The method chosen for this study is of a qualitative approach, which is preferred since the study is of an interpretative nature. A qualitative research strategy contributes to high validity due to the theoretical framework visualized in the data collection. The work is expressed in a natural language, uses small samples and is focused on particular individuals, events, and contexts (Gerring, 2017). Gerring (2017, p. 31) states that “If the work is qualitative, the inference is based on bits and pieces of non-comparable observations that address different aspects of a problem and are traditionally analyzed in an informal fashion.” Data collection of a qualitative study is of a non-standardized form, meaning that questions and procedures can be changed and emerge in a natural way during the research process (Saunders et al., 2016). The qualitative research design can be divided into two different forms; a mono-method qualitative study, using a single data collection technique and analytical procedure, or a multi-method qualitative study where one uses several data collection techniques and analytical procedures. The form used in this study will be of the multi-method qualitative form since an analysis of both written sources and interviews will be conducted. The study can be seen as an interview-based case study.
The literature sources used in this study have been checked to be up-to-date and expertise-based to give a broad theoretical background to the research and to ensure what has already been treated in the research area. The literature review included reading international scientific articles, books, government and municipal reports as well as previous university theses connected to theory regarding segregation, affordable housing and inclusionary housing in South Africa as well as around the world.
The master thesis was based on evaluating particularly Cape Town as a segregated city and as a potential for an inclusionary housing tool. In Cape Town, the study was conducted with the help of Professor Mikael Samuelsson at the University of Cape Town Graduate School of Business and with online assistance from Professor Mats Wilhemlsson at the Royal Institute of Technology in Stockholm.
A closer investigation was made on a specific development project in Sea Point where an inclusionary housing pilot project was going to be implemented. The aim of evaluating the pilot project in the study was to illustrate a practical example and gain deeper understanding on how inclusionary housing could possibly be used and implemented in Cape Town.
The advantages with conducting a case study includes that it has the ability to produce intense and in-depth research. It can identify and answer the questions about what and why it is happening, and possibly give an understanding of the effects from the situation and implications for actions (Saunders et al., 2016). Therefore, to conduct an interview based case study is the most suitable research strategy to answer our research questions.
According to Saunders et al. (2016) high reliability during a research study can be achieved by conducting the interviews with at least two researchers within the project. This can contribute to that the results, analysis and conclusions are evaluated and interpreted similarly and have been agreed upon the researchers’ perception.
Semi-structured interviews were supposed to be conducted in Cape Town with different stakeholders. Due to the circumstances regarding Covid-19, interviews were held via the electronic tools Skype and Zoom. Nine semi-structured interviews were successfully held among the requested. These interviews provided information on the various stakeholders’ view on inclusionary housing and how it can be implemented in Cape Town.
When using semi-structured interviews the researcher decides upon a theme and key questions that should be covered during the interview. However, the use of them and what questions will be asked may vary interview to interview (Saunders et. al., 2016). The order of the questions may also vary and questions can be added (Eriksson & Kovalainen, 2008). Since the interviewees will be stakeholders from different sectors, their opinions and views will most likely differ and therefore the interviews will possibly go down different paths. Hence, the semi-structured interview technique is preferable since it allows the interviewer to ask other questions than the questions prepared. This method is seen as a unique method among the different interview methods since it gives a high degree of relevance for the topic while remaining responsive to the interviewee (Bartholomew, Henderson, & Marcia, 2000 cited in Mcintosh & Morse, 2015).
Hanna (2012) discusses the advantages and challenges with using electronic tools such as Skype. There are a lot of practical benefits when using electronic tools. It is easier to set up an interview since there is no travel time for the participants and researchers as well as the freedom to shift the interview at the last minute. Further, it allows both the interviewer and interviewees to stay in a familiar setting, for example in their home, without imposing on each other’s personal space which can make the interview situation more relaxed. Additionally, one can easily record both the visual and audio interaction directly through the electronic tool which makes it much easier for the researcher at the transcription stage. However, one must bear in mind that some technical difficulties can occur such as problems with bad internet connection and that the camera or sound is not working properly.
Interviews were conducted with various stakeholders from the private sector, public sector, NGO’s and academics with the aim to provide their different perspectives on inclusionary housing and to answer the question if it is the right tool to help Cape Town become a more integrated city.
In this report, the private sector is represented by the developer FWJK, the South African Property Owners Association (SAPOA), Multi QS and the town planning consultancy firm Nigel Burls & Associates. The two last mentioned have also been involved in work with the public sector. However, in this report they are representing the private sector. A researcher at the Urban Real Estate Research Institute at the University of Cape Town represents the perspective of academics. The Secretary-General at the Good party in Cape Town represents the public sector. As a complement to the previous stakeholders the Development Action Group (DAG) and Ndifuna Ukwazi were interviewed representing the perspective from the NGO’s.
