literature review
Engvdapi
1-The_impact_of_site_and_serviceOnHousingNairobi.pdf
Reproduced w ith permission o f the copyright o w n e r Further reproduction prohibited without permission
1+1 National Libraryof Canada Biblioth&que nationaledu Canada Canadian Theses Service Service des thgses canadiennes
Ottawa. Canada K1A0N4
NOTICE AVIS
The quality of this microform is heavily dependent upon the quality of the original thesis submitted for microfilming. Every effort has been made to ensure the highest quality of reproduction possible.
If pages are missing, contact the university which granted the degree.
Some pages may have indistinct print especially if the original pages were typed with a poor typewriter ribbon or if the university sent us an inferior photocopy.
Reproduction in full or in part of this microform is governed by the Canadian Copyright Act, R.S.C. 1970, c. C*30, and subsequent amendments.
La qualit6 de cette microforme depend grandement de la quality de la th&se soumise au microfilmage. Nous avons tout fait pour assurer une quality sup6rieure de reproduc tion.
Sil manque des pages, veuillez communiquer avec I'umversite qui a conf6r6 le grade.
La quality d’impression de certaines pages peut laisser k d6sirer. surlout si les pages originates ont 6t6 dactylogra p h ies k I'aide d un ruban us6 ou si I'universii nous a fait parvenir une photocopie de quality inf6rieuro
La reproduction, m6me partielle, de cette microtorme est soumise k la Loi canadienne sur te droit d'auteur, SRC 1970, c. C-30, et ses amendements subs6quents
NL-339 (< 68/04] c Canada Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
THE IMPACT OF SITE AND SERVICE PROJECTS ON URBAN HOUSING MARKETS: THE CASE OF DANDORA, NAIROBI
by
Joyce Mwende Maloxnbe
Department of Geography
Submitted in partial fulfilment of the requirement for the degree of
Doctor of Philosophy
Faculty of Graduate Studies The University of Western Ontario
London, Ontario February 1990
©Joyce Mwende Malombe 1990
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
National Library of Canada
Bibhotheque naiionaie du Canada
C anadian Theses Service Service des theses canadiennes
Ottawa. Canada K1 A 0N 4
The author has granted an irrevocab's non exclusive licence allowing the National Library of Canada to reproduce, loan, distribute or sell copies of his/her thesis by any means and in any form or format, making this thesis available to interested persons.
The author retains ownership of the copyright in his/her thesis. Neither the thesis nor substantial extracts from it may be printed or otherwise reproduced without his/her per mission.
L’auteur a accorde une licence irrevocable et non exclusive permettant a la Biblioth^que nationale du Canada de reproduire, prfeter, distribuer ou vendre des copies de sa these de quelque maniere et sous quelque forme que ce soit pour mettre des exemplaires de cette these a la disposition des personnes interessees.
L’auteur conserve la propriete du droit d’auteur qui protege sa these. Ni la these ni des extraits substantiels de celle-ci ne doivent etre imprimes ou autrement reproduits sans son autorisation.
ISBN 0-315-55275-1
Canada Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
The IMVERSnYqf WESTERN ONTARIO Faculty o f Graduate Studies
In the interests o f facilitating research by others at this in s titu tio n and elsewhere, I hereby grant a licence to:
or substantial parts thereof, the copyright which is invested in m e, provided th at the licence is subject to the follow ing conditions.
1. O n ly single copies shall be made or authorized to be made at any one tim e , and o n ly in response to a w ritten request from the library o f any U niversity or sim ilar in stitu tio n on iis ow n b eh alf or on behalf o f one o f its users.
2. This licence shall continue for the full term o f the copyright, o r fo r so long as may be legally perm itted.
4. H ie Universal Copyright Notice shall appear on the title page o f all copies o f m y thesis made under the au th o rity o f this licence.
4. This licence does not perm it the sale o f authorized copies at a p ro fit, but does p erm it the collection by the institution or institutions concerned o f charges covering actual costs.
5 A ll copies made under the a u th o rity o f this licence shall bear a statement to the effect that the copy in question “ is being made available in this form by the a u th o rity o f the copy right ow ner solely fo r the purpose o f private study and research and may not be copied or reproduced except as p erm itted by the copyright laws w ith o u t w ritte n a u th o rity from the copyright o w n e r."
0. The foregoing shall in no way preclude my granting to the N atio n al L ib rary o f Canada a licence to reproduce m y thesis and to lend or sell copies o f the same.
T H E U N IV E R S IT Y O F W E S T E R N O N T A R IO
to make copies o f m y thesis THE IMPACT OF SITE AND SERVICE PROJECTS ON URBAN HOUSING MARKETS:
THE CASE OF DANDORA, NAIROBI
( signature o f witness ') (signature o f student)
May 1, 1990 Ph.D. Geography (date) (degree) (departm ent o f student)
with permission of the copyright owner. Further reproduction prohibited without permission.
THE UNIVERSITY OF WESTERN ONTARIO FACULTY OF GRADUATE STUDIES
CERTIFICATE OF EXAMINATION
Chief Advisor
Advisory Committee
The Thesis by Joyce M. Malombe
entitled The Impact of Site and Service Projects on Urban Housing Markets: The case of Dandora, Nairobi.
is accepted in partial fulfilment of the requirements for the degree of
Doctor of Philosophy
Date Chairman of Examining Board
i i
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
ABSTRACT
This dissertation is an evaluation of site and service projects as a strategy for meeting the housing needs for the low-income households in Nairobi, Kenya. These projects were an attempt by the government to help low-income households by increasing the housing stock available to these kinds of households.
The major source of data was from interviews conducted over a period of five months in the large Dandora project in Nairobi. Here, information was gleaned on changes in socio economic, housing and migratory variables.
The data confirmed that the project significantly improved the housing conditions of those who moved. The project, however, did not directly meet the housing need of the low-incomes group because of the high housing costs of this project. The Dandora project benefited only a small percentage of residents from the squatter settlements and these characteristically earned higher incomes than were intended for this project. The analysis showed that the project mainly benefited respondents who had moved from standard housing (housing constructed with permanent materials and approved by the City Council) . However, this does not necessarily indicate the failure of this housing strategy beceiuse the benefits to the ill housed in the squatter settlements could have been indirect ones resulting from
iii
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
filtering. The survey determined that 85 percent of those moving from standard housing left vacancies that could have benefited low-income households through filtering. An analysis of the migratory history of those surveyed indicated that most of the respondents had moved from the rural areas to standard housing prior to Dandora and not from the squatter settlements, suggesting that even with the second tier moves included, the site and service schemes' direct impact on the City's squatter settlements was limited. Even so, by enabling rural households to move to standard housing (which would not have been the case if the project had not been implemented), the project indirectly provided housing for migrants, who would have had more limited alternatives to the squatter settlements.
iv
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
ACKNOWLEDGEMENTS
I wish to express my sincere gratitude to several organizations and individuals whose support and contribution made possible the successful completion of this work. Special thanks must go to my chief advisor, Dr. W.R. Code, for his assistance and patience throughout the writing of this thesis. I am grateful for his having read the drafts and offered constructive criticisms, suggestions and guidance. My special thanks also go to Dr. D. Janelle, Dr. M. Goodchild, Dr H. Hosse, Dr. E. Bjorklund, Dr. A. Philbrick and Dr. D. Cartwright and indeed to all the faculty for their support and encouragement. I am particularly grateful to Kim Holland for his support and great help with the maps. As well, I am grateful to Tim Bailey for his friendly he’p. I also appreciate the on going support of my other colleagues: Catherine Hooey, Quentin, Bert, Gloria, Tom, Shuo Chao Fu among others.
I wish to acknowledge the help from the administrative staff and especially Diane Shillington. Outside the Department, my thanks are extended to many friends for their great support and encouragement. I am especially grateful to Linda McLean, Kimberley Granger, Michele Anderson, Ruth Hufnagel, Ben Hartford and Wambui Kiai for all their support and help and for making mv stay in Canada such a pleasant experience. I am also grateful to Angeliek Wick, Yaap
v
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Hogeling, Ruth Knutson, Easther Waruiru, Beatrice Adimola and Dr. Tom Mcfarlane.
I wish to thank the Canadian International Development Agency for the financial support. I am thankful to the University of Nairobi for giving me study leave. Special thanks go to Dr. Syagga and all the staff of Housing Research and Development Unit for their help during the field work. I am greatly indebted to Mr. Muindi Mulili, the research assistant, for his faithfulness and support. His experience in housing research at the university was of great help during the challenging field work. A very special thanks must also be extended to the individuals and families in Dandora who made this research possible. Finally I want to thank my family for their unfailing support. I especially want to thank my mother and my father and my brother to whom I dedicate this work. I greatly appreciated my brother's help in transporting me daily to Dandora. Above all, I thank God without whom this work would not have been completed.
vi
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
TABLE OF CONTENTS Page
CERTIFICATE OF EXAMINATION ............................... ii ABSTRACT .................................................... iii ACKNOWLEDGEMENTS .......................................... v TABLE OF CONTENTS ......................................... vii LIST OF APPENDICES ........................................ X LIST OF TABLES ............................................. xi LIST OF FIGURES ............................................ xiv
CHAPTER 1 INTRODUCTION .......................................... ... 1 1.1 Introduction ......................................... 1 1.2 Problem Statement ................................... 3 1.3 Operational Definitions ............................ 10 1.4 Organization of the Study .......................... 15 1.5 Description of the Dandora Community Project .... 17
1.5.1 Historical Development of the Site and Service Strategy ............................... 17
1.5.2 The Dandora Community Development Project ......................................... 19
1.5.3 Goals and Objectives of the Dandora Project ......................................... 21
1.5.4 Implementation of the Project ................. 25 1.5.5 Conditions of the Lease Agreement.... ......... 30 1.5.6 House Type Plan Options ........................ 31
CHAPTER 2 LITERATURE REVIEW ......................................... 34 2.1 Introduction ......................................... 34 2.2 Third World Urban Housing .......................... 35 2.3 The Filtering Process .............................. 43
2.3.1 The Filtering Process in Nairobi's Urban Housing Market ................................. 4 6
2.4 Urbanization in Kenya ............................... 49 2.5 National Housing Policy ............................ 51 2.6 Evolution of Housing Policy in Nairobi ........... 63
2.6.1 Housing Before Independence........ 66 2.6.2 Housing After Independence ................... 7 3
2.7 Housing Programmes .................................. 84 2.7.1 Nairobi City Council ........................... 86 2.7.2 National Housing Corporation .................. 92 2.7.3 Housing Finance Company of Kenya ............. 94
2.8 Conclusion ........................................... 98
vii
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Page CHAPTER 3 DATA SOURCES AND RESEARCH METHODOLOGY .................. 100 3.1 Research Design ..................................... 100 3.2 The Method of Data Collection ..................... 102 3.3 Sampling Procedure .................................. 103 3.4 Questionnaire Design ................................ 106 3.5 Field Investigation Problems ...................... 116 3.6 Method of Data Analysis ............................ 118 CHAPTER 4 SOCIO-ECONOMIC CHARACTERISTICS AND HOUSING CONDITIONS OF THE DANDORA RESIDENTS ..................... 121 4.1 Introduction ......................................... 121 4.2 Socio-economic Characteristics ..................... 122
4.2.1 Age Distribution and Length of Stay in Nairobi ................................ 124
4.2.2 Family Composition of the Dandora Households ...................................... 127
4.2.3 Jobs Held by the Dandora Residents .... 130 4.2.4 Transportation System and Cost to
the Residents ........ 134 4.2.5 Monthly Incomes of the Dandora
Residents........................................ 140 4.3 Building Process ..................................... 143 4.4 Housing Conditions in the Dandora Project......... 153
4.4.1 Number of Rooms Occupied by the Respondents in Dandora......................... 154
4.4.2 Available Services in the Dandora Project ......................................... 158
4.4.3 Monthly Housing Costs in the Dandora Project ......................................... 161
4.4.4 Reasons for Moving to Dandora ................ 166 4.5 Respondents' Housing Conditions Before
Moving to Dandora .................................... 169 4.6 Evaluation of Housing Improvements as a
Result of the Move to Dandora .............,....... 180 4.7 Conclusions ............. 191 CHAPTER 5 AN EVALUATION OF HOUSING BENEFITS AS A RESULT OF A DIRECT MOVE TO DANDORA FROM THE SQUATTER SETTLEMENTS ................................................ 194 5.1 Introduction ......................................... 194
viii
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Page 5.2 Socio-economic Characterise!cs of the
Respondents Moving from the Squatter Settlements to Dandora ............................. 199
5.3 Evaluation of Housing Conditions as a Result of the Move from the Squatter Settlements to Dandora .......................................... 207
5.4 Direct Move from the Rural Areas to Dandora ....... 221 5.4.1 An Evaluation of Housing Conditions
a Result of the Move from the Rural Areas to Dandora ......................... 228
5.5 Conclusions ......................................... 235
CHAPTER 6
THE MIGRATION FROM OTHER HOUSING AND THE INDIRECT IMPACT OF THE SITE AND SERVICE PROJECTS ON THE SQUATTER SETTLEMENTS ..................................... 239 6.1 Introduction ........................................ 239 6.2 Socio-economic Characteristics of the
Residents Moving from Standard Housing to Dandora .......................................... 242
6.3 An Evaluation of Improvement in Housing Conditions as a Result of the Move from Standard Housing to Dandora ....................... 249
6.4 An Evaluation of Indirect Benefit to Residents of the Squatter Settlements as a Result of the Move to Dandora from Standard Housing to Dandora ................................. 261
6.4.1 Did the People Moving from Standard to Dandora Begin a Vacancy Chain.............. 262
6.4.2 The Origin of Households Occupying Residences Made Available as a Result of the Move to Dandora ........... ............ 2 67
6.4.3 The Indirect Impact of the Dandora Project on the Growth of the Squatter Settlements .................................... 269
6.5 Conclusions ........................... 272 CHAPTER 7 SUMMARY AND CONCLUSIONS ...........................:..... 275 7.1 Introduction ........................................ 275 7.2 Summary .......,...................................... 275
ix
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Page 7.2.1 Household Characteristics of the
Dandora Residents ............................. 275 7.2.2 , Housing Conditions as a Result of
the Move to D a n d o r a ........................... 278 7.3 Conclusions .......................................... 281
7.3.1 Did the Site and Service Schemes Address the Housing Needs of the Target Income Group? .......................... 282
7.3.2 Did the Project Allow Incremental Housing Development? .................................. 284
7.3.3 Were Housing Costs Reduced Through Self-help and Residents' Involvement? ....... 285
7.3.4 Did the Project Provide Security of Tenure? ......................................... 287
7.3.5 Did the Project Provide Affordable Services? ...................................... 288
7.3.6 Did the Project Contribute to the Elimination of Squatter Settlements? ........ 288
7.4 Recommendations ..................................... 292
LIST OF APPENDICES I Grade II By-laws ................................... 295 II Questionnaire ....................................... 302 III Profile of Plot Owners and Tenants ................ 312
BIBLIOGRAPHY ............................................... 330 VITA ........................................................ 344
x
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
list of tables Description Page
2.1 Centres by Size of Urban Population 1948-1979 .... 52 2.2 Estimated Housing Need by Category 1983-2000 .... 52 2.3 Level of Crowding in Urban Areas in 1965 ........ 55 2.4 1977 Estimated Household Incomes in Nairobi ..... 65 2.5 Population Growth in Nairobi Since 1906 .......... 75 2.6 Estimated Demolicions of Squatter Settlements
Since 1960........................................... 77 2.7 Construction Date and Current Rents of City
Council Housing Units in Eastlands ............... 88 2.8 City Council Rental Housing Located in Other
Parts of the City .................................. 89 2.9 City Council Tenant Purchase Projects ............ 91 2.10 Yearly Mortgage Advanced by the Housing
Finance Company of Kenya Since 1966 .............. 97 4.1 Age Distribution of Dandora Residents ............ 126 4.2 Number of Years the Respondents have live in
Nairobi . ........................................... 126 4.3 Jobs Held by the Respondents ...................... 132 4.4 Distance of Place of Work frcm Dandora ...... 135 4.5 Monthly Transportation Cost for the
Respondents ......................................... 139 4.6 People Involved in the Construction Process ..... 147 4.7 Average Number of skilled People Employed
During Construction of Dandora Plots ............. 149 4.8 Average Number of Unskilled Workers Employed
in Dandora .......................................... 149 4.9 Estimated Construction Cost of the
Dandora Plots ....................................... 152
xi
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Page 4.10 Estimated Electricity Connection Cost
in Dandora .......................................... 152 4.11 Number of Persons per Room in Dandora ............ 157 4.12 Total Number of Rooms the Plot Owners
were Subletting ..................................... 160 4.13 Total Income from Subletting ...................... 160 4.14 Total Housing Cost Incurred by the
Dandora Residents.................................... 162 4.15 Reasons for Moving to the Dandora Project ........ 167 4.16 Location of Housing Estates the Residents
Lived in Before the Move to Dandora .............. 170 4.17 House Types the Respondents Lived in Before
Moving to Dandora ................................... 17 4 4.18 Number of Persons per Room Prior to the Move
to Dandora .......................................... 17 6 4.19 Rent Paid per Month Before the Move to Dandora ... 179 4.20 Available Services Before the Move to Dandora .... 179 4.21 Ranking of Housing Components ..................... 181 4.22 Number of Rooms Before and After the Move to
Dandora .............................................. 188 4.23 Number of Persons per Room Before and After the
Move to Dandora ..................................... 188 5.1 Jobs Held by the Respondents who Moved
Directly from the Squatter Settlements to Dandora .......................................... 203
5.2 A List of the Different Squatter Settlements the Respondents Lived in Before Moving to Dandora .............................................. 209
5.3 Number of Rooms in the Squatter Settlements and in Dandora .......................... 211
5.4 Number of Persons per Room in the Squatter Settlements .................................... 213
xii
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Page 5.5 Jobs Held by the Respondents Moving from the
Rural Areas to Dandora ............................. 224 5.6 Number of Persons per Household in the Rural
Areas and in Dandora ............................... 230 5.7 Number of Persons per Room in the Rural Areas
and in Dandora ..................................... 232 6.1 Jobs Held by the Respondents Moving from Standard
Housing to Dandora ................................. 245 6.2 The Housing Estates the Respondents Lived in
Prior to Dandora ................................... 250
6.3 Number of Persons in Standard Housing and in Dandora ............................................. 254
6.4 Number of Persons per Room in Standard Housing and in Dandora ............................. 256
y '.ii
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
LIST OF FIGURES
Description Page 1 Direct and Indirect Move to Dandora ............... 6 2 Map of Dandora ..................................... 2 6 3 House Type Design Options .......................... 31 4 Typical House Plans for Dandora Project
(Four-roomed Unit) 3 2 5 Nairobi Boundary Changes Between 1900 and 1963 ... 67 6 Nairobi in 1903 ........... ......................... 68 7 Land use in Nairobi in 1939 ........................ 70 8 Some of the Major Housing Estates in Nairobi ..... 79 9 Dandora Project in Relation to Estimated
Income Distribution in Nairobi ................... 80 10 Sampling Procedure .................................. 105 11 Respondents' Monthly Incomes ....................... 141 12 Owners' and Tenants' Monthly Incomes .............. 142 13 Plot Construction Time in Years ................... 14 5 14 Number of Rooms Occupied by the Respondents ...... 155 15 Housing Expenses for Owners ........................ 164 16 Comparison of Tenants and Owners
Housing Expenses ................................... 165 17 Available Services Before and After
the Move to Dandora ................................ 184 18 Rents Before and After the Move to Dandora ....... 190 19 Migration Histories of the Dandora Residents ..... 196 20 Migration Histories of the Residents Moving
from the Squatter Settlements to Dandora ......... 198
xiv
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Page 21 Monthly Incomes for the Residents who had Moved
from the Squatter Settlements to Dandora ......... 2 05 22 Locations of the Squatter Areas the Respondents
had Lived in Before Moving to D a n d o r a ............ 208 23 Housing Expenses Before and After the Move to
Dandora .............................................. 216 24 Available Services in the Squatter Settlements
and in Dandora ...................................... 219 25 Monthly Incomes for the Residents Moving from
the Rural areas to Dandora................. 227 26 Migration History of Residents Moving from
Standard Housing to Dandora ....................... 240 27 Monthly Incomes of the Respondents Moving from
Standard Housing to Dandora ....................... 24 8 28 Standard Housing Origin Areas of the Dandora
Residents ............................................ 251 29 Number of Rooms in Standard Housing and in
Dandora .............................................. 253 30 Housing Expenses in Standard Housing and in
Dandora .............................................. 2 58 31 Available Ser-ices in Standard Housing and in
Dandora .............................................. 260 32 Housing Left Behind by Respondents Moving from
Standard Housing .................................... 263 33 Migration Histories of the Residents Moving to
Dandora Indirectly from the Rural Areas .......... 268
xv
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER 1
INTRODUCTION
l.l introduction
The general theme of this study is the provision of housing for low-income families in the cities of the Less Developed Countries, and its specific concern is with such provision in Nairobi, Kenya. Most cities in the Less Developed Countries have experienced rapid growth in the last three decades, growth which has rarely been accompanied by the necessary development of infrastructure, and this deficit has frequently induced problems. One of the major problems is an acute shortage of housing for the majority of urban dwellers. The initial responses to this housing shortage were attempts to eliminate slums and squatter settlements and to limit the growth of cities. One of the most direct attempts to address the shortage was the construction of public housing projects. As Payne has commented, although dwellings in these projects usually conformed to high standards of construction and services provision, they were far too expensive for the households for which they were designed and required such heavy subsidies, that they were unable to meet more than a small proportion of the total housing demand (1984, 1).
