

Politico-juridical rules (meso)

-Planning system Mixed-income residential

developments 1: Programme 2: Finance 3: Duration Shaping Resources

(micro) -Land ownership

Regulating

7

Relevance of the study

The relationship between urban transformation and land ownership is not often investigated, particularly in the Netherlands (Buitelaar et al., 2008). Particularly the separate influences of planning instruments and land policy on development outcome are unclear and dependent on context (Evans, 2004). This study therefore aims to generate some valuable knowledge on the interplay of planning instruments through the different planning systems and land policies through land ownership. This study thus particularly adds the role of the variable land ownership to the academic debate on planning systems and how they evolve (Munoz-Gielen & Tasan-Kok, 2010; Janssen-Jansen & Woltjer, 2010). The research hereby aims to generate insight in considerations made by public landowners in London and Amsterdam. As Haila (2016) suggests, these considerations are key to generating conditions for the inclusion of affordable housing segments in a development. With the UK as a state where more neoliberal influences are present than in the more social-democratic Netherlands, the expectation is that public land would more easily serve economic goals, while in Amsterdam it would more easily support social goals, such as mixed-income housing. Since there are also indicators that under neoliberal influences, the Dutch planning system is evolving towards a more development-led system, the comparison with the UK might be particularly useful to transfer knowledge from the British practice to the Netherlands (Heurkens & Hobma, 2014). Then especially the Royal Albert Wharf developed under a development-led planning system, but on public land, might be useful for the Dutch context as public land ownership is common in Dutch municipalities. Is in such a more market-led development public steering through land still required or can the market create a housing mix on its own?

Additionally, the study is relevant as it connects planning theory to the contemporary challenges on urban housing markets in terms of inclusivity and accessibility. Such kind of topics as inclusivity and residential mix are often studied in the context of urban regeneration and gentrification, particularly in Amsterdam (Hochstenbach, 2017; van Gent, 2013; Uitermark, 2009; Musterd, 2014). These studies then often focus on the sale of social housing as neoliberal policy to mix less affluent neighbourhoods with more affluent residents and study the effects on the existing community and amenities (Arthurson, 2002; Lees, 2008). This study adds a “development perspective” within this field of housing and socio-economic mixing and specifically focusses on the development processes of new urban neighbourhoods and how a mix of housing segments can be developed. As Boland et al. (2016) and Raco et al. (2007) illustrated, mixed-income housing is not always realised. How do cities plan for new socio-economically mixed neighbourhoods and what factors influence this?

Finally, the comparative research aims to provide some practical knowledge on the opportunities and constraints of providing a diversity of housing segments in cities. The development of affordable housing segments is hot topic in both London and Amsterdam since the demand for it is high (Hulsman, 2018; Damen, 2016; Martin, 2016; Boterman et al., 2013). Despite the fact that city governments often have deliberate goals to add a mix of housing segments to the urban housing stock, these goals are not always realised (Colenutt et al., 2015). This study explores the housing policies with regard to mixing in both cities and links them to the development processes of the studied cases. This might indicate why certain housing segments are more frequently included in the housing mix than others. Through the case selection, the study aims to explain to what extent public land ownership can play a role to influence the housing mix in these developments. Is public land policy the key to facilitate more affordable housing or are planning instruments (or other factors) more important to facilitate mixed-income housing? By studying the two cities in both countries that have the largest problems on the housing markets in terms of accessibility and affordability, the comparative knowledge might be useful as well for cities with a lesser amount of challenges on these issues.

8

Structure of the thesis

The thesis starts with the twofold theoretical framework. This chapter will first make clear what mixed-income housing is and what the benefits are of mixing different housing segments in developments. The second part of the theoretical framework explains how mix-income housing can be facilitated in new developments through the housing system, planning system and land policy. Next, as the research compares two different cities in different countries, the thesis spells out the most important contextual differences between planning and housing developments in the UK and the Netherlands. Chapter 4 discusses the research questions the thesis aims to answer and by what methodological strategy. This chapter will also address the comparative strategy of this research and the case selection. Chapter 5 will discuss the research findings on the three sub-questions. Chapter 6 offers a comparative conclusion on the three sub-questions, while this comparison leads towards the answering of the main question of this research.

9

2. A theoretical framework on why socio-economically mixed

neighbourhoods should be developed and how it can be achieved

2.1. The importance of socio-economic mixing in urban areas

2.1.1. Mixing income groups in urban developments

In urban context, the idea of mixing reflects the ideal that a neighbourhood or new development should be accessible to live and visit by all sorts of residents. The composition of different housing segments is a powerful tool to enhance the social mix of a city, neighbourhood or area as a housing segment determines the income class who can live in it (Rowlands et al., 2006; van Gent, 2013; Hochstenbach, 2017). As social mixing comes with many, expected benefits, it is often a deliberate political goal and implemented by policy makers and urban planners as a way to promote diverse and sustainable communities (Arhurson, 2010; Musterd & Anderson, 2005; Lees, 2008). Mixing in spatial policy is mostly pursued to produce ethnically or socio-economically diverse areas (Arthurson, 2010). As this research is focussed on the housing aspect of urban developments, there is a focus on income-based mixing (socio-economical dimension) as a mix of different incomes is relatively directly attainable by providing or developing different housing segments (van Gent, 2013).