The aim of the study is not to make any political statement or provoke any part involved. Since different stakeholders will be interviewed it is important to stay objective throughout the study. Personal opinions will not take any part in the master thesis, only facts related to the research questions and the current housing situation in Cape Town. According to Saunders et al. (2016) participant validation is achieved when the interviewees have the opportunity to take part in the research study to confirm the accuracy before it is published. Since a qualitative method is used in this study, recordings will be carefully analyzed and safely stored with encryption.
Concept of residential segregation
Majority of papers see segregation as a residential separation of groups from a wider population within cities (Leal, 2012). Massey & Denton (1998, p. 282) define residential segregation as “the degree to which two or more groups live separately from one another, in different parts of the urban environment”. Residential segregation leads to a situation where a certain group is concentrated in neighborhoods with a higher amount of jobs, better public services and infrastructure and lower crime rates while another group is concentrated in areas with the opposite; areas with fewer jobs, poor public services and infrastructure and higher crime rates (Massey & Denton, 1998 through Condron et al., 2012).
Residential segregation can occur by income and ethnic inequalities of residents (Intrator et al., 2016). Racial segregation can be defined as physical and social isolation of specific races and minorities in homogenous and disadvantaged neighborhoods (Quane and Wilson, 2012). Reardon & Bischoff (2011) highlight the strong correlation between income and racial segregation. Racial segregation has the capacity to produce a certain level of income segregation even though no within race income segregation occurs. By definition, income segregation implies that on average, low-income households live in communities with low-income levels while high-income communities are occupied by high-income households (Reardon & Bischoff, 2011). Reardon & Bischoff (2011, p.1093) describe income segregation as “the uneven geographic distribution of income groups within a certain area”. According to Reardon & Bischoff (2011) income segregation may particularly be characterized by the spatial segregation of poverty (i.e. the extent to which the lowest-income households are isolated from middle- and upper-income households) and/or the spatial segregation of affluence (i.e. the extent to which the highest-income households are isolated from middle-and lower-income households). Income segregation can also occur due to spatial differences in land prices which are generating homogenous communities (Hardman & Ioannides, 2004).
Clark & Ledwith (2007) state that there is substantial literature regarding a household’s neighborhood choice. Most researchers agree that it is based on a complex interaction of income, socioeconomic status and preferences. Reardon & Bischoff (2011) and Clark & Ledwith (2007) argue that the neighborhood choice is based on households’ preferences such as neighborhood amenities (hospitals, schools, kindergartens etc.) and their ability to pay for housing. When a household is choosing a house or an apartment, it is choosing more than just a residence (O’Sullivan, 2012). It is also choosing a set of local public goods as well as neighbors who provide opportunities and interactions. O’Sullivan (2012) has adapted The Neighborhood choice model of Becker and Murphy (2000) regarding neighborhood choice and segregation due to income and race. The following model assumes that there is a competitive market for housing in each neighborhood, meaning that a house is sold to the highest bidder regardless of the neighbors wishes. This model highlights the importance of a tool like inclusionary housing to break segregation and foster greater socioeconomic integration.
Neighborhood choice model – Income segregation
This model considers a city with two different neighborhoods; A and B, and two different income groups; a high-income group and low-income group, each with 100 households. An assumption is that the positive externalities increase with household income, i.e. the attractiveness of a neighborhood will increase with the number of high-income households. The only differences between the two areas are the income mix and therefore the resulting neighborhood externalities. In this model the equilibrium point requires all households, both high- and low-income, to pay the same rent. In all scenarios, both curves show the rent premiums for the two income groups. They are positively sloped because of the assumption that high-income households generate positive externalities.
From the model, O’Sullivan developed three possible scenarios leading to different outcomes:
● Segregation equilibrium is unstable (segregation)
● Segregation equilibrium is stable (integration)
● Mixed neighborhood equilibrium
Segregation equilibrium unstable (segregation)
Figure 1: Segregation (O’Sullivan, 2012, Fig. 8-2, p. 209)
In this scenario, the horizontal axis measures the number of high-income households in neighborhood A, and the vertical axis measures the difference in land rent between the two neighborhoods. The assumption is that high-income households generate positive externalities, meaning that the rent premium is always positive. The larger number of high-income households, the larger premium.
Point k and j show that high-income households are willing to pay $8 more to live in neighborhood A rather than B and points respectively that low-income households are willing to pay $5 more to live in neighborhood A. Point i shows where the two neighborhoods are identical, i.e. the equilibrium. Point s shows where all high-income households are in neighborhood A, i.e. income segregation.