The end result of this mismatch in demand and supply has been continued expansion of squatter settlements and
1 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
increased densities in existing low income areas (Payne 1984, 2). The growth of these sguatter areas has created the need for a more realistic approach to the housing problem. Previous studies by J.F Turner and W. Mangin, among others, concluded that people were the best judges of their housing need, and that in most cases they were capable of obtaining it. The government's role was seen as that of providing tenure and basic services. Financial support by the World Bank led to the creation of site and service projects. Site and service projects provide urban land in small plots so that individual households can build their own dwellings. In these projects, lots are levelled and furnished with access roads, drainage, water, sewerage, electricity, health clinics, refuse collection, fire protection and other services. Projects of this type were implemented in Nairobi in 1975.
This investigation examines the impact of this housing policy as a strategy for meeting the housing need for low- income urban dwellers. By interviewing both owners and tenants in the large Dandora project in Nairobi, it was established the degree to which benefit accrued and to whom. The interviews identified whether beneficiaries came directly from squatter settlements as intended, or whether the project drew groups housed in better housing outside the squatter settlements, and if so, whether through vacancy chains opened other opportunities for people to move from the squatter settlements. The investigation also provided insight into the overall structure and functioning of the housing market and
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3
indicated whether this housing policy directly or indirectly addresses the housing need of Nairobi's low-income population.
1.2 Problem Statement
The purpose of this dissertation is to evaluate the impact of the Government's policy of providing housing to low- income people through site and service projects. Large scale implementation of these projects was started in Nairobi in 1975, with the implementation of the Danuora project. Site and service projects are seen as the government's tool for spreading housing benefits to the urban poor who had not benefited from the public housing constructed since political independence (1963). The policy addressed the cost of construction and services, building standards and land tenure. One policy goal was to reduce the cost of construction by generating community group involvement in the construction process. The reduction of housing cost was supposed to make it possible for squatters to move directly into these projects.
This study examines whether these projects improved the housing situation of low-income urban dwellers through direct movement to the projects, or indirectly through the filtering process. Change in the housing quality of the households is examined. Housing improvement is evaluated according to quality, level of services provided, and degree of crowding.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Housing quality is assessed according to the type of building materials used for construction. The level of services provided is determined by the provision of such basic minimum services as availability of water, bathroom and kitchen facilities. To assess the importance of these components, the respondents were asked to rank them in their order of importance. This ranking made it possible t ̂ give a numerical value to each of the components. Also, the respondents were asked to rate housing components according to their perceived importance. This ranking showed the relative weight the inhabitants place on the different aspects of housing, thus indicating whether or not the most important component of housing for low-income residents is provision of services as advocated by the site and service policy. These values were used to make a housing quality index which enabled ar assessment of the housing situation of beneficiaries before and after the move to Dandora.
Two other important aspects of housing considered are crowding and cost. Crowding is assessed according to the number of occupants per room. The level of overcrowding was determined by the residents themselves answering some of the questions addressed by the questionnaire. The cost of housing to the occupants is measured according to the percentage of their income devoted to housing. Residents spending more than 25 percent of their income on housing were considered to be paying too much.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
5
Central to the study are the migration histories of the beneficiaries. The analysis of migration patterns showed where Dandora residents came from, and their housing situation at each stage in their migration history. Four hypothetical categories of migration histories from which the respondents could directly or indirectly benefit from the creation of the Dandora project were identified (see Figure 1).
Case 1 This is a situation where a household moved directly from squatter settlements to Dandora. Squatter settlements have the worst kind of housing which is mainly built with temporary materials and is characterized by overcrowding and a lack of basic services. These settlements are usually entry points for migrants to the City and are growing rapidly in number. One of the government's methods of limiting the growth of squatter settlements was the implementation of site and service projects. The provision of basic services and the reduction of construction costs through community involvement in the building process was meant to reduce housing costs, thus make it affordable to low-income groups. This would in turn make it possible for squatters to move directly from squatter settlements to site and service projects. The study made it possible to quantify the degree to which households moved directly from squatter settlements as intended. The housing condition of the squatters before and after the move to Dandora is also determined. In the cases where residents moved directly from squatter settlements, the project met the
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
6
DIRECT AND MDIRECT MOVE TO DANDORA
Casa 1 Direct mova tha aquattar aattiamanta
Diract mova from tha rural
Dir act mova from standard housing
Mova from tha aquattar sattinmnnts to atandard housing prior to Dandora
Squattar Sattlamanta Sits and Ssrvics
Projacta
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
7
housing need of the urban poor in Nairobi as intended and thus was a viable way of directly alleviating the housing need of the squatters. The absence of moves from the squatter settlements is however not a sufficient condition to argue the lack of benefit to the poor. The project could have been beneficial to low-income families indirectly through the filtering process. The next three cases represent other situations where the site and service projects could have benefited the urban poor.
Case 2 The second case represents another direct move. This involves households who moved directly from rural areas. The numbers of migrants are significant given the high rate (7.5 percent) of migration to Nairobi. Most of the housing in rural areas is regarded as substandard because it 1? ks basic services and is usually constructed with temporary building materials. The previous housing conditions of these households are assessed and compared to that in Dandora to determine if these projects improved the housing situation of the new migrants. Most of the new migrants are young and have no housing in the rural areas. These projects would therefore have provided housing for them. Their entry point to the City, for the most part, is through squatter settlements or crowding in with relatives or friends. The provision of site and service projects would have prevented this situation for some and thus improved their housing situation. This confirms
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
8
that although the households may not have directly moved from squatter settlements, the project benefited them.
Case 3 In case three, the households moved directly from standard housing to Dandora, and through the filtering process made housing available to another household. Due to the acute housing shortage in Nairobi, much of the existing low-income housing is occupied by middle-income families. Their move to Dandora could have made housing available to low-income families as well as other middle-income families. Where lower income families moved to the vacated residence, it indicated that site and service projects indirectly benefited low-income households. In addition, it has been argued (Ward 1984) that site and service projects are too expensive for low-income families. If this is the case, the moving of a higher income group to site and service projects would not only have made housing available, but would have both provided the required finances for construction and increased the housing stock. Thus, through the filtering process, housing outside of site and service projects can be made available and affordable to low-income people.
Case 4 Case four represents a situation where the households moved from squatter settlements to minimum standard housing1 prior to the Dandora site and service project. Where
1 See definition on page 12
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
9
this was the case, it is assessed whether squatters indirectly benefited from site and service projects. The study attempts to determine whether the housing situation of the families that moved improved. If housing improved, the move to a site and service project could have been beneficial to the low- income households by improving the housing situation of the family that moved.
Improvement in the households' housing condition is assessed on the basis of whether the housing unit was constructed using either permanent or temporary building materials. Other aspects included are the level of services and the level of crowding. The move may have also made a standard dwelling available to another famixy, thus providing housing indirectly to low-income households. Hence, although the beneficiary may not have directly moved from the squatter settlement, Dandora provided them with better housing than they had before.
To assess the impact of the Dandora project on low-income households, the study answers the following questions: 1. Did the owners and tenants directly or indirectly
move from the squatter settlement or other substandard housing? a) What were their housing conditions in previous locations in Nairobi? b) What was their housing situation just prior to the move to Dandora? c) Has their housing situation and the level of available services improved over time? d) Did their move make housing available for another household?
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1 0
2. Are site and service schemes affordable by the low- income groups? a) What were the total housing costs before moving to the scheme? b) What are their total housing costs since the move? c) What was the total construction cost? d) What was the method of construction? e) What role did the bui'* ding groups play ir: housing construction? The analysis of both these questions and the four cases
provide information on the real beneficiaries of the project and whether site and service projects improved their housing condition. The housing index developed indicates what aspects of housing improvements resulted from the project. This study also evaluates whether construction cost was reduced or not.
1.3 Operational Definitions
One of the problems involved in housing research is the lack of commonly accepted operational definitions. This is especially true in Third World cities where the same words mean different things in different cities. Commonly used terminology such as "low-income housing," "slum" and "squatter areas" mean different things to different communities. Listed below are terms used in this study and their meanings.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Squatter Settlements
Nairobi's squatter areas can be divided into four types (Kayongo-Male 1988, 135). These include semi-permanent rural, semi-permanent urban, temporary urban and mixtures of temporary urban. Buildings in squatter settlements include a wide mixture of styles and quality. Many of these areas were originally villages on the outskirts of the City that were incorporated when the boundaries were expanded. There is a mixture of ownership patterns. In "squatter area," as many as one-third of the residents may own plots and are not really squatters. About 10 percent of the plots are often owned by absentee landlords. There are also parcels of land owned by squatters who form companies and built barrack-style housing. The rest of the plots are owned by the Central Governments and the City Council (Kayongo-Male, 137).
According to the above characteristics, squatter settlements refer to unregulated residential areas. S«'me of these are built illegally while others are legally built but were initially outside the boundaries of the City before extension. They are regarded by the City Councils as substandard because the owners did not follow the housing standards laid down by the municipality. Most of these settlements are characterized by overcrowding and a lack of basic services. Construction is done mainly with temporary materials. The term "slum" and "squatter" settlements are used interchangeably in this study.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Site and Service Projects
Site and service projects involve servicing urban land and making it available in small plots so that individual households can build their own dwellings. In these projects, lots are levelled and incorporated with access roads, drainage, water, sewerage, electricity, health clinics, refuse collection, fire protection and other services (Grimes 1976, 18) .
Squatter Upgrading
Squatter upgrading involves the provision of basic services to existing squatter settlements and improving them over time. Upgrading of squatter settlements has not been successful in Nairobi mainly due to mixed land ownership patterns which cause delays in the land acquisition process. Upgrading processes depend largely on community participation for housing improvement. Services are provided through loans given to the owners of the plots. The high rate of absentee landlords makes it difficult to mobilise community participation. Identifying who the owners are is also a difficult process. Once they are identified and the services provided, they increase their rents. The tenants are thereby displaced by a higher income group who can afford to pay the increased rents. Since the funds for these projects are borrowed, the delays associated with this whole process makes
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
it unfavourable to developers. There is also a lack of policy commitment to these kinds of projects because of the negative attitudes towards squatter settlement. Thus site and service projects have been preferred where funds have been available.
Overcrowding
Overcrowding is a housing market situation characterized by too many persons living in too few dwelling units. This was defined by the residents of Dandora as a situation where more than one person lived in one room. Most respondents would have preferred to have a separate room to co and do all their other household functions but the acute lack of housing prevents this from being feasible in the near future. "Overcrowding" in this study will include those with more than one person per room.
Low-income Households
Low- income refers to those households earning between 280-1,000 Kenya Shillings per month ( 1 U.S. $ = 21 Kenya Shillings, January 1990 exchange rate). This was the target income group for the Dandora project. Households income distribution estimate show that the lowest 40th percentile of Nairobi's population were within this income bracket in 1985 (Chana 1984, 18). This is assumed to have remained the same due to the slowing down of the economic growth of the country.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Due to the low incomes these households continue to live in the squatter settlements since this is the only housing that they can afford.
Minimum Housing Standard
This has been specified as a permanent house having two habitable rooms, a kitchen, a shower, and a toilet. This study however uses the word "standard housing" to mean any housing that was constructed with permanent materials and was approved by the Nairobi City Council. This includes one- roomed units constructed before independence. These do not meet the current minimum housing standard but are regarded as standard housing. Although the one-roomed unit policy was abolished after independence, most households continue to rent rooms in houses or apartments because they can not afford the whole unit. Thus in practice, most households continue to rent rooms. This is especially true in site and service projects where most of the households sublet one room.
Household
This term refers to a group of people living together. The household head is the one who paid the rent or the one to whom the plot was allocated.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Formal and Informal Sectors
Residents working in the formal sector include those employed in the wage sector by either the public or private sector. Those in the informal sector were self-employed.
1.4 Organisation of the Study
The dissertation is organized in seven chapters; the rest of this chapter provides information on the Dandora Community Development Project. It reviews historical development and the objectives of the project.
Chapter 2 provides a detailed literature examination of urbanization and housing problems in the Third World and in particular, Kenya. This is followed by an examination of housing policies and the historical development of the housing problems in Nairobi. Also discussed in this chapter are the housing programmes implemented by both public and semi-public housing agencies. Literature on the filtering process, which is the theoretical framework of this study, is also reviewed.
Chapter 3 provides information on the data and the research methodology. A detailed description of the sampling procedure, questionnaire objectives and method of data analysis follows.
Chapter 4 is the first of the chapters presenting the research findings. It provides information on the socio
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
economic characteristics of the respondents. These are discussed to assess the characteristics of the beneficiaries of these projects. Also discussed in detail are the housing conditions of the residents before and after moving to the Dandora project. The housing conditions in these locations are compared to assess whether the housing improved as a result of the move and in what aspects.
Chapter 5 provides information on the respondents who moved directly to Dandora from the squatter settlements or the rural areas. The issues addressed include the socio economic characteristics and the housing conditions before and after the move to Dandora.
Chapter 6 discusses the households who moved from standard housing to Dandora. The socio-economic and housing conditions are discussed to assess whether the move improved the residents housing conditions. Also discussed in this chapter is whether the respondents left housing behind that could have benefited low-income households.
Chapter 7 is the summary and conclusion and also contains policy recommendations.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1.5 Description of the Dandora Community Project
1.5.1 Historical Development of the Site and Service Projects in Kenya
The idea of site and service projects in Kenya dates back to the early sixties when mass urban migration began. The schemes were mainly intended for the resettlement of squatters who had lost their land due to the demolition of squatter settlements. The concept of site and service was however, adopted as Governments' policy of meeting housing needs of the low-income families in Kenya in the 1970-74 Development plan which stated:
These schemes will be a significant part of the housing programmes in urban areas. If no alternatives are available, the lowest income families will build the temporary houses they need anyway, as witnessed by the existence of large expanding illegal squatter areas near the urban centres. It will be the responsibility of the local authorities and National Housing Corporation to ensure that this activity is channelled into proper self- help schemes on serviced sites, lest the task of removing them or providing them with services later on become a grave problem. The degree of servicing such sites will vary from the mere laying out of the sites to houses and spaces for communal facilities to fully developed services systems commensurate with the type of community envisaged (Republic of Kenya, 519).
The small scale projects started in the sixties had a variety of standards, and in some cases, a roof plan was provided to the plot holders to assist them in building their house. Many problems were experienced during the
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
implementation of these projects and plot holders' satisfaction varied from project to project.
As a result of past experiences, and faced with no other alternatives, the concept of site and service projects received increased recognition as a realistic approach for meeting the residential needs of the greater portion of the urban population. A survey done in 1975 by the Nairobi City Council revealed a lack of funds for housing construction, a high percentage of absentee landlords, a lack of administrative procedure, slow development of plots, criticism of house design and delays in cadastral surveys. The housing construction standards were also found to be too high for a majority of urban dwellers.
The National Housing Corporation, in response to these findings, changed the site and service concept by limiting the contract work for these projects to the provision of infrastructure and services only. Plot holders were responsible for the construction of their own houses, with the aid of loans for materials. The projects were under the supervision of the technical staff of the Corporation, which also established a department to assist local authorities. The first of these large scale projects to be implemented was in Nairobi in 1975. This was the Dandora Community Development project which is the focus of this study.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1.5.2 The Dandora Community Development Project
The history of this project dates back to 1970 when the Nairobi City Council set up the Nairobi Urban Study Group to establish and formulate an urban metropolitan development strategy. This study group recommended, among other things, that there was an urgent need for about 20,000 affordable housing units for the lower income households (earning between Kshs. 200 and Kshs. 800 per month). Since the World Bank was willing to finance such housing projects, the Nairobi City Council formed the Dandora Community Development Department in 1975. The primary goal of this department was to implement the Dandora project in Nairobi as part of a loan agreement between the International Bank for Reconstruction and Development (IBRD), and the International Development Assistance (IDA) of the World Bank and the Kenya Government.
At this time, the national housing policy was developing interest in the site and service approach as a viable strategy to meet the housing need of the low-income earner. With the availability of funds, the Nairobi City Council and the commissioner of lands made available 350 hectares of land in the eastern part of Nairobi for development of 6,000 plots and related community and commercial facilities and services. This project was conceived as a pilot project so that the government and the local authorities could evaluate the potential of such projects for other urban centres in Kenya.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
20
Preparation work for the project started in January 1973 and continued until May 1974. The project appraisal report was prepared between June 1974 and May 1975. The Grade II By-laws (see appendix 1) of the building code were adopted for the project area. The plans for phase 1 were prepared in accordance with the Grade II By-laws requirements. In some cases, however, the standards used were higher than the ones stipulated by these by-laws. For example, the provision for the sewerage was at a higher standard and was based upon a waterborne system rather than pit latrines. Actual project implementation began in 1975.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1.5.3 Goals and Objectives of the Dandora Project
The goals and objectives of the project were specified as follows: a. To provide access to land and security of tenure on
a long-term basis primarily for residential use, with supporting community facilities including schools and clinics.
b. To control speculation and profit-making at the expense of the low-income sector.
c. To stimulate employment opportunity and industrial activities in the organization of local residents' associations and for credit, purchasing, equipment, training of special skills, management, legal assistance and marketing outlets.
d. To provide a framework within which residents can develop their own associations to administer the development of housing units and utility networks.
e. To provide communications and utilities channels which will stimulate transportation routes, and investment in residential, industrial and commercial activities, both within and near new communities (Njihia 1982, 88).