Mixed-income housing1 _{is associated with a number of benefits. First of all, there is a place-oriented} goal that emphasises a better quality of a neighbourhood by mixing different segments of housing (Levy et al., 2013). A diverse population of residents would lead to more diverse facilities and improved safety than there would be in particularly housing estates. This place-oriented goal is thus largely associated with preventing the existence of economically poor and segregated areas. The benefit is often used in urban regeneration context and mixing is then often used to make poorer neighbourhoods more diverse by the addition of more expensive housing segments (Lees, 2008; van Gent, 2013). Critics however argue, that it are most often poorer areas being mixed with more expensive, middle-class homes, but not vice-versa (Lipman, 2008; Hochstenbach, 2017). As this study investigates the development processes of brownfields, there is no existing neighbourhood on the developed site itself and thus place quality by mixed-income housing can be produced from the scratch (Dixon, 2009).

Creating a mixed supply of housing segments then also results in a second benefit, the improved accessibility for many types of residents to the local housing market. Developing a mix of housing segments can simply offer a suitable home for different income groups as someone dependent on social benefits could inhabit a social home, an architect might inhabit an intermediate rental home and a surgeon might buy a home. These different socio-economic classes then get the opportunity to live within the same neighbourhood. In new housing developments, the inclusion of a mix of segments is in no means a certainty, but it can be deliberately facilitated in the planning process (Boland et al., 2016; Tasan-Kok & Sungu-Eryilmaz, 2010).

A third benefit of mixed-income housing is linked to social cohesion and social mobility for low income residents. Many urban policies strive that the encounters between different income groups might

1 _{Other key terms used in literature for a housing scheme that consists of all housing segments are mixed tenure and}

inclusionary housing. The terms are relatively interchangeable, but this research sticks to mixed-income housing. Mixed

tenure is very UK specific and mostly only focusses on the segment and not on factors as size, while inclusionary housing is mostly US specific. Mixed-income housing is chosen as most neutral, straightforward term.

10 result in towards stronger ties between residents from different social-economic classes and that low income groups can lean on the social capital of more affluent income groups (Levy et al., 2013; Musterd & Anderson, 2015; McIntyre & McKee, 2012). Among researchers there is a debate on the positive external effects of mixing and at what scale housing segments should be mixed to gain the biggest positive effects (Arthurson, 2010). Research has however indicated that these effects are difficult to be measured, purely as a result of mixed-income housing (Levy et al., 2013). In some cases home-owners perceived their social renting neighbours as bad neighbours, and in general renters move quicker from an area than home-owners (Levy et al., 2013; Tersteeg & Pinkster, 2016). Such kind of factors make it difficult to create social engagement between different groups of residents. Gans (1962) also stated that within mixed areas, a certain degree of tenure homogeneity is required as people tend to identify themselves more easily with residents with a similar socio-economic status. However, with additional functions and actions, interaction between different groups of residents can be facilitated (Levy et al., 2013; Rowlands et al., 2016).

For this last benefit, mixed-income housing is mostly a means to achieve higher societal goals (Fainstein, 2000). For the first two benefits, mixed-income housing is a goal on itself to directly achieve a better spatial quality and, importantly, housing access to an area for a diverse group of citizens. As this study is focussed on the development process of these inclusive urban residential areas, it focusses on mixed-income housing as a goal on itself. The additional benefits for social cohesion and upward mobility are outside the scope of this research as they can only be measured after developments have been entirely completed for some time.

2.1.2. Mixing in a neoliberal context

Since the 1990s, cities over the world started to revive after a period of urban decline in the 1970s and 1980s (Scott, 2011). As cities became more popular to both live and work, the pressures on many of their local housing markets increased significantly, including Amsterdam and London (Boterman et al., 2013; Hochstenbach, 2017). Hence, housing in many major European cities became slowly more expensive and living in many parts of these cities remains a right for the most affluent groups (Hochestenbach, 2017). Regularly, these trends are also facilitated by (national) governmental policy, focussed on attracting more skilled workers and the sale of social housing, often in favour of urban regeneration (van Gent, 2013; Uitermark, 2009; Lees et al., 2008; van Kempen & Priemus, 1999). Particularly the inclusion of cheaper housing segments in the housing mix tends to be at pressure in new (large) urban developments, as governments take a more entrepreneurial approach to urban developments, desiring internationally appealing and competitive developments to promote the city for high-skilled workers (Swyngedouw et al., 2002; Boland et al., 2016; Raco & Henderson, 2009; Hochstenbach, 2017). According to Swyngedouw et al. (2002), many of these urban developments are guided by neoliberal policy based on values as deregulation and increased (international) competitiveness. These neoliberal forces have their influence on planning as they comprise a shift towards market-led planning and less regulation by public planning authorities (Tasan-Kok, 2011; Boland et al., 2016). According to some, planning is hence being downgraded to simply adjusting plans to meet demands of (market) actors.