In this scenario, the equilibrium point is unstable, due to a small change in the population will result in a new equilibrium. Since the high-income curve is above the low-income curve, the high-income households will outbid the low-income households and will continue to increase at the expense of the low-income households. Segregation will happen since the high-income households have a steeper premium curve which is a reflection of the benefits of living close to high-income neighbors.
Segregation equilibrium is stable (integration)
Figure 2: Integration (O’Sullivan, 2012, Fig. 8-3, p. 211)
In this scenario, the low-income households have a steeper premium curve. The equilibrium point i is stable since low-income households are willing to pay more than high-income households to live in a neighborhood with more high-income households. Any deviation from the equilibrium will be self-correcting, and low-income households will outbid the high-income households so that the low-income households will increase to the expense of the high-income households. This scenario will lead to integration.
Neighborhood choice model – Racial segregation
The neighborhood choice model can also be applied to racial segregation. When the premium curve for white household’s general preferences is laying above the premium curve for black households, the equilibrium point is unstable, leading to racial segregation since the white curve is above the black curve. The white households will outbid the black households and will continue to increase at the expense of the black households.
To reach a stable equilibrium and racial integration, the premium curve for black households needs to be steeper than the curve for white households at the origin. In this case, the black households would outbid the white households. The deviation from the equilibrium would be self-correcting since if the white population would increase above a certain amount, the black households would outbid the white households for the limited number of places in the neighborhood.
The definition of the term affordable housing varies across countries around the world, but the definitions do have some common ground. Affordable housing is in general described as housing, in different tenure forms, that is accessible in price to low- and moderate-income households through subsidizations, regulations or other arrangements. What differs is how countries identify the low- and moderate-income groups (Lawson & Milligan, 2007). Milligan & Gilmour (2012, p.58) defines affordable housing as “housing that is provided at a rent or purchase price that does not exceed a designated standard of affordability.” With affordability Milligan and Gilmour refers to “measuring whether housing costs exceed a fixed proportion of household income and/or whether household income is sufficient to meet other basic living costs after allowing for housing costs.”
Table of contents :
1.4 Research questions
2.1 Research design
2.2 Literature review
2.3 Case study
2.4.1 Interview technique
2.4.2 Interview participants
3 Theoretical background
3.1.1 Concept of residential segregation
3.1.2 Neighborhood choice
22.214.171.124 Neighborhood choice model – Income segregation
126.96.36.199 Segregation equilibrium unstable (segregation)
188.8.131.52 Segregation equilibrium is stable (integration)
184.108.40.206 Neighborhood choice model – Racial segregation
3.2 Affordable housing
3.3 Inclusionary housing
4 Literature review
4.1 Inclusionary housing around the world
4.1.1 Inclusionary housing in the United States
4.1.2 Inclusionary housing in Canada
4.1.3 Inclusionary housing in South Africa
4.1.4 Inclusionary housing policy in Johannesburg
5 Case study: South Africa, Cape Town
5.1 Spatial segregation in Cape Town
5.2 Housing situation in Cape Town
5.2.1 Current housing programs in Cape Town
220.127.116.11 BNG Program
18.104.22.168 Social Housing
5.2.2 Housing prices and incomes in Cape Town
5.2.3 Cost drivers
5.2.4 Inclusionary housing in Cape Town
5.6 The Fulcrum project
5.6.3 Target market and selection criteria
5.6.4 Tenure Form
5.6.5 Rental price and price adjustment
22.214.171.124 Do you think anything should have been done differently in the project?
6 Results and analysis of interviews
6.1.1 Definition of Affordable housing
6.1.2 Definition of Inclusionary housing and the Gap market
6.2 Advantages and challenges
6.2.1 What advantages and challenges do you see with inclusionary housing?
126.96.36.199 Public sector perspective
188.8.131.52 Private sector perspective
6.2.2 What are the challenges with implementing affordable housing in higher income areas?
6.3 Design of inclusionary housing policies
6.3.1 Lessons learned from South African policies
6.3.2 Program structure
184.108.40.206 Mandatory versus Voluntary approach
220.127.116.11 National versus Local approach
6.3.4 Target groups and selection criteria
6.3.5 Tenure form
6.3.6 Price adjustment and exchange of ownership
6.3.7 Alternative approaches
6.3.8 Overall structure recommendations
6.4 Stakeholder engagement and collaboration
6.4.1 Public sector
6.4.2 Private sector
6.4.3 NGOs and activist groups
6.5 Future outlook
6.5.1 How could a city like Cape Town become more integrated?
6.5.2 Do you think that inclusionary housing is a tool that could help the city to become more
6.5.3 How do you think the distribution and extent of inclusionary housing will look like in Cape
Town in 15 years?
7 Conclusion & further studies
7.2 Further studies