The four major housing components were specified as follows: a. All services other than those pertaining to the
preparation of the plots will be provided by the Nairobi City Council as part of its regular responsibility and will not be charged to the project. This means that the cost of health facilities, circumferential roads (other than those needed for the direct implementation of the project), educational and social facilities will be borne by the Nairobi City Council budget and only partly covered by charges and taxes paid by the inhabitants of the plots;
b. The built form, while circumscribed by the nature of the materials and equipment provided, will be left to the discretion of the builder-lessee subject to regulations pertaining to safety and sanitation. Design and production assistance will be necessary to insure sound investment by the individual;
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
22
c. The financial institution will be a quasi-private body with powers of eviction, transfer of title, loan moratoria and renewal. The financial institution will prepare building materials and equipment necessary to construct the housing unit. It will then make loans-in-kind in the form of materials and possible equipment (at interest rate calculated to cover its administrative expenses) to the builder-lessee. These loans will be repaid over a period of up to 25 years. The financial institution will also act as agent for the Nairobi City Council for purposes of collecting the plot rent. This will facilitate payment by the builder- lessee and reduction of administrative expenses;
d. The tied loans will be granted in amounts sufficient to enable the construction of a two room dwelling on the plot. The builder-lessee will construct the housing unit, perhaps in cooperation with other such individual local residents' associations. At least one of the rooms built will be occupied by the builder-lessee while one of the rooms could be rented to another qualified occupant. The responsibilities for payment of the loan and for payment for water, sewerage and refuse disposal and other chargeable services will rest on builder- lessee (Njihia 1982, 90). The Dandora Community Development Department(D.C.D.D.)
of the Nairobi City Council was established in 1975 and had the following functions:
a. To prepare and service 6,000 residential plots of 100 to 160 sq. meters, with individual water and sewerage connections, access roads, security/street lighting, and refuse collection services in the Dandora project site.
b. To construct the following wet cores and demonstration houses for the serviced plots: i. Options A 3,870 plots with wet cores (toilets and shower) on plot sizes of 100, 120 and 140 sq. meters. ii. Option B 1,800 plot with wet cores and one kitchen and store on plot sizes 100, 120 and 140 sq. meters. iii. Option C 300 plots with wet cores, kitchen, store and one room on 160 sq. meter plots and 30
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
23
demonstration houses to illustrate housing for option A and B plots. The option C plots were sold on market prices.
c. To operate and administer a materials loan fund amounting to Kenya pounds 1.5 million, to enable plot tenants for options A and B to borrow appropriate amounts for building materials required to expand such plots to have two rooms through self- help or contracting.
d. To construct community facilities including six primary schools, two health centres with day care facilities, one sports complex and 400 market stalls.
e. To construct trunk access roads to the project site.
f. To insure impartiality in the selection of prospective plot tenant, who must meet at least the following eligibility requirements: i. The total income at the time of application of the tenant and such members of his family as will live with him on his plot is between Kshs. 280 and Kshs. 500 per month for option A plots and Kshs. 450 and Kshs. 650 per month for option B plots. ii. The prospective tenant has lived in Nairobi for at least two years immediately prior to application for a plot and does not own any residential property in Nairobi. iii. The tenant's family (spouse, if any, or children) does at the time of application, and will upon allocation of a plot, reside with the tenant. iv. Prospective tenants will pay Nairobi City Council the appropriate fees for sewerage and water connection and a deposit of Kshs. 400 within sixty days of notification that they have been allocated a plot (Wanjohi 1977, 5).
To achieve the functions outlined above, the Dandora Community Development Department (D.C.D.D.) was organized into four major sections, consisting of the managerial section, technical section, the financial section and the community section. The functions of each of these are outlined below:
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
24
Managerial Section
This section was in charge of the overall activities of the D.C.D.D. and was to ensure that other sections conducted their specified tasks. It consisted of the Project Manager, Deputy Manager, and the Project Attorney.
Technical Section
The section was to "a. Supervise detailed planning, engineering, and
preparation of tender documents for site infrastructure, wet cores and community facilities;
b. Provide technical staff with specific building skill on site to show allottees how to perform technical skills, and
c. Illustrate the techniques of housing construction by erecting demonstration units on the site.
Finance Section
This section had the following tasks: a. Keep all project accounts involving expenditures
related to the project; b. Develop an accounting and management system; c. Prepare quarterly financial reports, and annual
project accounts audited by an independent auditor; and
d. Operate and administer the materials loan fund.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2 5
The Community Development Section
This section had the following tasks: a. Publicize the project; b. Solicit and process applications for the residential
plots? c. Orient and train allottees prior to the occupation
of the plots; d. Work with families during the construction phase;
and e. Assist residents in developing institutions and
programmes to enable them to create a genuine community (Wanjohi 1977, 6-7).
1.5.4 Implementation of the Project
The actual implementation of the project was carried out in two phases: phase 1, which consisted of residential area 1, and phase 2, which consisted of residential areas 2, 3, 4, 5, and a central spine of community facilities to be originally developed over a period of about four years (see Figure 2).
Construction of phase I started in October 1975 after obtaining approval from the Dandora Community Development Project Committee. However, in May 1976, the comments received from the Nairobi City Council did not agree with these plans and the City Council argued that certain Grade I building by-laws had to be maintained, although being a low- income area, Grade II Building By-laws should have been
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
MA P
OF
TH E
DA ND
O RA
P
R O
JE C
T
8
i.-.
v .v .v :-
>vW\ i f t f i 'f '- i
ty & y & y ■*y/y;yA
avoa oniu aaino > w b * o iOUJ f - o
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
27
followed. This concerned the designed standards for sewers, water supply, roads and ventilation. These plans were finally accepted and the first 1,000 plots were ready for allocation to selected families in December 1976.
During the construction, the applications for the plots were processed and the plot allocation was made well in advance. Initially, the applicants for the plots were interviewed personally to ascertain whether the information contained in the application forms was true and correct. A computer was used for random allocation but in spite of this, there were a few people who were allocated the plots illegally. Allottees were prepared through meetings and lectures to undertake the task of construction of their own houses. The nature and intent of the technical assistance to be given was explained to them.
The delays caused in obtaining the acceptance of phase 1 planning and infrastructure layout plans, in turn, resulted in delays in issuing house type plans to the plot allottees, and delays in the construction of demonstration houses in phase 1. Finally, the house type plans were accepted on the basis that it was too late to change them. This occurred in January 1977, while plots had been allocated in November 1976.
Phase 2 plans were based upon the same planning design standard but included many changes. Phase II consisted of 4,971 residential serviced plots, 3,180 option A, 1,536 option B and 300 option C.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
28
Residential Plots
The project provides 6,000 residential plots, each with individual water and sewer connection, and related basic services and infrastructure, including roads, security lighting and refuse collection. The plots vary in size from 100 square meters to 160, and are leased for a period of 50 years. The gross density of the project is 32 plots per hectare and the net density is approximately 45 plots per hectare. The plot occupancy rate is assumed to be 10 people pei plot, creating a gross living residential population density of 320 people per hectare.
At the time of drawing the plans, the total development cost of the plots was estimated as follows:
Option A between Kshs. 11,000-12,000 Option B between Kshs. 12,000-13,000 Option C about Kshs. 16,500
Financial Component
The Dandora project, estimated at Kshs. 200 million. (U.S. $ 30 million), is a joint venture of the Kenya Government (US$ 14 million.), the International Bank for Reconstruction and Development ($8 million), and the International Development Assistance (IDA) ($8 million).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
29
Loans to Allottees
The Dandora Community Development Department (D.C.D.D.) section operated a materials scheme. Cash loans were made for the value of the materials in that portion of the house already constructed. The rate of interest for loans (materials and plot) was 8.5 percent. Type A plots have a loan repayment period of 30 years and type B plots 20 years. All loans were repayable in equal monthly instalments (i.e. by way of annuity), consisting of principal and interest. Loan repayments were due at the end of each month, commencing the month after the plot was made available for possession. Repayment of the materials loan began 18 months after the date the first instalment of the plot loan is due. The interest rate accrued during the construction period on any material loan borrowed was capitalized, and the total interest was repayable by type B plot owners over eighteen and a half years. Type A plot-holders had five years grace period on the principal only. After 18 months they paid interest only for three and half years on any material loans borrowed and on which interest is accrued. On the expiry date of the three and a half year grace period, payment for interest commences and continues for 25 years.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3 0
1.5.5 Conditions of Lease Agreement
The lease specifies four conditions. The plot holder shall: a. Undertake construction of dwelling according to the
minimum standards laid down by the D.C.D.D., within 18 months of signing the lease. The materials used for the construction is either to be one's own or obtained under the materials loan scheme.
b. Pay all charges promptly and, in general, conform to agreements.
c. Sublet rooms only on conditions specified by the project administration and make the identity of tenant known to it prior to subletting.
d. Notify the administration of intention to leave the project and conform to the project rules and regulations regarding the transfer (Njihia 1982, 103-104).
1.5.6 House Type Plan Options
The project provided a range of alternative plot layouts, thus giving the allottees the choice of plan type (see Figure 3 and 4). On average, five rooms were to be constructed on each plot.
Each stage of house construction was monitored and a progress report maintained by a building inspector. The main stages which were monitored and inspected were the setting-up of rooms, foundation trenches and walls, ground floor slab, external and internal walls, wall plates, roof structure and finishings, and fittings. The intended purpose of monitoring and inspecting the house construction was: to ensure that the
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3 1
Figure 3
HOUSE TYPE DESIGN OPTIONS
6.3 m Eini"-in
coco
P
si
7.35m
Oi
Plot area 100 m2 “1
2
— i — 4 ir
H —I _L_
4 rooms 3 rooms 4 rooms 3 rooms 3 rooms
6.3 m r 3H 5 rooms 4 rooms 3 rooms 4 rooms
6.3 m
~\r "I*"! J L .
-I -I
_L5 rooms Upper floor 4 rooms 4 rooms 4 rooms
5 * _j —ir- JL
5 rooms 6 rooms
X JL
J l_
Comer plots areas: 180-240m
6 rooms ♦
Source: Ghana 1984, 26
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
R eproduced
with perm
ission of
the copyright
ow ner.
F urther
reproduction prohibited
w ithout
perm issio
n .
Figure 4
TYPICAL HOUSE TYPE PLANS FOR DANDORA PROJECT (Four Room Unit)
Pre-built core
ROOM 1
12.3 M2
_T\_ ROOM 2
Yard WC
10.2 M
VERANDAH
VERANDAH
8.3 M Yard
Source: Chana 1984,26
Figure 3
0 1 2 3 4 1 I I I I
metres
3 3
construction was based upon the building specifications; to meet the general standards of building and planning practice; to ensure the allottee sought material loans in the stage that D.C.D.D. specified, i.e. at completion of each stage; and lastly, to ensure that the MminimumN number of rooms were completed within the 18 month period.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER 2
LITERATURE REVIEW
2.1 introduction
The inability of Third World governments to meet the housing need of the majority of urban dwellers led to the implementation of site and service projects in many cities in the 1970s. To understand what general problems the projects were supposed to solve, a brief review of the urban housing problem and the response in the Third World is included. This review is followed by literature on the historical development of the filtering process and its application to the Nairobi case. The filtering process is central to this study, because it is one of the ways in which the urban households could have benefited from the development of site and service projects.
A detailed review of the urbanization process in Kenya and the national housing policy is included. This enables one to see the total picture in the urban areas. The site and service projects were not only supposed to benefit Nairobi residents, but also the other towns in the country. The discussion on this issue as well, shows the magnitude of the housing problem in the rapidly growing towns of Kenya. The housing policies made in the past have mainly been directed to Nairobi, which experiences an acute housing shortage. This
3 4
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
35
chapter discusses the general housing policy and the historical development of housing in Nairobi. Thus, a detailed discussion of the historical developm vt of the housing policy of the City is included. The historical development is important because it shaped the current housing policy in Kenya. This is followed by a discussion of the different public and semi-public organizations involved in housing development. The type and number of housing units developed greatly influence who the beneficiaries of the site and service projects are.
2.2 Third World Urban Housing
Cities of the Developing Countries have grown rapidly in the last three decades. This rapid urbanization process has not been accompanied by the necessary growth of housing and related services. The lack of these services is more noticeable among the low-income households who have housed themselves in squatter settlements. These settlements are very common in Third World cities and are said to characterize the urban landscapes of these cities. The settlements are also a direct manifestation of the high rate of urbanization in these cities. For example, "in the past 40 years Third World urban populations have grown at between 3 percent and 5 percent annually, from a total of under 300 million people to 1.3 billion people. In the 1970s alone, the number
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
36
increased by nearly 300 million, currently Third World cities are receiving 45 million new inhabitants each year, compared to 7 million in developed countries" (Harris 1989, 175). By the year 2000, some 45 percent of the people in developing countries will live in cities or towns (Linn 1983, 3).
The number of very large cities is also increasing rapidly. In 1950, only one city in the Developing Countries had more than five million inhabitants. By the end of this century, some 40 cities are expected to be at or above this size. This rapid urbanizetion process has been accompanied by the explosive growth of large cities. For example, in the mid 1970s, Mexico City and Sao Paulo each grew by over a half a million people annually, while cities such as Jakarta and Seoul grew by over a quarter of a million people each year. Small and medium sized cities are also growing rapidly (Linn 1983, 3).
This growth varies from one country to another, with the Latin American countries being the most urbanized and the African countries least urbanized. For example, in Latin America, Mexico city and Sao Paulo are already among the most populous cities in the world and it is predicted that they will swell to populations of well in excess of 20 million by the end of the century. Asia has more uniform growth with an urban population of 25 percent (kept at this low level by the huge populations in India and China) . Africa is much less urbanized, with some countries having less than 10 percent of their population living in urban areas (Drakakis-Smith 1987,
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3?
4) . However, African urban areas, particulary primary cities, now display the world's fastest rates of population increase. In addition, this growth will continue to be the most rapid in the world. The degree of urbanization and the rate of urban population in Africa varies considerably from country to country and from one region to another. The southern region of Africa has the highest level of urbanization (46.5 percent), the northern region has the longest tradition of urbanization (44.3 percent), the western and the central regions have the longest tradition of African urbanization (22.5 percent and 34.5 percent), and the eastern region is the least urbanized (16.4 percent) (Obudho 1988, 6).
The rapid urbanization that is taking place in these cities has not been accompanied by growth in other sectors, and has therefore brought numerous problems. Lack of adequate shelter is one of the most visible problems faced by Thiia World cities. This lack of adequate housing has been brought about by the existing deficit and the inability of most city dwellers to pay for decent housing. This is especially true of low-income households who have met their housing need by living in substandard housing in squatter settlements. As Linn states, "sprawling shanty towns, slums, or squatter areas; dilapidated shelter structures patched together from scraps of cardboard, corrugated iron, wood, and sometimes bricks; an appalling lack of basic services such as safe water supply, sanitation, and drainage; all these problems are glaringly obvious to anyone living in or visiting the cities
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
38
of developing countries" (124). The following examples illustrate the case: In 1977, out of the eight million inhabitants of metropolitan Calcutta, 5.3 million (67 percent) were living in squatter and slum communities. About 75 percent of Ibadans and 60 percent of Bajotas were living in similar conditions (Qadeer 1983, 2). Other cities with high percentages of slum and squatter dwellers include Addis Ababa (90 percent), Accra (53 percent), Abidjan (60 percent), Lome (74 percent), Mogadishu (77 percent), Dakar (60 percent) and Ibadan (75 percent) . In Asia the highest is in Colombo (SriLanka), with 43 percent, followed by Kuala Lumpur (Malaysia) with 37 percent, Manilla 35 percent, and Seoul 31 percent. In Latin America, Bogota (Colombia) has 60 percent, Guayaquil (Equador) 49 percent, Lima 40 percent, Caracas (Venezuela) 40 percent and Mexico City 46 percent. The above statistics reveal that the percentage of people living in slums and squatter settlements is highest in Africa (Linn 1983, p.12-13).
Most Developing Countries' response to housing shortages in the 1960s was characterized by policies which assumed that only the public sector could provide housing for the low- income groups. This led to policy objectives which were geared towards providing safe and decent, and sanitary housing for all people. These objectives were seen as supporting public sector shelter consisting of walk-up flats. These were built to high physical standards and required high monthly subsidies per unit. In reality, comparatively little such
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
39
shelter was ever built compared to the total needs of low- income groups (Van Huyck 1987, 340). Not only were there few housing units, but they could not be afforded by the majority of urban households due to their low incomes. Studies done by Grimes in the 1970s revealed that the cheapest new housing currently being built was still not affordable by the poor. Under reasonable repayment terms and at interest rates of 10 percent, one-third to two-thirds of urban families could not afford the cheapest new housing in the cities studied (Grimes 1976, 9). The total percentage of urban households unable to afford the cheapest new housing unit accounts for 68 percent of households in Nairobi, 90 percent in Cairo, and 50 percent in Rabat, Morocco, 55 percent in Mexico City and 47 percent in Bogota (Grimes 1976, 69). In Indonesia, conventional site and service programs and low-income housing were beyond the ability of the poorest 30-4 0 percent of the urban population to be served, despite subsidy estimated to amount to 65 percent of the total cost. Since the poor could not afford publicly constructed housing of this type, the housing units almost invariably were purchased or leased by high-income groups (Linn 1983, 138). Thus very few poor households obtained accommodation in public housing projects. The number of such units built annually fell far below needs in virtually all nations. Ambitious targets were set but rarely met. For instance, in Kenya, the 1979-83 Development plan admitted that "over the last plan period, only 8 percent of the low cost units planned were in fact completed and these
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
4 0
cost an average of five times the expected cost" (Republic of Kenya 1979, 50). Also, the criteria for the allocation of the few that were built often excluded the poorer households, for they lacked the proof of regular income which was demanded by eligibility criteria, or did not belong to officially recognized organizations or trade unions, whose members got first priority. The situation has become worse since then and the projections for most of these cities show a grim picture. It has been argued that these population increases have overwhelmed even the most extensive programmes for housing construction and improvement of social services, particularly in the rapidly growing major centres (Bourne 1976, 244).
Apart from public housing construction, the 1960s were characterized by a negative attitude towards squatter settlements. These settlements were viewed by the government as below national standards and therefore, should be cleared where they existed, and new settlements built below national standards, prohibited. This policy led to the tearing down of 'illegal* housing stock and harassment of low-income people who sought to shelter themselves at standards they could afford (Van Huyck 1983, 340). No housing was constructed to replace the units which were demolished. The owners moved to other parts of the city where they were not likely to be evicted, but they moved to the same kind of housing since that was the only housing they could afford.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
The failure of the conventional s.ielter policies became widely recognized in the 1970s. International donor agencies drawing from experiences in Puerto Rico, Peru, and India, among others began to support new approaches for housing low- income groups. These initiatives, called site and service projects, focused investment on the provision of serviced sites and minimum structures provided if any. These projects are defined by the World Bank as the subdivision of urban land, and its servicing with varying combinations and levels of public utilities and community facilities for residential and commercial use (World Bank 1974, 3). Generally, the objective of site and service schemes was to provide an economically accessible physical framework to a specific target low-income population for their shelter and related employment needs.
The necessary aid given by local, regional, national, and international institutions can be in the form of infrastructure, sanitary core, roof plans, supporting structures, standard designs and specifications, demonstration prototypes, instructions, supervision, fencing, and supply of certain materials and/or plant tools. Technical aspect variables would include such items as land acquisition and tenure requirements; credit arrangements; technical assistance; cooperatives for the purchase of materials, service and tools; building codes; standards and restrictions; the mix of individual, mutual and subcontracted labour and utilization of local leadership, voluntary associations and
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
42
informal networks (UNCHS Habitat 1982, 180). Site and service projects differ from conventional housing in that families who move into site and service neighbourhoods receive only incomplete housing, which means a degree of gradual construction of infrastructure or houses, or both, with a significant amount of total investment left until after occupancy. Also, they invest in their own housing, rather than pay for something decided on and provided by someone else. With self-help, the direct investment by families can take the form of either labour from family members and friends, or purchased materials and hired labour, or a combination of both (Skinner and Rodell 1983, 22).