The more entrepreneurial approach towards urban policy and urban regeneration started in the 1980s in the United Kingdom under the Thatcher administration, and spread to other European countries later this decade. This urban entrepreneurialism represents a neoliberal philosophy characterised by deregulation, liberalisation and privatisation, as a step away from the Keynesian socio-economic

11 tradition of the post-war period in Western Europe that led to the development of large stocks of cheap social housing in European cities (Tallon, 2013; Harvey, 1989; Scanlon et al., 2015; Carmona et al., 2003). Since the 1980s, there was a larger emphasis on the private sector in shaping urban policy and area development. Particularly in this phase, there was a relative “roll-back” of the state in urban policy making. Neo-liberalisation however does not imply a full retreat from the state in policy, but it implies a different role of the state to impose and facilitate market mechanisms in social spheres: the so-called roll-out neo-liberalisation (Pinson & Morel Journel, 2016; Peck & Tickel, 2002).

Since the late 1990s, and continuing till present-day policy, there has been a renewed focus on preventing social exclusion and involving local communities in urban policy (Tallon, 2013). Pinson & Morel Journel (2016) argue that especially since the 1990s the more traditional social-democratic policies have been intertwined with neoliberal thinking. According to Tasan-Kok (2011), present-day planning is in-between two counter currents. On the one hand, planning needs to be responsive to growing critics on the trends that have come with neoliberal planning practice (more privatisation, market-led development and regeneration), while on the other hand, planners need to work within this new system to do what they can to prevent possible negative impacts on cities and communities (Tasan-Kok, 2011). It is clear that this shift to market-oriented development entails a new role for planners to cooperate with private developers in order to achieve planning goals such mixing housing segments to foster inclusivity (Heurkens & Hobma, 2014). As a result of the retreating role of public parties in leading developments, developers have an increasing role in safeguarding these values. Moreover the inclusion of cheaper housing segments has benefits for the general plan quality, what would be appealing to all parties (Majoor, 2011; Boland et al., 2016). Traditionally, it is often understood that only the public partners of a development protect public interests as inclusivity, while private partners are guided by financial, short term goals. The reality is however less twofold, and both developers and planners act as market actors in a complex negotiating process on what will be developed on a location (Reynaers, 2014; Rainey & Bozeman, 2000; Adams & Tiessdell, 2011).

2.1.3. Conceptualising mixed-income housing

When mixed-income housing is developed, there should ideally be a balance between similar housing segments and differentiation of housing segments in a development scheme, but there is no standard or perfect rule (Gans, 1962; Alexander et al., 1977). Vale & Shamsuddin (2017) investigated 260 mixed-income housing developments in the United States and concluded that the term includes a broad group of different housing compositions. The concept is therefore differently operationalised by various authors. Alexander et al. (1977) mention that a housing mix in an area should be mixed in a sense that it can accommodate a similar population as the total population of a city. Holin et al. (2013) found that at least 20% of the residents in an area should earn below 30% of the area median income and 20% should earn more than 50% of the area median income. Baron (2009) uses a more straightforward mix of one-third public housing, one-third intermediate housing, and one-third market homes. Khadduri & Martin (1997) use a normative perspective, viewing mixed-income housing as a means to a particular social end, and thus focus particularly on including low income families with children in an area. Finally, Brophy & Smith (1997, 5) use a more general definition that describes mixed-income housing as a “deliberate effort to construct multifamily development that has the mixing of income groups as a fundamental part of its financial and operational plans”.

Based on Vale & Shamsuddin (2017) who studied 260 mixed-income housing developments, four different dimensions of mixed-income housing can be studied: 1)the housing programme, 2)the

12 finance of the mix, 3)the duration of the mix and 4)how interaction between the different income groups is promoted. The first three dimensions will be studied for this research. The fourth dimension is left out in this study as the research focusses on the development process of mixed-income housing. Moreover, only the first residents of the studied cases have moved in and it is therefore difficult to study the interactions between different income groups of residents.

First, and the most conventional way of investigating mixed income housing, is the distribution of the different types of residential units by tenure: the programme. Investigating the programme of the developed homes, provides a global image of the resident mix in the area. Decisions made about the different segments and their size has a result on the types of residents that have access to the area. For instance, a large development of studio homes will likely attract single person households, while social rental housing will house a relative large share of ethnic minorities. When investigating the proposed programme, it is also important to reflect on the extent to which housing is provided for the most vulnerable groups on the local housing markets (Arthurson, 2002). These vulnerable groups are not necessarily the least affluent groups. In Amsterdam, it are predominantly starters on the housing market and mid-income groups that find it hard to find a home (Boterman et al., 2013; Hochstenbach, 2017). The housing programme should thus ideally be to some extent demand-driven.

Secondly, it is important to investigate the financial considerations made by both public and private partners as they eventually result in the financial viability of the development programme. Including more affordable homes will normally result in less profit for developers and local authorities as well (Colenutt et al., 2015). It is therefore interesting to investigate how the affordable homes are financed through the development process. In general affordable housing can be financed through a number of ways. Firstly, the government can provide direct supply subsidies to affordable housing. This was a very common way in Europe till the 1990s (Burgess et al., 2007). Secondly, a local government could give discount on the land for the development of affordable housing. This approach is typically linked to many European countries, but particularly the Netherlands (Buitelaar et al., 2011; Cheshire, 2009). Thirdly, the cheaper housing segments could be financed from the profits that the market homes generate (Morrison & Burgess, 2013). Finally, subsidies could be given to tenants of housing themselves, making certain types of market homes more affordable and accessible for them (Vale & Shamsuddin, 2017; Schwartz & Tajbakhsh, 1997). In practice, the funding will be typically a combination of the mentioned ways.

Third, the duration of the mix is an important aspect of maintaining areas sustainably inclusive on a long-term basis. Housing tenure can change and particularly owner-occupied housing is subject to quick price changes (Vale & Shamsuddin, 2017). For intermediate private rental housing, the rents might increase as well over the time and the rent increase could thus displace residents with a middle income (Damen & Kieft, 2016). A mix of housing only upon completion of the development is not serving the long-term housing need, and is therefore important to investigate what kind of agreements local authorities made with developers to maintain the housing mix in an area over a longer time period (Rowlands et al., 2006).

13

2.2. Strategies to promote housing mix

2.2.1. Affordable housing provision embedded in planning and housing systems

In planning practice, achieving a housing mix can be quite challenging. It is particularly the provision of affordable and/or social housing that can be problematic, as affordable housing is simply less profitable than housing that can be sold or rented at market prices (de Kam, Needham & Buitelaar, 2014; Boland et al., 2016; Colenutt et al., 2015). In order to promote affordable housing within residential developments, in most countries and cities there is public policy that focusses on the regulation and promotion of affordable types of housing.

Particularly public parties traditionally take the leading role in facilitating affordable housing, and thus mixed-income housing developments, particularly as a housing mix is seen by policy makers as an effective means for achieving long-term broader societal goals which are linked to the public benefit for an area or the city (Arthurson, 2010; Musterd & Anderson, 2005; Musterd; 2002; Lees, 2008). For private partners -developers and to a lesser extent investors- these long-term goals are generally less important, as these parties have more economic and short-term goals (Kort & Klijn, 2011; Kenniscentrum voor PPS, 2006). The public policy that regulates affordable housing provision differs over different countries, and some countries have more effective legal possibilities to promote the inclusion of affordable housing within developments.

De Kam, Needham & Buitelaar (2014) created a framework to analyse the relation of affordable housing provision with public policy. The authors make a distinction between a country its housing system on the one hand, and the planning system on the other hand. The housing system hereby determines the wish for affordable housing in a country, while the planning system influences the possibilities for affordable housing provision. The differences of both the housing and planning systems between countries can be embedded in the society and welfare regimes of the different countries (figure 1). The role of land ownership is added through this study (see below).

Figure1: Basic relations between housing and planning systems and mixed-income housing provision

14

Housing systems: a shift towards home-ownership

De Kam, Needham & Buitelaar (2014) use Kemeny’s (2001) classification of housing regimes. Kemeny (2001) distinguishes dualist and unitary housing systems, which he bases on Esping-Anderson’s classification of welfare states. Dualist and unitary housing systems both have a different view on the role of social and affordable housing in society.

The dualist housing system can be mostly found in liberal welfare states such as the United Kingdom and the United States and Mediterranean countries. In the dualist housing system, the state takes a direct responsibility for the provision of social and affordable housing. The social housing sector is however seen separately from the regular housing market and the provision of social housing is seen as an activity for the most vulnerable people in society. The unitary housing system is found in more social-democratic and corporatist welfare states such as the Netherlands and Sweden. In these countries, the social housing sector is more integrated with the regular housing market and hybrid semi-public organisations (e.g. housing associations) exist to provide social and affordable housing. These organisations are often supported by the government and they do not only provide housing for the lowest incomes, but sometimes also for middle incomes (De Kam, Needham & Buitelaar, 2014; Lennartz, 2011). The type of housing system thus influences what type of social or affordable housing is typical for a country, as some types of housing are more institutionalised and/or supported by government policy. By this way, the housing system sets the definitions for the different types of housing and for which kind of citizens are eligible for living in them (De Kam, Needham & Buitelaar, 2014).

Housing systems are not static and evolve over time under political reforms. In many European countries, including the Netherlands and the UK, the national governments have started stimulating home-ownership of citizens since the 1980s (UK) and 1990s (Netherlands) (Ronald & Elsinga, 2012; Aedes, 2013; Scanlon et al., 2013). Because of the increase in welfare since the end of the Second World War, an increasing number of citizens were able to afford their own home (Musterd, 2014). The increase of home-ownership also fits within the neoliberal shift in many European countries in the 1980s. National governments saw home-ownership as way of providing individual welfare for its citizens as an un-mortgaged home can substitute a pension income (Castles, 1998). At the same time, the stimulation of home-ownership also meant a retrenchment of the welfare state and public expenditures. Kemeny (2005) and Castles (1998) call this the “really big trade-off” between home-ownership and the welfare state. This paradigm shift towards a “home-home-ownership society” depends on the increase of house values, as housing has been commodified as welfare asset (Ronald & Elsinga, 2012). It is outside the scope of this study to discuss the drivers of house price increases, but since the 1990s house prices in both the UK and the Netherlands started to increase more quickly than inflation. Changes in the UK and Dutch society and welfare regime have clearly influenced the way the housing system functions in each of these countries.