The other major initiative was called "squatter upgrading." This approach recognized the basic worth of informal settlements regardless of their existing standards and focused investment on providing the essential minimum infrastructure (e.g. water, toilet facilities and access roads) which was missing. Frequently this upgrading process was accompanied by efforts to provide secure tenure. Settlements upgrading in some countries clearly proved that when informal neighbourhoods were provided with infrastructure and security of tenure, the people themselves would invest substantially in the improvement of their own shelter. International donor support for these kinds of projects increased dramatically throughout the 1970s. However, in many countries there was a continued resistance to adopt these techniques as national shelter policy. Often international
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
43
resources were invested in these kinds of projects, whereas domestic resources continued to be invested in other higher standard shelter projects (Van Huyck 1983, 342).
The success of these projects in meeting the housing need of the urban poor has been questioned. Grimes and Payne, among others, have argued that these projects have not benefited squatters in Third World cities as intended. Payne states that these projects benefit the upper and middle-income groups, "but isolate the poorest groups, still from suitable housing and employment opportunities" (Payne 1984, 5). The same point is stressed by Grimes, who argues, that due to market imperfections, "it is common to see site and service projects that are designed for low-income groups be occupied in the end by middle-income groups" (89). In general, due to this high cost, it was found that these projects did not address the realities of low-income groups whom the site and service projects were supposed to help (Ward 1984, 152). This study evaluates the beneficiaries of these projects to see whether the low income households benefited directly as intended or indirectly through the filtering process.
2.3 The Filtering Process
The term filtering has been used widely in studying the urban housing markets. Its definition has varied depending on the emphasis of researchers. The history of the concept
Reproduced with permission o f the copyright owner. Further reproduction prohibited w ithout permission.
goes back to the British special committee report on slum housing in 1929 which stated "when the post-war building began, it was hoped that there would be a gradual movement of the working-class population out of the slums into better houses. This might occur in two ways, either the slum dweller might go directly into the new house or a process of "filtering up" might occur under which the slum dweller would move from the slum into a better pre-war house" (Davies 1976, 140) . However, Hoyt is said to have given the first systematic definition to the filtering process, which he referred to as a process whereby all groups "move up a step leaving the oldest and cheapest houses to be occupied by the poorest families or to be vacated" (Hoyt 1939, 122). He stated that obsolescence of higher-priced dwellings and population growth were the major contributors to neighbourhood movement.
A comprehensive definition of the term was developed by Ratcliff who defined the process as "the changing of occupancy as the housing that is occupied by one income group becomes available to the next lower income group as a result of decline in market place" (Ratcliff 1945, 1). This definition could be used to describe the filtering down of dwelling units or the filtering up of families. The argument is that filtering down will provide housing for the poor, and also eliminate substandard housing by building enough units. "Thus the solution for substandard housing conditions is a rate of housing production which will release a sufficient number of
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
used dwellings to successively lower income groups so that the substandard housing may be retired and the families at the bottom of the income scale may have the opportunity to live in housing of acceptable quality at no higher rents than they paid for substandard housing" (Ratcliff 1945, 1).
Fisher and Winnick departed from this definition and argued tha* this definition brought confusion between downward price movement, which they regarded as the actual filtering, and change of occupancy which they viewed as the effect. They dropped Ratcliff's key element - change in occupancy. They saw filtering as a movement of dwelling units and the test for filtering as in rent and price not in income. They defined filtering as "a given dwelling unit or group of dwelling units within the distribution of housing prices and rents in the community as a whole" (Fisher and Winnick 1951, 47). Grigsby argues that Fisher-Winnick's definition has the advantage of avoiding the problem of the changing price level. Their definition does not address the argument "whether filtering works or not; i.e. supplies lower-income groups with decent housing, for it is now only a statistical measurement incapable of illuminating this question" (Grigsby 1965, 91). Lowry later developed another definition which he used to examine whether filtering can provide adequate housing for lower-income groups. He defines it as "a change in the real value (price in constant dollars) of an existing dwelling unit" (Lowry 1960, 362). He stated that "the effectiveness of filtering as a means of raising housing standards thus
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
hinges on the speed of value decline relative to quality decline" Lowry's definition differs from Ratcliff's in its exclusion of the notion of changing occupancy and from Fisher- Winnick's by the standard against which the process is measured.
2.3.1 The Filtering Process in Nairobi's Urban Housing Market
Government involvement in the provision of housing to the poor can either be direct or indirect. The direct approach involves the construction or acquisition of housing units for occupancy for a specified needy group. The indirect approach consists of providing or stimulating the provision of housing for higher or middle income households in anticipation that the needy population will be accommodated in housing left vacant as these higher income households move to new housing. Although the term "filtering" has been used widely in studying urban housing markets in developed countries, considerable disagreement exists over its precise definition and measurement .
In the absence of price data, the study must return to Hoyt's original definitions who defined filtering as the "move up a step leaving the oldest and cheapest houses to be occupied by the poor families" (122). This move causes the real housing condition of the families involved to change. This definition has been chosen because it emphasizes the movement of poor people to better housing, and also includes
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
the improvement of housing. Thus, it addresses the two aspects to be investigated by this study. The price index of houses is a major part of other definitions but information on price index is not available in Nairobi.
Both the direct and indirect approaches discussed above have been adopted in Nairobi. As stated in the National Development Plan, "since independence the role of the government in housing had been to encourage the private sector to play an increasing role in building more houses and in assisting local authorities through the National Housing Corporation to enlarge public housing programs" (Republic of Kenya 1970, 505) . The efforts of both the private and public sector to provide housing could not keep up with the rapidly increasing population in Nairobi. They could not even meet the housing demand of middle and high income groups, on whose action the filtering process depends. Grigsby argues that dissatisfaction with housing must originate among this group or somehow be transmitted to them through the price mechanisms or by the actions of those who cannot afford new construction. However, for filtering to occur throughout the stock, the dissatisfaction must be sufficient to create an excess of accommodation for the entire economy.
The housing shortage in Nairobi did not allow whatever filtering took place to benefit low income households, "since the number of houses completed so far, were for incomes in the middle range, yet nearly 70 percent of demand for urban housing was by the lowest income groups" (Republic of Kenya
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
4 8
1970, 505) . The government, realizing this situation, decided to directly provide housing to low-income people through site and service programmes. In the 1970-74 Development plan they stated that "these schemes will be a significant part of the housing programme in the urban areas. If no alternatives are available, the lowest income families will build the temporary houses they need anyway, as witnessed by the existence of large expanding illegal squatter areas near the urban centres" (Republic of Kenya 1970, 519). These projects were meant to benefit low-income people who were living in squatter settlements. It was hoped that the effect of low-priced construction would be felt much sooner at the bottom since these projects were designed to benefit low-income people directly, or were positioned so that the impact could be transmitted a shorter distance through the vacancy chains (Grigsby 1963, 101). These projects were supposed to benefit the low-income people directly.
This investigation addresses the situation in Nairobi in order to determine whether these projects have either directly or indirectly, through the filtering process, benefited low income people is intended or whether they expanded the range of housing stock for the Nairobi middle class. This is done by using the migration history of the residents which indicates whether the beneficiaries (directly or indirectly) moved from squatter settlements and whether their housing situation has improved as a consequence of the site and service project.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2.4 Urbanisation in Kanya
Apart from limited areas, large-scale urbanization in Africa as a whole was initiated almost entirely as a result of colonial rule. In Kenya, before the construction of the Kenya-Uganda Railway, there were virtually no urban centres except along the coastal area. The human settlements were highly rural and of a dispersed nature. The introduction of railway routes led to the establishment of commercial and administrative centres along the routes, which in turn had a great effect in determining the number, size and distribution of urban areas today. Other colonial policies initiating commercial and administrative centres together with the strategies of provision of infrastructure and agricultural development in certain areas, also led to the growth of other urban centres which were not along the railway network.
Urbanization as a major element of Kenya's development is therefore comparatively recent. The number of urban centres (defined as population centres inhabited by 2,000 people or more) is relatively small in size compared to urban centres in many parts of the world. However, the growth of such towns during the last three decades has been rapid and follows general urbanization trends in other developing countries. Kenya has therefore been identified as one of the most rapidly urbanizing countries in the world in the sense of having a largely rural population, few existing urban centres and services, and a rapidly expanding rural and urban
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
50
population. It has been estimated that the urban population would expand from only 2 million to 8 million in the last 25 years of this century (Obudho 1987, 180). This rapid growth is clearly shown by the rate of increase of urban population. For example, the annual rate of growth of urban population during the 1962-1969 intercensal period was 7.2 percent compared to 3.4 percent of the national population. During the 1948 population census there were 17 towns with a population of 2,000 people or more having a total population of 276,240 or 5.1 percent of the national total. In 1962 the number of towns doubled and the urban population numbered 671,000. The 1969 population census illustrated further the rapid growth of the urban population to a total of about one million people distributed over 48 towns. Urban centres increased from 48 towns to 91 between 1969 and 1979, making the urban population increase from 10 percent of the total population in 1969 to 15.1 percent (2,309,000 people) in 1979 (see Table 2.1). This table shows a rapid increase of the urban population, especially in the bigger towns. For example, in 1948 Nairobi's population accounted for 41.6 percent of the total population. In 1962 this was 33.8 percent, in 1969 it was 47.2 percent while in 1979 the City had 35.9 percent of the total urban population (Obudho 1987, 39).
However, the provision of housing has not kept pace with the rapid increase of population. As a result, overcrowding and squatter settlement development has increased, especially
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
in the big urban centres. According to the 1983 housing survey, the estimated number of housing units provided between 1976-1982 was 44,600. This averages to 6,400 units per year. The greater part of this output comes from the National Housing Corporation (NHC), whose houses and site and service projects together account for 64.4 percent of the seven year total. The private sector contributed 25 percent, other public agencies (mainly local authorities) accounted for 6.4 percent and the Ministry of Works and Housing and Physical Planning a further 4.2 percent, chiefly for government employees (1983 Urban Housing Survey, 12) . The survey estimated the housing need between 1983 and 2000 to be as shown in Table 2.2. The survey stated that the percentage distribution of structures by status of permanency was 66.7 percent (33,452) permanent structures and 33.3 percent (16,692) of both semi-permanent and temporary structures.
2.5 National Housing Policy
The first phase of the colonization of Kenya was characterized by a government housing policy where housing was provided by the individual branches of government and private sector. But as the African settlement increased, the African employees from the rural areas were housed in "labour lines," which were grass thatched huts, clustered together in a location with no facilities at all. In the urban centres,
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 2.1 Centres by Size of Urban Population 1948-1979
Urban Population size 1948 Number of
1962 Urban centres
1969 1979
100,000 1 2 2 3 20,000-99,999 1 2 2 13 10,000-19,999 2 3 7 11 5,000-9,999 3 11 11 22 2,000-4,999 10 16 25 42 Total 17 34 47 42 Total Urban in '000 276 671 1080 2309 % of Total Population 5.1 -j • 00 9.9 15. 1
Source: Republic of
Estimated Housing
Kenya, 1986, p.11
Table 2.2 Need by Category 1983 -2000
Housing Need 1983-88 1989-90 1990-2000 New Population 70440 32310 187020 Depreciation 29395 10588 52790 Inadequate 17715 7086 35431 Total 117550 50484 275241
Source: Field Data
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
the provision of housing was the responsibility of the individual households until the 1939 strike in Mombasa and 1941 in Nairobi (Obudho 1987, 17). Until this time, housing for Africans had been planned on the assumption that the African labour force was and would remain one of single men. It was assumed that there would be no necessity to make provision for the worker and his family. This policy was reflected both in the low wage level and the provision of "bed space housing" (Obudho 1987, 18).
The housing problem had by this time become too big for the authorities to ignore so in the early 1940s, different measures were taken to alleviate the problem. One such measure was the creation of a Central Housing Board (CHB) in 1943. This organization was created to oversee the government loan terms. The housing fund was to be used for houses constructed by local authorities or as loans for individuals to build their own houses. The CHB wanted to encourage actual house ownership in addition to local authority rental housing. The CHB was very active in the 1950s in providing housing to consolidate tne Africans, to make them property owners. This marked the beginning of the provision of public rental housing in the major urban local authorities. Most of the accommodation provided was in the form of single rooms in a row of terraces with communal sanitary facilities. In contrast to the Africans, the standard of European and Asian housing improved simultaneously with urban development. The first houses were made of whatever materials were available,
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
but mainly timber. Little by little, housing for Europeans changed to solid stone structures with tiled roofs, showing a strong link with the Pritish upper class country house. Asians, on the other hand, brought their own traditions and building systems (Republic of Kenya 1986, 20).
When Kenya attained self-rule in 1962, it had an urban population of 671,000. The majority of these were Africans, but the rate of housing development had been so slow during the colonial period, that the housing sb^tage had become acute. The Government in 1965 invited United Nations experts to study the situation and make recommendations. The report noted that ?ike many other African cities, the Kenyan situation was characterized by "migration to cities from rural areas; the pressures of overcrowding and housing shortage; lack of capital with which to build houses and disparities between what the average urban family can pay for the shelter and the cost of producing it" (Bloomberg and Abrams 1965, 1). In addition, one-room dwellings predominated African housing comprising 70 percent in urban areas and 80 percent in Nairobi. Overcrowding was extreme, and only 49 percent had washrooms available to them. Separate kitchens for the African population were virtually non-existent and cooking was usually done in the rooms used for living (Bloomberg and Abrams 1965, 5).
Overcrowding was an exceedingly serious problem among the urban African households (see Table 2.3). Even at the extremely high ratio of 3-4 persons per room, 49 percent of
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 2.3 Level of Crowding in Urban Areas in 1965
Total Percent Nairobi Percent
Number of Rooms 2+ 3+ 2 + 3+ 1 75 54 78 57 2 61 35 60 34 3 47 22 49 20 4 38 16 33 10
Source: Bloomberg and Abrams 1965, 15
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
56
all urban households were overcrowded (52 percent in Nairobi) . Overcrowding was even higher in the one room units. Most of the people in the urban areas could not afford any type of housing, thus even if housing was provided for as little as 30 Kenya Shillings (Kshs.) (U.S. $ 3.5 in 1965) it could not be afforded by 50 percent of the households in Nairobi (Bloomberg and Abrams 1965, 24). Given the fact that most people could not afford to pay for any type of housing, the report recommended that government programmes must seek to develop schemes which would provide shelter at the lowest possible cost to the occupants. The report also pointed out that it would be necessary to provide larger housing units in the future because of the expected increase in household size in urban areas.
According to the report, Kenya was experiencing serious housing problems, particularly in the cities. The natural increase in population, accentuated by the migration into the cities, had put a burden upon the housing agencies which had outstripped their capacity to meet it. Simultaneously, low income and unemployment put a brake on the ability of people to afford better housing (Bloomberg and Abrams 1965, 3). The Mission recommended that the use of government funds should be directed towards serving the lower-income groups because this group was almost entirely removed from the benefits of privately constructed housing. The practical limitation on the cost of permanent housing made it improbable that the lower limit for this group could be substantially reduced without
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
direct subsidy. Since most families could not afford their own houses at the minimum costs, they would have to rely on minimum standard rental housing financed with long term low interest rate government loans. The report also recommended a rapid move towards owner housing for those segments of the population for whom ownership was in their best interest. According to the report, it was expected that with the high rate of migration to the City, overcrowding and squatter settlements would continue to grow. "Without the means to do otherwise, the only alternative is to attempt to control temporary housing and make it conform as far as possible to a plan by which it would eventually be replaced when families can afford it" (Bloomberg and Abrams 1965, 52) . The Mission believed that the best way to do this was through site and service projects.
The outcome of Bloomberg and Abrams' study was the first housing policy in the government publication known as "Sessional Paper no. 5." This paper set no specific policies on increasing the number of low-income dwellings, although this was the most critical problem at the time. Instead, the policy was to move away from providing "bedspace" accommodation. To deal with the problem of the slums and squatter settlements, the paper stated that "If towns are not to develop into slums, and centres of ill-health and of evil social conditions, low-income urban housing and slum clearance must continue to form the major part of the nations housing programme" (Sessional Paper no. 5 1966, 7). The policy was
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
58
to use a high percentage of development funds loaned to the NHC to finance rental housing for families who could not afford the cost of owning high or medium cost housing, since NCC could not meet the low-cost housing needs. In these circumstances, the mobilization of private capital for low- cost housing were to be encouraged and site and service projects were to be a significant part of the housing programme in urban areas.
Other documents that contain government policies since independence are the Development plans. The first Development plan for the 1964-70 development period had no priorities as far as housing was concerned. However, "the plan reaffirmed the government's housing policy by focusing attention on such issues as the problem of rural to urban migration and the inability of the majority of the urban population to secure housing of reasonable quality and quantity" (Chana T.S. et al, 20) . In spite of the inability of many to afford housing, the government discouraged any type of subsidy. Site and service schemes were seen as the practical approach to the problem of low-income groups in urban areas. During the 1964-70 development period, the role of the government in housing was to encourage the private sector to play an increasing role in building more houses, and in assisting local authorities through NHC to enlarge public housing programmes (Republic of Kenya 1970, 505). The private sector concentrated on middle and high income earner demand, whose housing demand over the years has outstripped supply. Thus, the private sector not
Reproduced with permission of the copyright owner. Further reproduction prohibited w ithout permission.
only did not get involved with low-income housing but could not meet the demand for the higher income groups. Since independence, the public sector had produced or helped to produce 9,500 housing units and sites on an average of 1,900 per year. It was stated that the number of houses completed so far were for incomes in the middle range "yet nearly 70 percent of the demand for urban housing is by the lowest income groups" (505). Although the private sector had increased their output since independence, it did not produce a large number of dwellings. To stimulate their activities, the government established Housing Finance Company of Kenya (HFCK) at the beginning of 1966. The objective of establishing HFCK was to make loans available to people wishing to acquire their own houses in the main centres.
The 1964-1970 development period witnessed a shortfall of 7,500 units per year (a total of 52,500 in seven years). This shortfall was met by individual families who had squatted on public and private land and built whatever poor shelter was within their means, usually fashioned of mud, wattle, cardboard and tin. This shortfall was an indication of how critical the housing shortage problem was becoming because the population was at the same time increasing. Another factor was that by 1963, 44 percent of the African households were estimated to be living in employer-provided housing. However, the Federation of Kenya Employers decided in 1970 to bring an end to that provision of workers' housing (Tribe 1972, 8).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
6 0
This was a disadvantage to the lower-income group since the employers' housing was no longer accessible to them.