Planning system: plan-led versus development-led

The planning system determines the possibilities for including affordable housing within development plans. The planning system determines what legal abilities public and private partners have to initiate, influence and steer on developments, including the housing mix (Munoz-Gielen & Tasan-Kok, 2010; Kang & Korthals Altes, 2015). The planning system is therefore an important means of safeguarding the inclusion of affordable housing within spatial developments (Carmona et al., 2013). Through the planning system, governments can intervene in property rights by means of e.g. compulsory purchase or the creation of land-use plans to ensure landowners provide an amount of affordable housing as

15 well (De Kam, Needham & Buitelaar, 2014; Webster, 2005). Especially when planning systems allow detailed prescriptions about the required tenure split for developments, the planning system can prevent the development of purely market housing.

Based on the work of Faludi (1987), academics generally differentiate two groups of planning systems. On the one hand, there is the plan-led (or conforming, or Continental) system, found in most European countries, and on the other hand there is the development-led (or performing, or British) system, found in the United Kingdom and the Anglosphere (Kang & Korthals Altes, 2015). Plan-led systems are characterised by legally binding land-use plans or zoning ordinances that determine the preconditions on which developers step in when they want to propose a new development. The central idea behind this is that this legal certainty prevents municipalities and planners against making arbitrary decisions in favour, or against, specific developers/parties. The legally binding guidelines, based on the rule of law are typically linked to the civil (Napoleonic) law in many continental European countries (Kang & Korthals Altes, 2015). In contrast, in development-led systems land-use plans are not legally binding, but they are more indicative. They indicate a general way of thinking about the future by the public authorities (Munoz-Gielen & Tasan-Kok, 2010). Based on these general plans, developers can initiate their plans for development. The way the initiatives fit within the land-use plans, is largely dependent on negotiations between local authorities and developers. After the negotiations, just before the development phase, the plans become legally binding. This negotiated aspect of the development-led system is called “development control” (Kang & Korthals Altes, 2015). Development control is often seen as problematic by academics as it means that there is generally less certainty about the outcome of developments and there are less possibilities to regulate a development by public parties (Kang & Korthals Altes, 2015). The development-led system is typically linked to the British common-law, which draws rules from preceding cases.

In general it can be said that, particularly due its legally binding and detailed land-use plans that preserve the existing land-use of sites, plan-led systems are considered to offer better legal instruments for public bodies to steer on development outcomes (Munoz-Gielen & Tasan-Kok, 2010). For mixed-income housing, these planning systems with legally binding land-use plans are seen to be better in facilitating the inclusion of cheaper housing segments, as development proposals consisting of only market homes could be more easily and legally refused. Contrastingly, the development-led system is more focused on a public body facilitating new developments. Together with more discretion and room for negotiations, it would be more easy to make exceptions on affordable housing targets and allow more market housing (De Kam, Needham & Buitelaar, 2014).

Planning systems are however country specific and a strict generalisation is difficult to make. In the UK for instance, there is a very widely established value recapturing mechanism to let private parties contribute to finance facilities for the public benefit such as affordable housing, but on the other hand there is (as expected) less regulation through land-use plans as they are not legally binding (De Kam, Needham & Buitelaar, 2014; Janssen-Jansen & Woltjer, 2010; Webster, 2005). In addition many plan-led countries, including the Netherlands, have characteristics of the development-plan-led system as well. Land-use plans are legally binding instruments in the Netherlands, but they are regularly made through a process of negotiations (Munoz-Gielen & Tasan-Kok, 2010; Buitelaar et al., 2011).

As De Kam, Needham & Buitelaar (2014) argue, the extent to which governments find intervening in property rights desirable -and how the available planning instruments are used- differs quite a lot between countries. Like the housing system, the use of the planning system is dependent on the type

16 of welfare system and political context of a country and/or city. In general, countries with liberal welfare states find intervening with property rights less desirable, while countries with a social democratic or corporatist welfare regime see it as a good means to capture land value and redistribute wealth (Calavita & Mallach, 2010).

The inclusion of land ownership

One aspect that is not directly included in the original model by De Kam, Needham & Buitelaar (2014) is the role of land ownership. Land ownership is an important factor which is related to the planning system since the planning system determines the legal possibilities for landowners versus the government. The specific situation of who owns the land is also important for steering on the outcome (and thus affordable housing provision). Land ownership is a crucial factor for affordable housing development, as the landowner (or the person holding the rights to the land) has the right to bring forward land for development (Needham, 2006). If a landowner is not willing to keep the land for development, he or she can also decide to sell it to other parties. If a developer sees a high potential for the location and the price for the land that is paid is high, it becomes more challenging to develop affordable housing as this type of housing generates less profits than market housing, a profit that is needed to cover the land acquisition costs. This is particularly the case in cities, as the land available for housing developments is generally scarce (Cheshire, 2009; Evans, 2004).