Thus, in the 1960s the official Kenyan housing policy was directed towards legally acceptable formal housing which was only available and affordable by middle-income inhabitants (Amis 1983, 122). The other policy was the demolition of unauthorized settlements. The 1970s were characterized by changes on the minimum standard housing. Thus, according to the 1970-74 Development plan, "the prime objective of government policy in housing is to move towards a situation where every family in Kenya will live in a decent home whether privately built or state-sponsored which provides at least the basic standard of health, privacy and security " (Republic of Kenya 1970, 513). The plan noted that this kind of housing will be out of reach of the majority of the families because of their incomes. The plan fixed a maximum cost of 24,000 Kenya Shillings (Kshs.) (current U.S. exchange rate $1=18 Kshs) per unit for housing financed by government development funds. As shown later, the housing constructed using government funds cost much more than the stated amount. This was especially true of the projects implemented by the National Housing Corporation which is the government arm for implementing housing policy. This plan also stated that since the shortage of housing was so acute, the available finances would be used to expand the total housing stock rather than to clear slums or purchase existing dwellings. This was a significant change from the previous plan, which stated its
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
commitment to slum clearance. Although a number of slums were cleared in the 1970s, this demonstrated the beginning of approaches that recognized the existence of slum and squatter settlements. During the plan period, the estimated 50,000 urban units needed were not built. Instead, only about 25,000 units were completed.
The 1974-78 Development plan continued on the same lines as the 1970-74 Development plan. However, high priority was placed on the rapid improvement of housing standards whereby the desire was to streamline housing design and construction with government-determined standards. This was done by ensuring that each housing unit constructed in an urban setting would be expected to have a minimum of two habitable rooms, plus a kitchen, toilet and shower. The raising of the minimum housing standard affected the activities of the local authorities' housing production. This was especially true of the Nairobi City Council which could not afford to build housing of the minimum standard and charge the stipulated rents. The end result of this was a shift from rental housing to tenant-purchase housing. This not only reduced the supply of new housing, but was also out of reach of the majority of the urban dwellers. In addition, the plan advocated action against further spread of squatter settlement and slum clearance, through the process of resettlement and improvement of substandard urban housing. The plan estimated that approximately 110,000 units would be required in urban areas, during the plan period to house the increase in population,
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
6 2
plus over 50,000 units to meet the existing shortfall. The two plans following this one had no new policies, since the previous ones had not been fulfilled.
The above policy statements seem to run counter to the existing housing situation. The United Nations' study revealed a very acute housing shortage, which was characterized by the rapid development of squatter settlements. The report further pointed out the inability of the majority of the urban households to pay for any type of conventional housing. The report also showed the necessity of subsidy if these low-income people were going to get any type of standard housing. Housing units were to be increased at a very fast rate in order to deal with the existing deficit and the rapidly increasing new population. However, the development plan initially stressed demolition of slum and squatter areas. Although this was changed in the 1970s, it was replaced by high standards which the majority of the urban dwellers could not afford. Construction of standard units became much more expensive; this slowed down the number of housing units produced. Meanwhile, the urban populations were increasing rapidly and providing substandard housing for themselves in the squatter settlements. Most of the above discussed policies have been directed to Nairobi, the biggest city in the country. Its size and rapid growth have made the nousing situation more critical in this city than elsewhere in the country. The different policies made, affected the different housing programmes implemented and these in turn
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
have served different incone groups in the City and greatly influenced the urban housing market. The following section discusses urban housing policy implementation in Nairobi before and after political independence.
2.6 Evolution of Housing Policy in Nairobi
The current estimated population for Nairobi is 1.5 million people. The City is currently growing at a rate of 7.5 percent per annum and is estimated to have a population of 3 million by the year 2000. The current population is predominantly youthful with 50 percent of the male population being below the age of 24 years, 75 percent below the age of 35 years and 90 percent are aged 44 years or less. The f male population has 56 percent of the population below 19 years of age, 80 percent under 29 years and 90 percent under 39 years (Obudho 1987, 43).
In the past ten years, there has been an average of 14,800 additional households in the City and its outskirts each year. The major part of this growth is from migration into the City from other areas. Most of the migrants are poor and unskilled and many cannot find formal employment. Thus, while they continue to place additional burdens on urban services, they are unable to afford the type of housing provided and contribute little to the City's revenues. In the period since 1973, the rate of economic growth in Kenya has
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
64
slowed down drastically but a high rate of migration has continued. As a result, per capita incomes in the City have fallen significantly. According to Amis, "even if the economic growth recovered to its former high levels the continuing high rate of population migration would result in virtually no increase between 1975-85 and the real per capita incomes in the City would remain below 1972 level. The level of households demanding inexpensive forms of housing will thus remain very high" (Amis p.2/3). According to 1977 estimates, the household income distribution in Nairobi was as shown in Table 2.4. This situation had not changed drastically by 1983 when the minimum wage was Kshs. 480.
It is estimated that 11,000 plots will be needed annually to meet the minimum needs of the increasing population. This rapid increase of households has not been accompanied by a similar development of housing. Thus, approximately 500,000 Nairobi residents are cu»- *-ly living in housing which is unplanned and unauthorized. Given the growth of households, the annual need for housing ranged from 9,000 units in 1972 and 17,000 ;n 1985. This did not take into account the accumulated backlog from past years. In addition to the monumental task of housing and the increased population, the problem of improving its older and now dilapidating housing units must be considered. To understand the current housing situation in Nairobi t one has to look at the historical development of the City. This section discusses housing development before and after political independence.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 2.4 1977 Estimated Household Incomes in Nairobi*
Cumulative % of Families
Monthly Rent in Kshs.
Monthly Rent in $ U.S.
0-20 0-500 0-66 21-40 501-1050 67-126 41-6C 1051-2000 127-240 61-80 2001- 000 241-482 81-100 Over 4 000 Over 482
U .$ = 8 Kenya Shillings in 1977
The 1977 estimated incomes have been used in this study because the real incomes of residents are sa not to have changed because of the slowing dovn of economic growth. Source: Chana 1984, 18
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
6 6
2.6.1 Housing Development Before Independence
Nairobi's history dates back to 1896 when a transport depot was established there. Its irtval growth however, began with the arrival of the Kenya-I ^anda Railway in Nairobi in 18*9. It was then that the railway authorities decided that Nairobi would be their headquarters. With this decision came extensive development of Nairobi. The City's first boundary was defined as 1.5 square miles from the office of the subconunissioner in 1900. There were subsequent boundary changes in 1919: 9.8 square miles, 1948: 32.2 square miles, and 1953: 266 square miles (Figure 5 and 6). Figure f> shows the layout of the City in 1903. Then the size of the City was 1.5 scruare miles.
The City grew rapidly, and by 1911 the agricultural activities began to flourish and there was a new wave of investment. The City began to grow in every direction; the settlers began to build comfortable residences on large plots in the Npirobi hilly areas to the west and northwest and in the Parklards to the North. Africans were already living in the eastern sections of the City as well (Temrle 1973, 28). By 1913, racial segregation was already being exercised to ensure that the choicest land was preserved for the Europeans and that the Africans could be controlled. In spite of all this, the African population continued to grow and by 1948 when the first census was taken, the population had increased from 29,864 in 1926 to 118,976 in 1948. By 1962 the
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
6 7
> zoe lu< LU o o coo oj m to Cft O ) O l o )
Reproduced with permission of the copyright owner. Further reproduction prohibited w ithout permission.
TO
M O
M B
A S
A
68
Figure 6
NAIROBI 1903
otoc^ ■ * Mathare River
McQueensmx J> House / ^
*PARKLAND African villages
B.E.A. Rifles African Market
1,1 mil,
Gov.. Officers' Houses
5ldUQnt6mOUS9 Govt
Offices
a V ^ ^ O l d Indian Bazaar Indian Bazaar a
r<y\H European Bazaar NAIROBI HILL Railway Hous
Railway Officers’ " Houses Q Cemetery
Source: Hake 19/7, 25
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
population had more than doubled to a high of 266,794 people. The African population comprised 60.3 percent of che total in 1926, 54.1 percent in 1948, and 59.5 percent in 1962 (Temple 1973, 32). This high African population was ignored as far as provision of services was concerned until the late 1930s. In the first 20 years almost all the Africans lived in unregulated settlements (see Figure 7). These were later demolished and the inhabitants were obliged to live in a demarcated native location. This location was called Pumwani and was opened up in 1922 for Africans and Arabs to rent plots from the Municipal Council and to erect lodging houses out of traditional materials called MPumwani,, (a place of rest) . Apart from this, other settlements existed like Kibera and Pangani. Outside the municipality, the Railway Administration provided housing for their employees. Until the Second World War, housing was provided under the assumption that the Africans who sought employment had come to the City for only a temporary period. Colonial policy evolved to a situation where only the actual labour force was permitted to reside in town. All other dependents were supposed to remain in the rural areas. Perhaps more important as far as urban housing was concerned, a unique population structure became characteristic, wherein adult African males predominated. In 1953, it was estimated that in the 'natives and other' category there were 23,750 males to 4250 females (a ratio of almost six men to one woman) (Stren 1978, 186)
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
R eproduced
with perm
ission of
the copyright
ow ner.
F urther
reproduction prohibited
w ithout
perm issio
n .
Figure 7
LAND USE IN NAIROBI MUNICIPALITY, 1939
OLD CITY AREA
Km
I I European Residential
I 'M l Asian Residential
H U African Residential
Source: Kingoriah 1980,224
Asian and European Residential European Commercial
Asian Commercial
i i
m
Industrial Area
Other uses: Mainly Undeveloped Schools and Other Public Buildings
•J c
The 1948 census revealed that 64,397 Africans, 43,749 Asians, and 10,830 Europeans, a total of 118,976, lived in the town. This led to further developments in housing. Estates were developed along the axis of what was then called Dolnholm Road, to be called Bahati (luck), Gorofani (two storeys) and Mbotela. Further areas were serviced and made available for employees. The government completed Starehe and the Railway Administration added 350 stone and tile rooms in Makongeni in 1947. By 1951, considerable progress had been made, but the government had to report that "despite the efforts, the shortage of African housing in urban areas remains acute, and there is no doubt that, owing to the expansion of industry, supply lagged further behind demand at the end than at the beginning of the year" (Hake 1977, 58). City Council reported that a significant feature of the year was shanty towns on the boundaries of the City and new African locations. Already it was becoming clear that neither the government nor the City Council could hope to keep pace with the demand for housing.
In all, the stock of housing available for Africans grew by over 30,000 bed spaces between 1946 and 1957. During that period, the African population grew from 63,000 to 115,000, an increase of 52,000 in spite of stringent controls. This shortfall must have been well over 30,000, and consequently, the overcrowding in Nairobi's saturated housing estates was worse than ever before (Hake 1977, 58). The actual housing provided consisted of blocks of dormitories within which
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
72
individuals were allocated bedspace accommodation. The policy for the first half century of Nairobi's growth was based on the concept of the "bachelor bedspace" defined as 40 square feet, or a space five by eight feet, enabling three to share a room ten by twelve feet. This was the module for most buildings up to 1956. Sometimes two renters shared a room, groupings of more residents were not rare. Other utilities such as kitchens, toilets, and showers wore used on a communal basis (Hake 1977, 59).
Around 1957, there was a change in policy towards housing for Africans. The policy swung towards providing family accommodation. Apart from the Makadara Estate, which was developed from 1954 onwards as an area where individual landlords or employers could build housing at their own expense, and the plots in Bahati on which richer Africans could build their own houses, efforts were concentrated on the new ofafa development later know by the residents as Maringo (posh), Jerusalem (the heavenly city), and Jericho (not too far from Jerusalem).
Between 1958 and 1962, 5,500 houses were produced (Hake 1977, 80) . It was clear that conventional housing schemes were no longer adequate to satisfy the needs of the unhoused and unemployed population, and that they were incapable of expanding sufficiently to cope with the crisis. From 1960 (the period of emergency 1952-1960 saw the removal of most illegal structures in such areas as Mathare Valley by colonial powers) onwards, the people turned for a solution pattern that
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
had been tried on a small scale since the beginning of the century. They built unauthorized shelters and they took up unregistered jobs but this time they did it on a grand scale.
2.6.2 Housing Development After Independence
Kenya got political independence in 1963, and with independence came radical changes in urban housing policy. The general policy became one whereby Africans were to be allowed to evolve into a permanent citizenry of urbanites and not just transient workers. To this end, the urban authorities became committed to creating self-contained units that would permit workers to bring their own families to live in town. This new City Council implemented its directive by building partitions within the colonial dormitories and grouping some rooms into single units. This measure eventually created more socially acceptable housing conditions for some low-income Africans, but it was also instrumental in decreasing the existing housing stock.
Another change which came with independence, following the repeal of the restrictive rural-to-urban migration laws, was dramatic population growth. The inflow of people was given an added momentum by changes that were taking place in the rural areas. Particularly in the districts of Murang'a, Kiambu and Nyeri, where a land reform program was reaching maturity, thousands who lacked the minimum accepted landholding rights, found their way to Nairobi. The annual
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
growth rate of 6 percent that characterized the period 1948- 62 rose to 9.3 percent between 1962 and 1969 (see Table 2.5). Most of the growth was experienced among the African population, who between 1962 and 1969 registered a net population increase of 195,000 (Muwonge 1982, 59). The impact of this growth on housing was that the traditional areas of African housing in eastern Nairobi quickly filled up, as residential densities of 50 persons per hectare in 1962 rose to 200-300 persons per hectare by 1969.
The Central Government, realising this acute housing shortage, invited United Nations experts to study the housing situation and make recommendations in 1965. The UN study confirmed the acute housing shortage in Nairobi, where most of the urban population lived at this time. In addition to this shortage, there was serious overcrowding with the majority of the African households living with two or more persons per room (Bloomberg and Abrams 1965, 15). Even at the extremely high ratio of three or more persons per room, 52 percent of all African households were overcrowded, with overcrowding being a greater problem in the one-room units. Another significant element that affected housing need was the estimated increase in the sir* of African households from 1962-1970. This meant more space requirements, although there would be fewer households relative to the population increase. The report further revealed that demand for housing was substantially less than need. Thus even if housing was provided for as little as Kshs. 30 per mrnth, it could not be
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
75
Table 2.5 Population Growth in Nairobi Since 1906
Year Europeans Asians Africans Total
1906 559 3582 7371 11,512 (4.9%) (31.1%) (64%)
1926 2665 9199 18000 29,864 (8.9%) (30.8%) (60.3%)
1936 5600 16000 28000 49,000 (11.6%) (32.3%) (56.5%)
1944 10400 34300 64200 108,900 (9.6%) (31.5%) (59%)
1948 10800 43749 64200 118,976 (9.9%) (36.8%) (54.1%)
1962 21476 86454 158865 266,794 (8.8%) (32.5%) (59.5%)
1969 19185 67189 422912 409,286 (3.8%) 13.2%) (83%)
1979 850,000 1981^ 1,150,000 1985 + 1,500,000
♦Estimates Source: Temple 1973, 32 and Central Bureau of Statistics
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
76
afforded by over half of the households in Nairobi. The report estimated that the housing requirement for Nairobi would be 3,400 additional units per year from 1962 to 1970 (a total of 27,200 housing units in eight years). Given this housing situation, the report recommended that government funds be directed towards serving the lower-income groups. The best way to implement this recommendation would be through the provision of site and service projects. The study also recommended the building of units consisting of single rooms. Along with this, was the development of minimum housing financed with long term low interest rate government loans.
In spite of the acute housing shortage, housing in the 1960s and 1970s was characterized by developments that mainly served the higher-income groups. The private sector channelled its funds into higher-income housing and the City Council was absorbed in the remodelling of colonial dormitories, resulting in a serious shortage of rental accommodation. Shelters were first erected in vacant areas within the central city and close to traditional areas of African housing. But this pattern did not hold for long because the land belonged to the City Council; periodic demolitions of squatter settlements was very often the rule. Table 2.6 shows the major demolitions since 1960.
Apart from these demolitions, housing development was also taking place. From 1966 onwards, new estates blossomed around the City in Uhuru, Kimathi, Harambee, California, Jamhuri, and Madaraka and within five years, the City Council
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
7 7
Table 2.6 Estimated Demolitions of Squatter Settlements
Since 1960
Year Place Individuals Dwelling Units
1963 Pumwani Not Known Not Known 1966 Langata 400-500 80-100 1969 Grogon Road Not Known 200 1970-71 Kaburini 48,000 8,943 1971 Eastleigh 22,000 3,900 1978-79 Nairobi River 10,000 Not Known
Source: Amis, 123
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
had built 3,800 houses (see Figure 8). By 1970, the NHC was building over 1,300 units a year in the City* and private building operations were adding 300-350 houses or flats to the total. In spite of this notable achievement, this still did not fully match the need of the City. The majority of the urban dwellexs provided themselves with substandard housing in the squatter settlements which were growing very fast. By 1971, it was estimated that one third of Nairobi's population- some If' .000 (estimated at 500,000 in 1985) were living in unauthorized housing. One such settlement is Hathare Valley which to date is the largest squatter settlement in Kenya. The population in this settlement quicKly jumped from less than 5,000 in 1965 to 90,000 in 1970 and to more than 150,000 in 1979. The Mathare Valley and the adjoining areas now form the main area oJ very low-income residence in Nairobi (Hake 1977, 93) (see Figure 9).
By 1970, it had become increasingly clear that no conventional housing programme could meet the demand for xow- income earners. To deal with this problem, it was realised that people have to be involved in the housing development themselves. Given these conditions, the housing need of up to 40 percent of the total population could only be met by the provision of a range of site and service facilities. T^e present minimum cost of Kshs. 24,000 was outside the economic limits of at least 30 percent of Nairobi's households, where the need for adequate shelter was most acute. The project was formulated to help satisfy the habitation needs of this sector
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
<Jf »
** LOCATION OF SOME OF THE MAJOR HOUSING
ESTATES IN NAIROBI
^5 O andcxa P ro je ct
Ki'mnri K*og*m rto ita N Fkrvita tala** W aithalra U#-'u Mufen KararVtangaM KtaaiWmxhy
to Gort Couraa tt Natntt* tWa*tSau**i
tnduatrrat Araa Mugum&m Embahaai Dandcva
Sourta MAZING IRA 1983. ?7 Zmmn 78 Pangan 3 t Ngara Waat 32 Ngara Eaat 33 K a n M w g i 34 Mktha ra 39 E a d a g h 38 At<«Ogoahoftigomongo
Harambos lumurrfca Makadara Katotarv Makongam M bolala Bahaa Manngo Uhuru UuVxrwa C ity Cantre
ACKNOWLEDGEMENTS
I wish to express my sincere gratitude to several organizations and individuals whose support and contribution made possible the successful completion of this work. Special thanks must go to my chief advisor, Dr. W.R. Code, for his assistance and patience throughout the writing of this thesis. I am grateful for his having read the drafts and offered constructive criticisms, suggestions and guidance. My special thanks also go to Dr. D. Janelle, Dr. M. Goodchild, Dr H. Hosse, Dr. E. Bjorklund, Dr. A. Philbrick and Dr. D. Cartwright and indeed to all the faculty for their support and encouragement. I am particularly grateful to Kim Holland for his support and great help with the maps. As well, I am grateful to Tim Bailey for his friendly he’p. I also appreciate the on going support of my other colleagues: Catherine Hooey, Quentin, Bert, Gloria, Tom, Shuo Chao Fu among others.
I wish to acknowledge the help from the administrative staff and especially Diane Shillington. Outside the Department, my thanks are extended to many friends for their great support and encouragement. I am especially grateful to Linda McLean, Kimberley Granger, Michele Anderson, Ruth Hufnagel, Ben Hartford and Wambui Kiai for all their support and help and for making mv stay in Canada such a pleasant experience. I am also grateful to Angeliek Wick, Yaap
v
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
R eproduced
with perm
ission of
the copyright
ow ner.
F urther
reproduction prohibited
w ithout
perm issio
n .