Land is crucial for any human activity, such as living, working and leisure to happen. It is however a complex commodity that has multiple functions. In particular, a tension exists between the producers and consumers of land (Montgomery, 1987). The developer of the land aims to achieve a high as possible return by producing new housing or other facilities on land, while the consumers -often residents- benefit most from how the land is finally used (and using the land and facilities for a decent price). To structure this complex interaction between producers and consumers of land, many countries use a form of land-use planning to regulate these economic benefits versus social, public, needs (Montgomery, 1987). On the one hand, the government can use its planning system to regulate spatial developments, but on the other hand, public land ownership can be a powerful tool to facilitate housing mix, such as the inclusion of affordable housing within developments (Terhorst et al., 2003). This is especially beneficial when land is scarce and high in value (Montgomery, 1987). When the (local) government owns the land, it generally has additional powers to steer on developments either by active land-use planning or giving discount on the land price for affordable/social housing developments (Buitelaar et al., 2011; Montgomery, 1987; Martin, 2017). The public landowner also has the choice to procure the land and let developers compete on certain values, such as housing programme, sustainability, but also land price (van Gelder, 2016).

A crucial aspect of public land ownership is the goal that ratifies the ownership of land by the (local) government (Haila, 2016). Land ownership itself is rarely a goal by itself but it is often a means to serve a greater good. Montgomery (1987) distinguishes four functions of public land ownership, which can function both separately or together (figure 2). First, the land could serve as a means of production. The government can make the land available for development, something which is common in the Netherlands. Second, apart from just delivering land, the government can make the land and the to be developed buildings available for consumption. In this case the government makes land available under the conditions that certain facilities or a share of affordable housing will be provided. Third, the goal can be circulation, serving a wider community than just the direct residents or workers on the site. In this case the government would make the land available for e.g. public infrastructure or

17 conservation of valuable buildings. Finally, the land can serve as investment for the state, and like regular property developers, the public body owning the land could use the land to make financial profits by using land in anticipation of future increases in land value.

When the government is a landowner, it automatically puts the government in a difficult position as it has an economic interest in the land (as market player) while it also has the responsibility to safeguard the public interest by e.g. providing affordable housing (as market regulator) (Montgomery, 1987; Buitelaar, 2010). Buitelaar (2010) refers to this governmental dilemma as the “double hat problem”. As public land can thus also serve as a simple means to fill budgetary gaps, it is important to understand the intentions and reasons that lie behind the land policy of a governmental institution (Besussi, 2016; Martin, 2017).

Figure 2: The functions of public land ownership

Source: Montgomery (1987).

Society and welfare regimes: the importance of political context

As mentioned, the wish and possibilities for a mixed-income housing programme, but also the use of public land, are strongly influenced by the society and in particular the political context. Although there could be housing policy and planning instruments to facilitate housing mix, the reality is that planning outcomes differ in terms of mix and how inclusive the housing mix is (Colenutt et al., 2015). Aaron Wildavsky (1973) analysed this discrepancy between planning expectations and reality and noted that in many cases expectations do not meet up with the outcome. According to Wildavsky, the planning process stands between its actors and society, as the society influences how actors perceive problems and what they want to do about it. In other words, if a certain housing segment is not desired in society, nobody will make effort to include it in the housing mix, even though planners might personally want it and planning instruments could facilitate it.

According to Wildavsky (1973, 132), planning decisions thus have a strong political component: “planning is politics”. According to Wildavsky, planning is an instrument used within the political arena to achieve political goals (Wildavsky, 1973). As local politicians want for instance more owner-occupied housing (e.g. guided by national housing policies), then the planner’s efforts to develop more social/affordable homes can be nullified. According to Wildavsky, in planning practice there are different wills competing between different political groups, planners and private parties, all having different intentions. What is developed is dependent on the dominant plans in mind. The question therefore is, who’s intentions are realised? Wildavsky’s answer to this is that in practice, the most

18 dominant intentions arise from the political arena: governmental bodies are after all responsible for a large part of the funding of developments. The planning process is in this view mostly there to control and mitigate the political choices into a executable plan. Wildavsky (1973) notes that the political and societal context is under constant change. A new government, might change policy focus (on e.g. other housing segments) or the housing and land markets fluctuate. These changing factors indicate how intentions might change and the policies to implement them. Planners and plans must simply adapt to these changing circumstances.