Figure 9
Dandora Projact In Relation to Estlmatad Inoomo Distribution In Nairobi
NAR08I NATIONAL P A W
AnnuN Income (in Konya Pounds) Incoma lovol By Word
□ o-soo
1001-3000 Source: MAZMGRA 1M3. f» Km3001 and OrerHqh
00 O
of population. Eligibility of the project population included those households able to afford dwelling charges not exceeding Kshs.. 120 per month, representing monthly incomes of Kshs. 600 and annual income of Kshs. 7,200. Thus, the site and service strategy was seen as the only realistic way to deal with housing problems of the low-income households. This was also in keeping with the government's housing policy between 1970-74, which stated that "if no alternatives are available, the lowest income families will bui-d the temporary houses they need anywhere as witnessed by the existence of large and expanding squatting areas near the urban centres (Republic of Kenya 1970, 519). The site and service approach was specifically geared towards implementation of affordable urban services, but it could also contribute to the creation of low- cost construction industry jobs. Projects were also designed to create other informal sector employment through the provision of goods and services with the projects.
The purpose of the scheme was to ensure that the eligible participants contributed as much of their efforts, skills and financial resources to their own dwelling houses as possible. Financial assistance in the form of material loans and technical assistance was provided to ensure proper control and adherence of procedures in the implementation stage. The projects were also seen as an alternative to squatting, and provided the local authority with a controlled development in its area of jurisdiction. By providing essential services, the programmes enabled a higher density with a proper
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
8 2
environmental standard to be achieved (Obudho 1987, 190). This was started in 1970, when Nairobi City Council agreed to lower the building standards so that Nairobi residents could play a key role in the construction of their own houses (Obudho 1987, 180-181) (see Appendix 1).
The Nairobi City Council established the Dandora Community Development Department, in 1975, as part of a loan agreement between the International Bank of Reconstruction and Development (IBRD), the International Development Assistance (IDA) ant the Government of Kenya. The housing Development Department within NCC was expanded in 1978 to implement the Dandora project and all similar low-income housing development projects in Nairobi. The Dandora project was therefore intended to benefit low-income households earning between Kshs. 280 and Kshs. 650 per month by allocating option A and option B plots, which accounted for 5,670 plots (95 percent of the plots), while option C plots were intended to benefit all other income households through sale at market prices and the derived surplus used to cross- subsidize about 50 percent of the plots, all in option A.
The planned monthly charges were to range from Kshs. 73 to 153 for option A and B plots. Based upon the assumption that expenditures for shelter and services was about 25 percent of a monthly household income, Option A plots (65 percent of the total) , would be affordable by household in as low as the 20th percentile (earning Kshs.500 or below) of the City's income distribution in 1975. Option B plots (30
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
83
percent of the total) would be affordable to households in the 30th to 40th percentile (earning Kshs. 551-1,050 per month). Furthermore, the plot and material loan repayment terms were made favourable by charging an interest rate of 8.5 percent per annum, and 20-30 year repayment periods for option A and B plots respectively. In order to encourage faster house consolidation, allottees were granted a grace period of five years on building loans. The project was therefore planned on the basis that all the capita, costs of the projects would be recovered from the allottees and the City Council.
In conclusion, the laying out of housing policy after independence suggests that social needs were allowed to prevail over economic realities . Thus, whereas the minimum acceptable standard during the colonial period was bed-space accommodation in units built of permanent materials, under the new order it became a two-room unit in permanent materials together with a kitchen, toilet, and shower. Although the households needed this kind of housing, they could not afford it as shown by the United Nations report on the housing situation in Nairobi. This is especially true given the fact that most people needing low-income housing worked in the informal sector and did not even earn the minimum wage which in 1983 was Kshs. 480.
Another aspect not given attention, was the increasing cost of housing which was not matched by an increase in incomes. Thus with the increasing cost of living in Nairobi,
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
8 k
many households in the lower-income groups continued to experience financing constraints, especially the lowest 20 percent. For example, in the late 1960s a house with two
Kshs.40,000 while the same house now costs Kshs. 100,000. Such property, even if subsidized is out of reach of the low- income groups, and is usually occupied by middle-income
Development Unit in 1977 indicate that some 70 percent of the population earns less than Kshs.500(U.S $ 1 - 8) . The official minimum standard of a two-roomed house with kitchen, toilet and shower costs Kshs. 35,000 and requires a loan repayment of Kshs. 400 per month. This kind of housing could not be afforded by many, and theoretically the site and service projects were adopted as a remedy to allow the iow- income households to have access to housing they could afford.
2.7 Housing Programmes
Existing housing available in low-income groups is provided through public, private formal and private informal or popular sectors. The public housing supplied by Nairobi City Council and National Housing Corporation similarly provides for about one third of the city's population. However, the majority of the public housing, especially that built in the 1960s and 1970s, caters for the medium and high
habitable rooms cost Kshs. 24,000 In 1970s it was
classes. Estimates made by the Housing Research and
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
85
income groups. The private formal sector provides the remaining one third of the housing for these income groups, including staff and servants' quarters provided by the employer for the low-income workers and rental tenements in some of the low-income settlements. This sector reported housing completions of 13,858 between 1963 and 1985 (Central Bureau of statistics 1963-85). The performance of the public and private sector since 1973 has only resulted in a supply of about 200 units per annum, each sector providing about 50 percent of this amount. Such a poor performance has resulted in high annual housing deficits, especially for the lower income groups who have continued to live in unauthorized settlements. The informal activities in the uncontrolled squatter settlements are a response to the failure of the public and private formal sectors to supply enough housing at affordable prices to meet the needs of the low-income households. These low-income settlements, which are located mainly in the eastern and western outskirts of the City, provide shelter for over 1/3 of the City's population (Chana 1984, 19).
As already discussed, the public housing agencies have been the major providers of low-income housing in Nairobi. This section discusses the housing programmes that are totally or partially financed by public funds. These include the Nairobi City Council (NCC), National Housing Corporation (NHC), which is responsible for implementing the government housing policy and the Housing Finance Company of Kenya. Housing Finance
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Company of Kenya, though mainly involved in middle and high income housing, is included in this category because it is partially financed by the Government.
2.7.1 Nairobi City Council
The Council's involvement in housing development goes back to colonial times. Thus by 1961 the Council had provided a total number of 10,782 dwelling units. Most of the housing constructed before independence was based on the "bedspace" policy and had very low rents. Thus 8,189 of all the housing constructed by 1961 had a rent of Kshs. 100 or less while only 2,593 of the units had rents of more than Kshs. 100. The situation changed after independence, where only a small portion of the housing constructed had a rent of less than Kshs.100 (Obudho 1987, 179).
By 1975, the Nairobi City Council had provided 20,667 housing units which included 16,599 rental units, 2,900 tenant purchase houses and 1,198 site and service projects (Coopers and Lybrand 1975, 19). A total of 7,440 of all the units had rents of Kshs. 80 or less, 6,451 had rents ranging from Kshs. 81-160 while 345 units had rents ranging from Kshs. 161-320. Tne units with rents ranging from Kshs. 321-640 included 1,535 of all the units while the remaining 828 units had rents of more than Kshs. 640.
As illustrated by Tables 2.7 and 2.8, the rents have significantly increased after independence and today only 56
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
housing units are rented for less than Kshs. 100. The remaining units range from Kshs. 103-1,725. Most of the rental housing constructed is in the eastern part of the City where the low-middle and low-income people mainly live. This area was predominantly occupied by the low-income households before independence.
As can be seen in Tables 2.7 and 2.8, only 4,2 54 rental housing units have been constructed by the NCC since independence. This compares very unfavourably with the number of increasing households per year, currently estimated at 14,800 households. This does not take into consideration the old housing, most of which has been long overdue for renovation. In addition to this, most of the rental housing constructed after independence was very expensive and could not be e 'forded by the low-income groups. For example, the cheapest housing constructed had rer.es starting from Kshs. 115. Assuming that one was willing to spend 25 percent of their income on housing only those earning Kshs. 460 could afford the cheapest housing constructed. In addition, very few houses were in this rent bracket. Thus only 922 housing units had rents ranging from Kshs. 115-400. The remaining 78.3 percent had rents ranging between Kshs. 401-1,725.
Compared with the period before independence, housing production was very slow and the housing produced was very expensive. This slow housing production has left many urban dwellers, who have little income, with no alternative but providing themselves with or renting substandard housing in
Reproduced with permission of the copyright owner. Further reproduction prohibited w ithout permission.
Table 2.7 Construction Date and Current Rents of City Council
Housing Units in Eastlands
Estate Year
Constructed Units Rents in Kshs.
Old Pumwani 1923 516 103 Shauri Moyo 1929-38 678 138-212 Ziwani/Gorofani 1941-46 1447 115-264 Kaloleni/Bondeni 1945-1948 806 115-667 Bahati/Landhies 1952-59 2C22 69-345 Mbotela/Ofafa 1 1953-58 2228 161 Maringo/Jerusalem 1955-59 1900 287 Embakasi/Kariobangi 1960 602 138-333 Dargoretti 1 1960 96 115-256 Lumumba/Jericho 1961-63 3004 207-402 Uhuru 1 and 2 1967-1969 886 267-356 New Pumwani 1969 224 575-690 Outer Ring 1970 360 517-690 Kariobangi Timber 1972 27 414-650 Harambee 1971 96 575 Huruma 1977 586 851-966 Kariobangi South 1979 720 851-966 Buruburu 1981 386 517-1380 Pumwani Maternity - 6 256 Total 16635
Source: Field Data
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
8 9
Table 2.8 City Council Rental Housing Located in Other
Parts of the City
Estate Year Constructed Units Rents in Kshs.
Old Ngara 1945 78 690-1104 Bachelors Quarters 1945 26 575-690 Joseph Kangethe 1949-58 286 920, 725 Pangar.i 1954 48 1,104 Jevanj ee 1955 26 1,300 Meru Road 1956 6 920, 104 New Ngara Road 1958 136 920, 150 Juja Road 1970 11 977 Caledonia Road 1960-61 2 1,500 Mariakani 1963 240 1,104 Kariokor 1965 240 1, 104 Jamhuri 1969 72 1, 150 Madaraka 1972 600 1,035, 1,150 Ngong Road 1973 30 115-356 Total 1801
Total Rental 18194 Total Rental Before Independence 13940 Total Rental After Independence 4254
Source: Author's Survey
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
squatter settlements. In addition, the shortage of housing has led to overcrowding of both the old and the new housing estates. As pointed out by the United Nations study, after independence most people needed cheap rental accommodation and only the public sector could develop this kind of housing. However, this does not seem to have been the case and the rents of the housing units constructed before independence not only increased but the newly constructed housing had very high rents.
The Council concentrated its resources on tenant purchase housing units which were not only few, but were constructed at a slow pace (see Table 2.9). Apart from the U.S.A.I.D sponsored Umoja project no other tenant purchase housing projects have been constructed since 1975. The major focus has been on the site and service project where housing has been produced at a slightly faster rate. The Dandora site and service project has 6,000 plots each, with five to seven rooms or more. Other site and service projects have been implemented or are being implemented. In conclusion, the small number of housing units constructed by NCC since independence illustrates, first, a low level of response to a problem of immense scale, and secondly, a growing emphasis on the provision of housing for the middle-income groups. This is true also of both National Housing Corporation and Housing Finance Company of Kenya. The majority of the housing built with funding from NHC costs more than Kshs. 24,000 which is beyond the reach of the low-income groups while HFCK
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 2.9 City Council Tenant Purchase Projects
Estate Date of
Construction No. of Units
Jamhuri 1969 156 Pumwani Relief 1970 216 Pumwani Redev. 1968-69 100 Harambee 1 1968-70 154 Uhuru 3 1968-69 236 Kariobangi 3 1968-69 206 Kimathi 1969-72 345 Uhuru 4 1970-72 294 Kariobangi Re-cast 1970-71 67 Kariobangi 4 1970-72 390 Mathare Valley 1972-73 557 Kriobangi 5 1973-74 28 Harambee 2 1973-74 151 Umoja 1976-78 3009 Total 5753
Source: Field Data
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
provides mortgages for middle and upper income groups. The Councils' funds were spent on units which were beyond the reach of half the City's population.
2.7.2 National Housing Corporation
The National Housing Corporation (NHC), which was instituted in 1967, was preceded by the Central Housing Board which began to function in 1943. The only loans made to local authority by CHB were for the development of site and service projects where individuals could build their own houses. Other loans made to local authorities indirectly were for lending to employers to enable them to build housing for their employees,
After the recommendation of the United Nations Mission to Kenya on housing in 1965, the Central Housing Board was transferred to the National Housing Corporation in 1967. The main objective was that the NHC should be a separate corporation more independent of the Ministry of Housing. Another reason was to give the corporation a wider field than just to local authorities, so as to undertake projects on its own. The Corporation is the government's main agency under the Ministry of Urban Development and Housing charged with the responsibility of implementing the government's housing policy.
Although NHC has activities all over the country, their major activities are in Nairobi. The housing developed has
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
included mortgage housing, rental housing, tenant purchase housing and site and service projects. Between 1967-1970 NHC completed 722 housing units in Nairobi. These were units mainly foe middle and high income housing. The 1970s saw an increase in the Corporation's activities where between 1971- 74, 5,518 housing units and sites were completed. A total of 62 percent of the total finance was used to construct housing units costing above Kshs. 24,000 (this was the maximum amount to be spent on dwellings constructed using public funds, according to the housing policy). During this time the average cost per unit was Kshs. 42,820. For example, in 1971 the total average cost per unit in Nairobi was Kshs. 24,000. This same trend was followed in 1972, except for three projects namely: Mathare site and service, Mathare tenant purchase projects, and Askari housing. The average cost ranged from Kshs. 28,800-91,000; 51.5 percent of the funds were spent on this category of houses, which consisted of only 408 units. The remaining 1,838 units were constructed with 48.5 percent of the finances. In 1973-74, only high-income housing was constructed, and this had a minimum cost of Kshs. 42,320. Similarly, between 1975-79, 5,260 housing units and sites were completed. Between 1980-86, a total of 2,041 units were completed. These included 385 mortgage housing units, 600 tenant purchase units and 1,056 site and service projects. The site and service units were part of the Dandora project which was started in 1975. Although there were more site and
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
94
service projects constructed at this time, most of the funds were used for middle and high income housing*
Although it is evident that the National Housing Corporation has been very artive since 1967, the concentration has been on middle and high income housing. Besides site and service projects, no other low-income housing has been provided by the NHC. This shows the important rc e that the site and service projects are expected to play in providing housing for low-income households. The production of these sites has been slow and the building costs high. They have however increased the housing stock available to the low income households.
2.7.3 Housing Finance Company of Kenya
The Housing Finance Company of Kenya (HFCK), the largest mortgage company in Kenya, was incorporated in 1965, as a national mortgage company with the main objective of implementing and supplementing the government policy of promoting thrift and home ownership by providing savings and mortgage facilities to the Kenyan public. The Company was founded on the ashes of the First Permanent Building Society which had been incorporated in Northern Rhodesia in the fifties and extended its operations in East Africa. The Company took over the Kenyan operations of the demised Building Society; thereby filling the vacuum created by the demise in the area of mortgage financing.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
9 5
Initially, the operation of the Company was restricted to the zoned residential areas within the City of Nairobi and Mombasa municipality. The price or valuation of the houses acceptable as securities ranged between Kshs. 30,000-100,000. In order to encourage new construction, it was the Company's policy that no less than 70 percent of the initial capital available was to be used for new houses. The maximum mortgage advance was 75 percent of the valuation, and the repayment period was between 10 to 15 years. Initially, estate development was disappointing, especially because there were hardly any private developers who could handle sizeable developments. The Company, therefore, had to depend on the National Housing Corporation which had been established for that purpose. The first major project to be financed by the Company was Woodley Estate in Nairobi: the first phase of this Estate consisted of 37 housing units with selling prices ranging between Kshs. 75,000-117,500. After Woodley was completed, the company financed other Estates such as Ngei, Kibera, Nairobi West etc. As the Company continued to grow, its area of operation was extended to include other urban areas outside Nairobi and Mombasa. Tne maximum limit of advances was increased to Kshs. 120,000. In 1970, HFCK acquired all the share holdings of the First Permanent (East Africa) Limited, thereby acquiring the latter wholly-owned subsidiary - Kenya Building Society Limited. At this time the Kenya Government became an equal shareholder with the
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
9 0
Commonwealth Development Corporation in the Housing Finance Company of Kenya.
Over the years, the Company has endeavoured to work towards its main objective of providing mortgages to Kenyans wishing to acquire their own homes. From its first year of operation, the Company has provided mortgage advances amounting to Kshs. 1,882,500,000 (see Table 2.10). The figures in Table 2.10 represent the amount of mortgage advances made directly to individuals to acquire homes between 1966-1985. In addition, the Company has advanced Kshs. 14,816,000 to the municipalities of Kisumu, Kitale, Nakuru and the City Ccuncil of Nairobi to assist in the local authorities' tenant purchase housing projects. Altogether, the Company has made it possible for approximately 15,000 families to be housed directly, in addition to many more housed indirectly through local authorities.
Housing Finance Company of Kenya does not directly undertake any development of housing estates. The Company is, however, very much involved in housing estate development, firstly, as a facilitator of the developments by guaranteeing availability of end finance to the prospective house purchasers, and secondly, by actually providing mortgage finance to the purchasers who require and qualify for the finance. HFCK has been involved in estate development in a slightly different way through one of its subsidiary companies, the Kenya Building Society Limited in respect of the Buru Buru Estate phase 5 and 5 infills. It should be
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 2.10 Yearly Mortage Advance by HFCK* Since 1966
Year Amount in Kshs. Year Amount in Kshs.
1966 4,000,000 1976 68,900,000 1967 5,300,000 1977 18,900,000 1968 7,800,000 1978 70,900,000 1969 21,500,000 1979 94,300,000 1970 16,100,000 1980 222,000,000 1971 32,600,000 1981 214,600,000 1972 68,200,000 1982 275,400,000 197^ 37,700,000 1983 152,400,000 1974 96,600,000 1984 204,500,000 1975 68,200,000 1985 202,300,000 Total 1,882,500,000
Housing Finance Company of Kenya Source:Field Data
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
98
noted, however, that it is Commonwealth Development Corporation who developed the first four phases of the Buru Buru estate - the largest single housing estate in Kenya. They also supervised the development of phases 5 and 5 infills of the same estate. The Buru Buru estate consists of about 5,000 units and is considered to have a population of more than 30,000 peop'e.
2.8 Conclusions
The housing shortage which is escalating rents at a rate higher than the growth of income, has been brought about by the concentration of investment in high standard housing and the high cost of units. Other reasons that have brought about housing shortage include the demolition of dwellings, the NCC low rental policy, and the restriction of authority approval to permit construction of dwellings by the low-income groups to standards within limits of their income.
All these activities have led, first of all, to a disproportionate amount of public resources being devoted to building the more expensive houses, which could have been financed from the private sector sources. Secondly, the policy of building tenant purchase units (50 percent of all units constructed by Nairobi City Council between 1965 and 1974) had the effect of transferring a large proportion of the new housing into the hands of a select group of private
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
individuals who could, by subletting on a big scale, consolidate their position as landlords with capital assets, in many cases provided for by the public purse and paid for by the tenants. Thirdly, while maay urban wage-earners got better housing, many more found that the new dwellings, costing more and more each year, were beyond their means; they, therefore, sought their shelter in the squatter settlements.
The post-independence approach has been criticised against a background of the doubling of Nairobi's population since 1962 and a country-wide disparity of income distribution (Coopers and Lybrand 1975, 5). Thus, the combination of greatly improved house sizes and standards, rapid inflation in land and building costs, and a scarcity of financial resources has resulted in a deceleration in public housing provision at a time when total housing needs have been growing at an unprecedented rate. Moreover, the higher rents or purchase prices pertaining to the additional houses have put them out of reach of the majority of new City residents who fall into the lower income groups. The inevitable result has been the overcrowding of authorized units (both public and private) and the mushroom growth of unauthorized accommodation, much of it well below the legal minimum standards of construction, density and amenity, often combined with totally inadequate provision of water and sanitation.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER 3
DATA 80URCE8 AND RESEARCH METHODOLOGY
3.1. Research Design
This study addresses the impact of site and service projects as a strategy for meeting the housing needs of the urban poor. It determines: the beneficiaries of the project; whether they moved directly from the squatter settlements or other substandard housing; and whether middle class residents created chains of vacancies such that squatters could indirectly improve their housing situation.
Data were obtained from two sources: primary data gathered by the researcher through personal interviews, utilizing a formal questionnaire, and existing secondary data. Survey data were essential due to substantial gaps in the general secondary data and the non-availability of information for the past and present situations in Dandora. The secondary data include information from the following organizations:
Ministry of Works and Housing Reports since its formation in 1966 National Housing Corporation Annual Reports since 1967 Nairobi City Council Annual Reports since 1963 Dandora Project Reports since 1975 World Bank Reports on the Dandora project
1 0 0 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
10 1
- Census Data , 1962 , 1969 , 1979 Statistical and Economic Surveys from the Central Bureau of Statistics since 1963 The two housing surveys conducted hy the Central Bureau of Statistics Information from the Housing Finance Company of Kenya
The lack of information on both tenants and current landlords necessitated the survey. The infonmation available from the government files for the plot owners in Dandora includes the original plot owners only. The many changes that have taken place over the past fourteen years are not documented, and no information exists on the tenants who occupy more than 80 percent of the rooms in the project. The questionnaire was, therefore, administered to both tenants and landlords (see appendix 2) . The two criteria used for selecting those to be interviewed were, that the respondent had to be a household head, and that the subject had to be currently residing on the project. The current resident criteria increased the likelihood that the person interviewed could provide the relevant information. With regard to tenants, it included those who rent the room. For plot owners, it included the person to whom the plot was allocated. Interviewing the household heads was particularly important because they were the ones who knew the details of the construction process and current housing expenses. The current resident criteria ensured that an absentee plot owner was not interviewed. Many of the plot owners either live
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
102
elsewhere in the City or in the rural areas, and only come to the plot once a month to collect rent. Others have relatives or friends on the plot to whom they give a free room or allow to pay minimal rent to maintain the plot and collect the rents. Those maintaining the plots knew little about the plot and the construction process. The absentee plot owners were also left out because although they benefited financially from the project, they were not benefiting from the actual housing, and were thus beyond the scope of this study.
3.2 The Method of Data Collection
The data were derived from detailed interviews of 300 household heads, consisting of 150 tenants and 150 plot owners. Initially the plot owners were supposed to sublet part of their plot to another household. Based on this assumption, an even split of numbers was used to ensure that the survey represented both types of beneficiaries. The inclusion of both tenants and plot owners was to ensure that the study got information that would enable analysis which would fulfil the objectives of the study. To assess whether the respondents benefited in either a direct and indirect way, one had to interview tenants who were not the intended beneficiaries, and yet comprise a majority of all living in Dandora today.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1 0 3
The two main languages used for interviews were English and Swahili. A few of the older respondents did not understand either Swahili or English. In these cases the research assistant was used as an interpreter because he was fluent in Kikuyu.
The field work conducted in Dandora extended over four months. Most of the interviews were conducted in the evenings and weekends because that is when the residents were available.
3.3 8aapling Procedure
The sampling procedure was determined by the characteristics of the population to be interviewed. The income of the residents for Dandora was assumed to be roughly homogeneous, since the allocation of the plots was based on the assumption that the beneficiaries were within certain income brackets. The different income groups were represented by the different options offered by the project. Thus out of the 6,000 plots, 65 percent were categorized as option A, and were meant for the lowest income group earning Kenya shillings (Kshs.) 280. A total of 30 percent were categorized as option B and were meant for those earning Kshs. 650 ( this was raised to Kshs. 1000). The remaining 5 percent of the plots were in option C. These were sold at market rates and were not meant for low-income groups. Thus, they are beyond the
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
10 4
scope of this study. These options had different levels of services. For example, option A had only a bathroom, shower and toilet, Option B had the above service plus one room, while Option C had the same service and two rooms constructed.
The sampling procedure chosen was stratified random sampling. The sample was stratified, in that 65 percent of the plots were in option A, while 35 percent were in option B. This was done so as to represent the two options that were meant for the low-income households. A random sample of 300 cases was done using random numbers. The sample frame obtained from the City Council was used to ensure that the random cases selected consisted of 65 percent option A plots and 35 percent option B. The cases selected included 150 owners and 150 tenants. Out of the 150 owners interviewed, 65 percent (98) cases) were those owning option A, while 35 percent (52 cases) were those owning option B plots. The same procedure was repeated for tenants who lived in option A and option B (see Figure 10). The selected plots were marked xi the layout plan with the interviewing starting in phase 1 (see Figure 2).
Before the survey was started, a draft questionnaire was administered to ten residents. As a result of this test, the questions were rearranged to facilitate a better flow of questions and a few questions were added. Some of the questions added included questions on who the respondents lived with, when they first moved to Nairobi, and whether they paid for services. This was added because it was found out that many of the respondents lived with others on their
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
105
Figure 10
SAMPLING PROCEDURE
150 Owners
Option A 98 Owners (65%)
Option B 52 Owners (35%)
300 Respondents
150 Tenants
Option A 98 Tenants (65%)
Option B 52 Tenants (35%)
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
10U
arrival to Nairobi. The question on services was added after it became obvious that most of the respondents' rents included payment for services*
3.4 Questionnaire Design
The questionnaire design was based on the objectives of the study which included an examination of who the beneficiaries of the project were and whether they benefited directly, as intended, or indirectly through the filtering process. To assess who the current beneficiaries of the projects were, the residents were questioned on their current socio-economic and housing situation. To determine whether they benefited directly or indirectly, migration histories were used. Migration histories provided accurate information on the respondents housing condition before and after moving to Dandora.
To collect the above information, a two-part questionnaire was developed and administered to both owners and tenants. Questions one to nineteen were the same for both tenants and plots owners. Questions twenty to twenty three were only for the plot owners because they addressed the construction process. The questionnaire addressed the following different aspects of the project:
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Questions l to 4: Household Demographic characteristics
Questions one to four provided information on the demographic characteristics of the households, and included gender, age, marital status and the number of children. This information provided insights into the current beneficiaries of the project. The age variable showed whether the project benefited a predominantly young population or an older population. The questions on marital status and residence of the spouse, and the number of children provided information on the occupants of the plots. Most of the residents left their spoi’-e in rural areas before independence. The residence of the spouse provided information on whether this trend of the household heads leaving their family in the rural areas had changed. If this has changed and most residents live with their spouses, they would require more room than they did before. Another aspect dealt with is the number of relatives living with the respondents on the plot. Most people stay with relatives or friends when they first move to the City. This information shows whether this in fact is true for Dandora.
Question 5: Occupation and Transportation
Section two of the questionnaire provided information on the occupation and means of transportation of the household head. The occupation held by the respondents determined the income
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
x o a
they earned, which in turn determined whether they could afford living in this settlement. The means of transportation and cost are important because of the location of the project from the city centre and the major employment areas. Information relating to distance from their place of work was also included. This data indicated who can afford to live in these projects, given the different work locations and the fact that most of these projects are located in the outskirts of the City. This is especially true in Nairobi, where the two large site and service projects implemented after Dandora are next to Dandora (Mathare North and Kayole projects all having a total of over 7,000 lots).
Question 6: Income Levels
This question provided information on the incomes of the beneficiaries. This information is central to this study, given the fact that the income level determined who could afford to pay the required housing expenses. This information also enabled the evaluation of whether the project benefited the low-income groups as intended. This is one of the major questions raised by this study. The income levels of the current Dandora residents enable an analysis of whether the residents are paying too much for housing or not.
Collecting information on incomes is usually very difficult, especially in the low-income areas where, the respondents are reluctant to talk about the specifics of their
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
incomes. The disclosing of income was especially a threat to Dandora residents because the project was meant for low-income households. Those earning higher income knew that they were not supposed to be in Dandora; and therefore, feared the possibility of getting into trouble with the authorities. This problem was minimized by grouping the residents income. This avoided the potential infringement of the respondents mentioning the actual sum they earned. The highest income range was also left open ended. The income earned from subletting was also addressed in a separate question.
Question 7 to 10: Migration History
These questions addressed the number of years the residents lived in Nairobi, the places lived in, the type and cost of housing, the level of services, the number of rooms occupied and the number of people per household. Another question asked concerned the housing left behind by the respondents when they moved to Dandora. This information was important because it allowed discussion of the migration histories, which in turn enabled one to assess whether housing was improved for the residents of Nairobi directly from the project, as intended, or indirectly through the filtering process. Information on the four types of respondents who benefited from the project was derived from these questions. The first two cases involved a situation where the resident benefiting would come directly from the squatter settlements
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1 1 0
or new households moving to the City. The other two cases showed whether respondents benefited indirectly through the filtering process. The respondents could have benefited through the filtering process had they initially moved from squatter settlements to standard housing and then to Dandora. As well they could have moved directly from standard housing to Dandora, thereby creating a vacancy which could have benefited other low-income households. The information from these questions also facilitated discussion on whether the project led to improved housing by reducing the number of persons per household. The migration pattern also allowed assessment of whether the housing r * increased or whether the move improved any other aspects of the beneficiaries' housing situation. Detailed information on the place the residents moved from prior to Dandora also allowed comparison of housing conditions at different levels to determine whether the move allowed improvement or not.
The use of migration history data was necessitated by the lack of relevant information on both the current and past housing conditions of Dandora residents. This information was used to infer whether the housing left behind could have indirectly benefited low-income households. Filtering studies done in the past have usually used information from the records on the housing previously occupied by the respondents. This information was not only unavailable from records, but would have been difficult to obtain from the residents since the residents would not have been willing to discuss the
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Ill
housing left behind due to various reasons. For example, some residents had moved from units that they had shared with other relatives or were illegally subletting from their friends. In these situations the residents would be suspicious of the researcher. This is because they would assume that one is from the City Council and wants to track down those who are illegally subletting public housing. Illegal subletting is quite common in situations where a housing shortage is acute. Another problem with giving detailed information on previous place of residence was the fact that most of the residents had previously stayed with relatives or friends and would therefore be hesitant to tell where these lived, mainly because they could not trust the motives for seeking information that they considered personal. Another reason that made it hard to get information on previous places of residence was the fact that even when one knew the address, many of the houses were not numbered and the residents only knew the location. The City Council records are also outdated, because when residents move from City Council housing, they do not usually report this but sublet it to others usually at a higher rate than the one charged by the City Council. This is illegal and collecting information on this housing would make the respondents suspicious. These problems made it difficult to directly get current information on the housing units previously occupied by the respondents.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Question 11: Reasons for Moving to Dandora
112
This question provided information on the respondents' reasons for moving to the project. This information enabled an evaluation of the benefits of the project as perceived by the current Dandora residents.
Questions 12 to 14: The Tenants' Present Housing Situation and Cost
These questions addressed the housing condition of the tenants in Dandora. The questions include the number and cost of rooms rented, the availability of electricity and kitchen facilities, the total number of people living on the plot and their opinion on the acceptable number of people per room. These questions provided information on the number of persons per room, to enable an assessment of the level of crowding. The cost and level of services available to the respondents as a result of the move to Dandora enabled comparison of these housing conditions with those found in the respondents' place of residence.
The questions dealing with the acceptable number of adults and children per room was important because it showed the respondents definition of crowding. The residents were incorporated in the definitions of adequate room. This enabled a more accurate definition of crowding and thus avoiding the biases inherent in the different definitions of crowding.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1 1 3
Question 15 and 19: Residents' Evaluation of Different Housing Components
The tenants were asked to rate and rank the services according to importance. Likert scales were used to determine what the most important services for tenants were and the level of that importance. This enabled the assessment of the importance of these services to the respondents and also allowed unbiased discussion of these services. Also discussed in this question is whether the tenants would like an extra room or a kitchen and why, and whether they preferred services to a permanent structure. This enabled one to assess whether what the respondents needed was more space or cooking facilities which most of them did not have. It was assessed whether the critical need for the residents was services, as assumed by the project, or permanent housing.
These questions also generated information on the most important components of housing to respondents. The questions addressed both provision of services and the materials used for the construction of the structure. Likert scales were used to measure the importance of the different housing components to residents. This scale eliminates the need for one to judge "by getting the subjects to place themselves on an attitude continuum for each statement" (Oppenheim 1966, 133) . The positions range from most important to least important. For scoring purposes these positions were given weights which range from 1-7. Each respondents was asked to check one of the seven positions. A. high score represented
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1 1 1
a favourable response, while a low score represented an unfavourable one. After scoring each item from one to seven, the item score was added to obtain a total score. Thus, for the six items, the possible range of total scores was from 7- 42 (7 x 6) for each subject. After scoring, a numerical score was given and total scores calculated. After this, an item- analysis was carried out to decide the level of importance of each item. A housing quality index was developed from these values. This index facilitated the evaluation of the residents' housing situation before and after moving to the project. Also, services were provided in these projects on the assumption that this was what the residents needed most. The answers to these questions indicated whether this has been accomplished.
Question 16 to 17: Housing Situation and Cost to Plot Owners
These questions generated information on the number of rooms occupied , whether these were shared or not, the repayment to the City Council and the number of people on the plot, the rooms constructed, the cost of services, the total number of people on the plot and the income from subletting. These questions provided information on the current housing situation of the plot owners. This enabled the analysis of the current housing conditions and comparison of these to the previous ones, to assess whether the housing situation had improved or not. The plot owners were allowed to sublet some
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
115
of the rooms to get access to more income, which was hoped to enable them to repay their loans to City Council. This information is important in assessing whether the money made from subletting improved incomes to the point where the owners could repay their loans.
Question 18, 20 to 23: Housing Construction Process and Cost
These questions addressed the method of construction, the total cost of construction, the number of people employed and sources of finance. The cost of services like electricity connection and monthly repayment were also included. These aspects are central to the site and service projects because they were supposed to make the significant difference in reducing housing cost, thus, making them affordable to the low-income households. The cost would be reduced by the lowering of the building standards and the people's involvement in providing labour during construction. The project was also expected to generate employment. These questions showed the level of plot owners' involvement and what the approximate cost of construction was. The number of people employed was also included and these show the approximate number of jobs created by these projects. Also discussed is the time taken to complete the construction of the plots. This enabled assessment of whether progressive construction took place. This type of construction was expected to enable the plot owners to build when they could.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1 1 u
This was hoped to make it possible for low-income households who do not have all the funds required for housing readily available, to benefit because they would only build their houses whenever they had the funds. Sources of finance for construction were also addressed by these questions. The actual amount owed was not available, because most of the respondents did not understand the housing finance mechanisms. They did however, know the different sources from which the money had been borrowed.
3.5 Field investigation Problems
The problems faced included language barriers where the respondents did not understand either English or Swahili. Those not understanding English or Swahili included a few of the older plot owners. All these spoke Kikuyu, in which the research assistant was fluent. In these situations the research assistant was used as an interpreter. Another problem was the high rate of absentee landlords. This was a very common problem; one spent a lot of time finding the plot owners who stayed on the plot. This problem was complicated by the fact that even those plot owners living on the plot worked either in the formal or the informal sector and were thus not available during the day. Most of those working in the informal sector were not available even in the early part of the evening because they worked until it was dark. These
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
could not be interviewed after it was dark because of security problems in the settlement. Although the project had street lighting provided, most of these are not operational and the settlement is dark and unsafe during the night. All these problems meant a lot of walking without finding the plot owners. Given the fact that only the plot owners residing on the plot could be interviewed, this caused unnecessary delays and thus prolonged the interviewing time. This was further complicated by the size of the project. The project is 350 hectares and interviewing involved a lot of walking. The layout plan with open drains made walking very difficult. This was worse when it rained because most of the roads were flooded. Another problem not anticipated was the fact that the interviewing schedule coincided with the City Council's eviction of the residents who had not paid their dues to the Council. This caused problems in that most people were unwilling to open the gates for strangers and especially if they had not paid all their dues. Although this problem lasted for only a few weeks it caused unexpected delays.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1 1 H
3.6 Method of Data Analysis.
The two techniques used to analyze the data included basic frequency statistics and cross tabulations. The analysis begins with the assembling of field information in chapter 4. The main objective is to provide information on the socio-economic characteristics and housing conditions of the respondents. A descriptive statistical package using frequencies and measures of central tendency is employed to summarize the raw data for a preliminary presentation of the beneficiaries of the project and their housing conditions before and after moving to Dandora.
To assess the different relationship between different variables, cross tabulations were also used. Cross tabulation is an "analytical technique which involves sorting of unit characteristics along two or more variables, such that each cell in the table represents different combinations of the categories of each variable, and the entire table includes all possible combinations for the particular category system being used" (Chadwick 1984, 434) The purpose of such a cross tabulation is to exhibit at a basic level, relationships existent between the variables (Grosof 1985, 233). This technique indicated if any relationships existed among the different variables and consequently showed who the real beneficiaries of the project were. Variables tested included sex, family size, income, occupation, rent, repayment rate, services, transportation, number of rooms, total number of
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1 1 9
people in the plot, migration history, method of construction and construction cost.
To evaluate the housing conditions before and after moving to Dandora, a housing quality index was developed. This included four measures of housing quality. These included the level of services, the number of occupants (to measure the level of crowding) , the housing cost and the building material used in the construction. The importance of these qualities was determined by the respondents and was addressed in the questionnaire. The respondents were asked to state the adequate number of persons per room. The respondents also rated and ranked the importance of different housing components. Using Likert scales, the components with the highest ratings were identified and ranked in order. This was used to identify the most important components to the residents.
To assess whether the respondents benefited indirectly through the filtering process, migration history was used. To analyze this, the housing conditions in different locations were identified. It was established where the residents moved from, what services were available to them, the housing expenses incurred, the number of rooms and the number of persons occupying these rooms. Four different moves were assessed to determine the level of benefits of the projects to the respondents. Direct movement includes those moving directly from the squatter settlements or the rural areas to the City. Indirect movement included those moving
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
120
from standard housing and leaving housing behind that could have benefited low-income households.
The Pearson Correlation was done to test relationship and strength of association. The variables tested included income and transportation cost, number of rooms and rent paid, income and rent and the distance travelled, among others.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER 4
SOCIO-ECONOMIC CHARACTERISTICS AND HOU8ING CONDITIONS OF THE DANDORA RESIDENTS
4.1 Introduction
The main objective of this study is to evaluate whether site and service projects are a viable way of meeting the shelter needs of the low-income families in Nairobi. The study identifies the beneficiaries of '•he projects and examines whether they benefited directly or indirectly through the filtering process. The chapter analyses the respondents1 socio-economic conditions as well as their past and present housing conditions. In addition, the chapter discusses the construction process of the Dandora plots. Chapter 5 and 6 evaluate how th.*s project directly or indirectly benefited the low-income households through the workings of the vacancy chains. Direct benefit includes the households who moved to Dandora directly from the squatter settlement or the rural areas. Indirect benefit includes the households wno benefited as a result of higher income groups moving from standard housing, and leaving housing behind that may have benefited the low-income households. It is assessed whether the moves improved the housing situation of the households.