In response to Wildavsky, Alexander (1981) argued that the planning process is not only dictated by political factors, but that the planning process is much more influenced by other factors such as the presence of community groups and the location. Local community groups can intervene in the planning process and can for example promote more social housing in a plan. The location might be on a brownfield, where a local community lacks, or the land values on an inner-city area might make it less viable to develop affordable housing segments. The planning process is the result of a social interactive process between different stakeholders. According to Alexander, the political arena might control the goal of a development, but the planner’s role is to become actively involved in the planning process and negotiate with other stakeholders to come up with the optimal strategy within the constraints of the societal and political context. The planning process is a process after all, and is often reflective. A planning process involves many options and choices made will be evaluated (Alexander, 1981; Faludi, 2000). This process thus also influences how planning instruments will be used to facilitate mixed-income housing.

Resuming, Wildavsy (1973) gives a much greater role to the political, societal context in determining the housing mix of developments. As the contemporary societal context is guided by neoliberal influences, these influences should have an important influence on the mix and planning processes of both cases. On the other hand, Alexander (1981) would put a much greater emphasis on the planning process and the actors in determining the housing mix in both cases. More recently, Raco (2018) illustrates that political pressure is still shaping local planning processes and outcomes. He hereby especially gives importance to the national government in reshaping planning powers of local governments.

2.2.2. Facilitating affordable housing within the housing mix of new developments

To plan mixed-income housing, there are three main ways through which local authorities can shape market conditions for developments, and promote a more inclusive housing mix in an area (Schwartz & Tajbakhsh, 1997; Adams & Tiesdell, 2010 & De Bruijn & Dicke, 2006). In practice these ways work more or less together to result in a mutual agreement by public and private parties.

First, the national and local governments can enforce the development of mixed-income housing by simply enforcing regulations (Adams & Tiesdell, 2010). Without conforming these regulations, developers can simply not get planning permission for their proposed development (Schwartz & Tajbakhsh, 1997) These regulations are commonly implemented in the form of zoning ordinances or they can be prescribed in land-use plans. In Amsterdam, the municipality regularly enforces a specific mix through procurements, which is possible in case of public land ownership (van Gelder, 2016). 30 percent social rent, 40 percent middle priced homes and 30 percent market homes was for instance a regularly used formula (Kieft, 2017). In the United Kingdom, Section-106 forms a legal ground for local governments to force developers to include certain aspects in plans, including affordable housing (Kang & Korthals Altes, 2015; Webster, 2005). De Bruijn & Dicke (2006) refer to regulations as

19 hierarchy. By implementing certain standards or regulations, local authorities and planners do not only enforce private developers to conform these standards, but they also conform themselves. Hierarchy, or regulation, is thus a relatively “strict” way of safeguarding a specific housing mix, and therefore local authorities might sometimes choose for other ways of promoting the public interest (De Bruijn & Dicke, 2006).

Secondly, national and local governments can stimulate the development of mixed-income housing within an area (Schwartz & Tajbakhsh, 1997; Adams & Tiesdell, 2010). Stimulating is a more encouraging approach than regulation, and has a more positive connotation. Stimulating can be done by providing subsidies to private developers when they develop a specific amount of affordable housing or by specific grant programmes. For instance, in New York the local authority provided tax-exempt for projects that reserve at least 20 percent of their homes for low- and moderate income households (Schwartz & Tajbakhsh, 1997). A local authority could also give a price discount on land it owns for the development of certain types of housing, such as social housing. In sum, stimulating measures are meant to lubricate private parties to provide less profitable goods for the public benefit such as social housing (Adams & Tiesdell, 2010).

Finally, private developers can provide a mixed housing development by themselves (Schwartz & Tajbakhsh, 1997). Not all developers aim to maximise their profits by developing purely high-end homes and developing a mix of housing segments might above all spread out sales risks (Rowlands et al., 2016). Therefore they are likely to respond to the demands on the local housing markets by providing a certain mix of housing. Moreover, especially for brownfield developments, a developer would aim to create a well-designed and integrated urban area to achieve place quality. To facilitate the market to develop mixed-income housing, planning authorities should make sure that the market conditions are fitting the desired developments for the areas (Adams & Tiesdell, 2010). Adams & Tiesdell (2010) call this method shaping. Shaping the market conditions can be done by making indicative urban design visions or housing visions. Shaping might seem somewhat similar to regulating, but it is a more soft way of influencing mixed-income housing. By shaping a specific vision for an area, private developers would ideally conform these visions with their proposals before hard regulation is needed. Since developers are then aware of the expectations of the planning authority, they will adjust their proposals to fit these visions (Rowlands et al., 2016). Good shaping can thus generate better conditions for negotiations between local authorities and developers (De Bruijn & Dicke, 2006).

20

3. An institutional framework of housing development in London and

Amsterdam

As there are differences between the way housing is developed in London (United Kingdom) and Amsterdam (Netherlands), this chapter spells out the institutional differences between both cities and countries with regard to housing development. The first section compares how the different planning systems shape planning practice in both cities (see figure 3). The second section discusses the roles of the different institutions governing housing policy. The third and final section spells out the differences between the housing segments existing in each country. This contextual chapter supports the understanding of the empirical findings and analysis for each city.

3.1. Two different planning systems

Figure 3: Overview of the most central differences between the British and Dutch planning systems

Source: own production, 2018.

3.1.1. London embedded in the British development-led system

The British (this paper focusses on the English and Welsh practice) planning system is embedded in the British common-law, and planning is mainly used to regulate developments, rather than to control or initiate them (as the Dutch system). Public intervention through planning is predominantly done through the assessment of new development proposals. This process is called development control (Janssen-Jansen & Woltjer, 2010). As part of the development control, a developer submits a planning application to the local authority with the aim to get planning permission for development. Local authorities (known as boroughs in London) are responsible for the most important planning document, the Local Development Framework (also known as the Local Plan) (Janssen-Jansen & Woltjer, 2010). This, since 2004 existing, document contains most indicative rules and desired developments for housing, spatial planning and recreation. According to Janssen-Jansen & Woltjer (2010), this document is in principle guiding to test development plans before planning permission is given. The document is however not legally binding and planning permission can also be given when a development is not conforming the Local Development Framework. Achieving planning permission is in practice often a result of extensive negotiations between local authorities and private developers. In the end, planning permission is mostly granted based on argumentation, and not entirely based on a land-use plan (Janssen-Jansen & Woltjer, 2010; Kang & Korthals Altes, 2015).

The British planning system is furthermore characterised by a high level of discretion (Janssen-Jansen & Woltjer, 2010; Heurkens & Hobma, 2014). This discretion is possible because of the non-legally

UK (England) Netherlands

Development-led Plan-led with development-led characteristics Planning to regulate new developments Planning to preserve existing land-use

Reactive, facilitating planning Proactive Municipal-led planning initiating spatial change

Indicative Local Plans provide general guidance for planning

Potentially highly detailed land-use plans form the legal guidance for planning

High degree of discretion allowing flexible ad-hoc solutions

Lower degree of discretion through the land-use plans

Affordable housing typically financed through planning gain

21 binding status of the Local Development Frameworks. Local authorities are therefore free to make separate, and thus different, agreements with different developers and for different projects. This also makes the British planning system much more flexible than the Dutch system and it gives room to political judgement, which is lesser the case in the Netherlands where plans are essentially judged by the law (Janssen-Jansen & Woltjer, 2010).

Affordable housing as part of the housing mix in the UK and London is typically facilitated through planning gain, which is embedded in Section-106 of the 1990 Town and Country Planning Act (Kang & Korthals Altes, 2015; Janssen-Jansen, 2010; Dixon, 2009; van der Veen & Korthals Altes, 2010). By use of Section-106 agreements, local authorities can enforce the developer to reallocate profits (the planning gain) from the development towards social goals such as infrastructure, affordable housing and schools within developments, even when the local authority is not the landowner. When agreed, the Section-106 agreement is a legally binding deed (van der Veen & Korthals Altes, 2010). Planning gain and the Section-106 agreements have been widely criticized for being inconsistent, not transparent, unfair and the negotiations take normally much time (Janssen-Jansen & Woltjer, 2010; Oxley, 2006). Moreover, local authorities might demand either much higher or lower contributions than reasonable for social infrastructure (Oxley, 2006). Moreover, the agreements do not necessarily contain sanctions for when developers do not meet the negotiated requirements (van der Veen & Korthals Altes, 2010).

The relatively small role of public land ownership in London

Land in the United Kingdom is traditionally owned by the aristocracy, as a result of the feudal system. In the 20th _{century, there was an significant increase of public owned land, and the share of public land} (1/6th) reached its top during the Second World War (Home, 2009). The Conservative governments of the 1980s and 1990s, however advocated the sale of many private lands and since then the share of public land has dropped. In London, about 10 percent of the land is publicly owned, while another 10 percent is owned by public-related bodies, such as Transport for London (Dixon, 2009). The remaining 80 percent of the land is owned by private landowners, including the traditional aristocracy, individuals or large international investment firms (Dixon, 2009; Haslett, 2017). Like in Amsterdam and the Netherlands, land is scarce in the UK and London (Home, 2009; Monk et al., 2013). This has led towards increasing and high land costs for developments, particularly in London (Home, 2009; Haslett, 2017; Adonis, 2015). Public land ownership in urban developments in London, is thus much less common than in Amsterdam. Moreover, many public land is already in use, although Dixon (2009) notes that especially many brownfield locations are owned by public bodies.

As most land in London is in private ownership, local planning authorities have a smaller tradition than the Municipality of Amsterdam to actively control developments towards certain planning objectives, such as mixed-income housing developments (Bowie, 2015). The most common way for local authorities to intervene in developments for achieving those social objectives is through Section-106, but this instrument is not linked to public land ownership (Kang & Korthals Altes, 2015; Janssen-Jansen, 2010; Dixon, 2009). Local authorities in the UK can however make use of compulsory purchase to enforce the sale of private land to the local authority. However, in practice this is not often used, as high costs come with this (especially in London) (Dixon, 2009). Moreover, there are less instruments in the UK that can help local authorities to financially gain from active development, making it not very interesting for public parties to buy up lands (Heurkens & Hobma, 2014). Local governments in the UK thus depend on the private land market and planning gain to mix housing programmes of new developments (Bowie, 2015).