12 1
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
The survey established that the project mainly benefited households who had lived in Nairobi for many years. Most of the respondents had jobs in both the formal and the informal sectors. Although a few of the respondents had temporary jobs, the majority of respondents had permanent jobs. Some did not have any occupation but indicated that their plots were their only source of income. Only a small percentage of the respondents did not have any source of income, and they had friends or relatives paying their rents. Most of the respondents earned more than Kshs. 1,000 and had not come from the squatter settlements but rather from other types of housing.
4.2 Socio-economic Characteristics
The variables to be discussed in this chapter include age, gender, marital status, number of people per household, occupation, cost of transportation, means of transportation, and income. These characteristics are important in identifying the beneficiaries of the project. The objective of the project was to meet the housing need of those who had not previously benefited from conventional housing provided since independence. The strategy promoted home ownership, which up to this point was mainly dominated by middle and high income groups. The middle and high income groups had access to mortgage facilities, while low-income households did not
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
12 3
earn enough money to pay the down payment or to qualify for mortgage loans. The Dandora project allowed lower standards so as to provide affordable housing to the low-income groups. In Dandora services were provided at a minimal cost and the residents were expected to develop their plots. The loan terms were geared to low-income earners rather than high incomes households as they were before. Thus, the low-income households who worked in both the wage sector and in the informal sector were able to benefit from the project.
According to the survey, the project beneficiaries comprised 42 percent female and 58 percent male household heads. The female household heads were characterised by lower incomes when compared to the male household heads. This is because at the time of independence, women were not allowed to come and live in the City. Over time, when they finally came they worked mainly as maids and in the informal sector. They were also among the least educated and did not benefit from the jobs offered in the City immediately after independence. Their low incomes had excluded them from benefiting from the conventional housing constructed since independence. This project made it possible for a high percentage of female households to possibly obtain their own dwelling units.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1 2 •
4.2.1 Age Distribution and Length of Stay in Nairobi
Nairobi has a predominantly youthful population. Among males, 50 percent of the male population is below 24 years old, 75 percent below 35 years and 90 percent are aged 44 years or less. The female population has 56 percent of the population below 19 years, 80 percent under 29 years and 90 percent under 39 years (Obudho 1987, 43). According to the survey, the respondents were aged between 18-90 years. The mean age was 39 years. The female household heads aged 19 years accounted for .3 percent while those aged 29 years or less comprised 7.*? percent of all the female household heads. Those aged 39 years and under accounted 18 percent of all the female respondents. ^he remaining 24 percent were aged between 40 and 90 years. Thus the project served much older female household heads. The male household heads were also generally older than that of the average age in the City. A total of 5 percent of the males were aged 24 years or less while 30 percent were 35 years or less. A total of 45 percent were 44 years old or less with the remaining 13 percent being 44-76 years. Thus, the project served a much older population of both male and female household heads as shown by Table 4.1.
As can be seen in Table 4.1, a small percentage of household heads were less than thirty years of age. Thus only 33 percent were aged between 18-30 years while 67 percent of the households were aged between 31-90 years. However, most of those under 10 years were tenants. Thus 63 percent of all
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1 2 5
the tenants belonged to this age bracket compared to 3 percent of the plot owners. The plot owners who were under 30 year? were not among the original plot owners; they either inherited the plot from parents or bought it after allocation of the plots by the city Council. The above analysis shows that the majority of the young population who benefited from this project were tenants, while in terms of ownership, the plot benefited an older population.
The respondents had lived in Nairobi for a long time as indicated in Table 4.2. A total of 39 percent of all the respondents had lived in Nairobi for a period ranging from three months to 10 years. The remaining 61 percent had lived in Nairobi for a period ranging from eleven to sixty years. Except for 7 percent of the respondents who were born in Nairobi, all the other respondents had come to Nairobi from rural areas. Those born in Nairobi were aged between 27-60 years. The plot owners had lived in the City longer than the tenants. Thus the average number of years lived in Nairobi for the plot owners was twenty three compared to nine for tenants. The highest percentage of those who had lived in Nairobi for a short time were tenants. This consisted of 42 percent of all the tenants who had lived in Nairobi for five years or less. Only a small percentage of 9 percent of all the plot owners had lived in Nairobi for 10 years or less. This compares with the majority of 69 percent of all the tenants who had lived in Nairobi for a period ranging from three months to 10 years. The percentage of tenants living
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1 2 6
Table 4.1 Age Distribution of Dandora Residents
Age No. from Sample Percent
18-30 97 33 31-40 90 30 41-50 56 19 51-90 55 18 Total 298 100
Source: Field Data
Table 4.2 Number of Years the Residents
Lived in Nairobi
No. from General Years Sample Percent
.3-5 62 21 6-10 53 18
11-15 39 13 16-20 48 16 21-30 57 19 31-40 28 9 41-60 11 4
Total 298 100
Source: Field Data
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1 2 7
in Nairobi for 11 to 20 years accounted for 20 percent of the tenants which compares with 38 percent of the plot owners. The remaining 11 percent of the tenants lived in Nairobi for periods ranging from 21-47 years. This compares with 54 percent of the plot owners who had lived in Nairobi for periods ranging from 21-60 years. The above data shows that a high percentage of plot owners were older than tenants. This is mainly because at the time of the implementation of the project, 14 years ago, those applying for the plots were required to have lived in the City for a minimum period of two years. Also, to get the required funds for construction, they needed to have worked for a number of years. Unlike the tenants, the plot owners had lived in Nairobi for a long time, and many were established in the informal sector; and therefore, had savings that they could use for construction.
4.2.2 Family Composition of the Dandora Households
The respondents interviewed included married and single household heads. The number of respondents who were married or single showed the housing requirements of the different households. In addition, the project was intended for those with families, so as to enable them to live with their families in urban areas. The housing units constructed before independence were characterized by "bedspace" accommodation, which catered for single males who were not expected to bring
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
12 8
their families to the urban areas. This housing policy was adopted because the colonial authorities had no intentions of letting Africans become permanent urban residents. The end result of the bedspace accommodation policy was the predominance of one-unit dwellings which by independence consisted of 80 percent of all the dwellings in Nairobi (Bloomberg and Abrams 1965, 5) . Although this policy was changed after independence, the one-units dwellings continued to house most of the urban residents, and especially the low- income households. This was mainly because of the few units constructed in spite of the rapid growth of the City. In addition, whatever houses were constructed mainly benefited middle and high income groups rather than low-income households who were increasing at a much faster rate. The low number of rooms constructed, and the low wages, had made it hard for families to live together in the urban areas. This led to most male household heads living on their own in the city while leaving their wife and children in the rural areas. In some instances, the household heads stayed with their wives on the plot but left their children behind with relatives in the rural areas. The Dandora project was expected to make it possible for the families to live together. This was stated in eligibility criteria which required that the applicant live with the family or plan for the family to join them. These requirements excluded unmarried residents, especially those without children.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1 2 9
A total of 56 percent of the respondents were married and 44 percent were single. Some of the married respondents left their spouse and some of their children in the rural areas. Thus out of 56 percent of the respondents who were married, 33 percent lived with their spouse in Nairobi while 21 percent did not. Those not living with their spouse had either left them in the rural areas (19 percent) or they lived in other towns (3 percent). Although the number of household heads living with their spouse has increased since independence, 38 percent of all the married respondents had to leave their spouses in the rural areas. The reasons given for leaving the spouse behind were low incomes and inadequate housing. These reasons had also made it difficult for some residents to stay with all their children in the urban areas. The children were thus left in the rural areas with relatives or wives.
The number of children per household in the project is important, given the limited number of rooms per household. Not all the respondents had children; a total of 15 percent of all respondents had no children while 85 percent had children ranging from one to sixteen. Those with children included both married and unmarried respondents. A total of 51 percent of the households had one to four children while 34 percent had five to sixteen children. Those who left all their children in the rural areas comprised 36 percent while 49 percent had one to ten children on the plot. In addition to immediate families, 30 percent of the respondents lived
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
130
with relatives ranging in number from one to eight. These figures indicate large households with most of the married respondents living in the urban areas with their families. This increase has not been accompanied by an increase in the number of available rooms. This has led to high level of crowding. Although the project encouraged family housing, the housing shortage and low incomes make it impossible for some respondents to live with their families in Dandora. The single households were also characterized by large household sizes, this was mainly because these lived with relatives and friends and they also had children. Most of the respondents did not earn high enough incomes to rent mere than one room, and crowding levels were high even among the single households heads. The large household sizes and the slow pace at which the dwellings are being added to the housing stock will mean that most households in Nairobi will continue to be crowded. Also, some households will have to leave part of their family in the rural areas.
4.2.3 Jobs Held by the Dandora Residents
The occupation of the household head dictated the income, which in turn determined who was eligible for Dandora housing. A stable income was required to be able to pay rent and other monthly housing expenses. Those without jobs had friends or relatives pay their housing expenses or had some other source of income. This group included those who were unemployed,
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
13 1
retired or students. All those without any source of .income were tenants and consisted of 4 percent of all the respondents. Not all plot owners had occupations, but most of them had income from subletting their plots. There were 16 percent who solely depended on income from subletting their plots. The remaining 80 percent worked either in the informal or the formal sectors (see Table 4.3).
A high percentage of these respondents worked in the service sector as casual workers, barmaids, musicians, machine operators, factory workers, bus drivers and conductors, masons, carpenters, clerical officers, messengers, and cleaners among others. Most of the women were either self- employed in the informal sector or had their own plots as their only source of income. For example, 41 percent of all the women employed were self employed, compared to 24 percent of the men who were self employed. Another 29 percent of the women had their plots as their only source of income, compared to 7 percent of the men.
Lack of jobs in the formal sector has led many Nairobi residents to be involved in the informal sector. Most of the people in this sector are self employed and work as shopkeepers, vegetable sellers, tailors, dry-cleaners, hairdressers among others. Some of these jobs are also not permanent. Temporary jobs provide no job security and in the informal sector there are no agreed-upon wages or working conditions. Most of the respondents in the informal sector were also self-employed. Thus, 89 percent of all respondents
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1 3 2
Table 4.3 Jobs Held by Dandora Residents
Occupation Percent informal
Percent Formal
No. from Sample
Percent Total
Self-employed 27 0 82 27 Office Support Staff 0 24 70 24 Drivers/Conductors 0.4 5 17 6 Machine Operators 1 4 13 4 Electric Mechanics 1 4 13 4 Services 0.3 3 11 4 Building Trades 2 1 10 3 Barma ids 2 2 10 3 Casual Workers 0.3 2 8 3 Factory Workers 0 1 3 1 Landlords 49 16 Unemployed 14 5 Total 34 46 300 100
Source: Field Data
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1 3 3
having permanent jobs in the informal sector were self employed. The remaining 6 percent were in the building industry, while the others were equally distributed (one percent each) as barmaids, working in hotels, electrical mechanics, drivers or conductors. Similarly, 75 percent of all temporary workers in the informal sector were self- employed. This was followed by 13 percent in the building industry while the others were equally distributed (4 percent each) as barmaids, casual workers, and electrical mechanics.
A total of 46 percent of all the respondents worked in the formal sector. These comprised 41 percent who had permanent jobs while 6 percent had temporary jobs. Those employed as office support staff comprised of the highest percentage of those with permanent jobs in this sector. This accounts for 49 percent of all those permanently employed in this sector. The remaining respondents were distributed as follows: 12 percent were either drivers or conductors, 8 percent worked in hotels, 7 percent worked as machine operators and the others were distributed in many small sectors. Only 5 percent of those in the formal sector had temporary employment. These were distributed as follows: 38 percent were casual workers, 25 percent worked in the private or public sector and 12 percent worked as machine operators. The others were are evenly distributed (8 percent each) as barmaids, electrical mechanics, drivers and conductors. Except for those with temporary jobs, most residents had stable incomes.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
134
4.2.4 Transportation System and Cost to the Residents
The eleven kilometre distance of this project from the City Centre and other employment areas made the transportation system and cost to and from Dandora important. High travel costs would affect the target income for the project. This section examines where the household heads worked, what their means of transportation were and how much it cost to get to work. The distance travelled varied from those working in Dandora to those walking twelve kilometres or more (see Table 4.4) .
The respondents who had no distance to travel included those who were unemployed, landlords who did not work, and people who worked on the plot. Many of those working on the plot included those self employed with shops, laundry facilities, tailors etc. From Table 4.4, 26 percent of the respondents worked in Dandora while 5 percent worked between one and five kilometres from Dandora. The remaining 47 percent worked six or more kilometres away. Most people in Nairobi working within five kilometres walk to work. This is especially true of the low-income workers who cannot afford to pay for transportation. Most of the respondents, who worked six or more kilometres away, worked in the City Centre, industrial area or beyond the City centre. Another of the project's objectives was to have an industrial component in Dandora and thus reduce travel cost. This part of the project
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 4.4 Distance to Place of Work from Dandora
Distance in Km No. from Sample Percent
0 67 22 Within Dandora 77 26 1-5 15 5 6-11 123 41 12 + 18 6 Total 300 100
Source: Field Data
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1 3 O
was not implemented and most of the people working in Dandora were self-employed.
The respondents either walked, took public transport, used their own car or had transport provided by the employer. Those who walked to work comprised 28 percent. Most of them lived within five kilometres from Dandora. The most popular type of transportation was public transportation which was used by 38 percent of all the residents. The public transportation includes public buses and small buses called 'Matatus*. The Nairobi Matatus are the equivalent of the one time "Jitney" of North America, the "Taxi Bleus" in French West Africa, the "Jeepney" in the Philippines, and the "Public Light Bus" in Hong Kong. These small buses are one of the most important means of transportation and are used interchangeably with the public bus system. These were classified under public transportation since payment is the same as for public buses and the intra-urban traveller has to use both to go anywhere. Only one percent of the respondents had private cars. Another popular means of transportation was provided by the employers. Some of the big companies usually sent buses to Dandora every day to pick up those people working for them (12 percent) . The employers who sent buses to Dandora had mainly shift workers. They sent buses to areas with a higher concentration of workers. The reasonable rent in Dandora for most of these workers made it possible for them to live in the settlement, which gave them access to free transportation. Some of the residents had moved to Dandora
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1 3 7
because of the availability of employer-provided transportation. Increasing these types of settlements may encourage more employers to provide transportation for their workers. This would make transportation cost a minor expense for some of the residents working elsewhere.
The type of occupation and the place of work influenced the respondents' choice of transportation. Thus, 68 percent of those permanently employed in the informal sector walked to work; 30 percent took the public transportation system; while one percent took employer-provided transportation; another one percent used their own cars. Those working as temporary workers followed more or less the same pattern with 67 percent walking to work; while 29 percent took the public transport system, and the remaining 4 percent took employer- provided transportation.
Distance to work for those in the formal sector varied from 14 percent who worked within Dandora, to 86 percent who worked one to over twelve kilometres away. These comprised 7 percent who worked one to five kilometres away; 67 percent who worked six to eleven kilometres away; while the remaining 12 percent worked twelve or more kilometres away. A total of 13 percent of respondents walked to work, 58 percent took the public transportation, one percent took their own cars while 28 percent took employer-provided transportation. Almost all the respondents taking employer-provided transportation (92 percent) were permanently employed in the formal sector. In sum, more respondents from the inform 1 sector walked to their
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1 38
place of work than those in the formal sector. The majority of those in permanent employment in the formal sector (58 percent) took public transportation, while 68 percent of all those in permanent employment in the informal sector mainly walked to their place of work. The high percentage of those walking in the informal sector can be explained by the fact that most of these respondents were self-employed or worked within or near Dandora.
Transportation costs varied from those who did not pay for transportation to those who paid Kshs. 700. Those not paying include 57 percent of all the respondents. As can be seen in Table 4.5 the average transportation cost was Kshs. 66, with a majority of 92 percent of the respondents paying Kshs. 200 or less. Only a small percentage of 8 percent paid more than Kshs. 200. Some of the respondents who indicated that they walked to work paid for transportation (11 percent) . These were those respondents who walked daily to their place of work, but travelled once or twice a week to get their supplies. This was especially true of those who were self- employed and worked as vegetable hawkers; they had to go to the main markets to get vegetables or other supplies from time to time. A majority of 86 percent of all who paid for transportation took the public transportation system and they paid between Kshs. 24-400. Most of the residents paying Kshs. 301-400 worked twelve kilometres or more away. Only one person paid more than Kshs. 400 and they used their own personal car.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1 3 9
Table 4.5 Monthly Transportation Costs for Dandora
Residents
Cost in Kshs. No. from Sample Percent
0 170 57 24-100 20 6.7 101-200 100 33 201-300 6 2 301-400 3 1 401-700 1 0.3 Total 300 100
Source: Field Data
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
14 0
According to the survey, no significant relationship was displayed between the means of transportation and income. For example, 26 percent of all those who walked earned between Kshs. 301-2,001 or more, while 38 percent of those who took the public transportation system earned the same amount.
4.2.5 Monthly Incomes of the Dandora Residents
This section addresses income, which was the crucial determinant as to who the beneficiaries of the project were. Affordability was one of the major issues discussed at the implementation of this project. As shown in Figure 11, only 28 percent of the total number of respondents earned Kshs. 280-1,000. The largest percentage of respondents (37 percent) earned more than Kshs. 2,000. This was followed by 35 percent of the respondents who earned Kshs. 1,001-2,000. T h e percentage of those earning Kshs.1,000 or less decreased for plot owners while it increased for tenants. When the incomes of the plot owners are examined, only 17 percent of the plot owners earned less than Kshs. 1000. A higher percentage of tenants earned Kshs. 1,000 or less (see Figure 12). Thus, 39 percent of all the tenants earned Kshs. 1,000 or less. The high percentage of tenants earning Kshs.1,000 or less is explained by the fact that they only paid rent and did not need to raise extra funds for construction. The plot owners, however, needed more finances to enable them to do the
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Fi gu
re
11 R
es p
o n
d en
ts ’
M on
th ly
In
co m
e in
K sh
s.
14 1
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
S o
u rc
e :
Fi el
d D
a ta
Figure 12 Owners’ and Tenants' Monthly Incomes
P e r c e n t 60
50
40
30
20
10
0 ■ I 3 0 0 or le s s 3 0 1 - 5 0 0 5 0 1 - 7 5 0 7 5 1 - 1 0 0 0 1 0 0 1 - 1 5 0 0 1 5 0 1 - 2 0 0 0
In c o m e in K sh s.
O w n e r s T e n a n ts
2000*
S o u r c e : F ie ld D a ta
1 1 3
construction. As shown later, plot owners' housing expenses were higher due to loan repayments and payment for services.
4.3 Building Process
This section deals with the process involved in the building of Dandora houses. The strategy adopted progressive development, where the plot owners could build whenever they had finances to do so. This came into effect after the construction of the first two rooms, which were supposed to be constructed in 18 months. After the allottees were given the plots with services, they were expected to do most of the construction. Type A houses had no rooms built while type B had one room built. The plot owners were expected to build a total 4-6 rooms. Some of the plot owners had built up to nine rooms by the time of the survey, thus making the plot coverage 90 percent in certain cases. Instead of leaving a courtyard where the occupants could sit or use for other purposes, some of the plots only had corridors leading to the rooms.
The plot owners were encouraged to join self-help groups. These group were expected to reduce the construction cost of these plots by providing cheap labour. The study revealed that only 7 percent of the plot owners were involved in these groups. For those who joined these groups, only one room per household was constructed with the help of the group, while
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
others simply paid money to the group and did not participate in the actual construction of the room.
The construction of the first two rooms was supposed to be completed in eighteen months after the date of allocation. The other rooms were expected to be completed when the landlords could do so. At the time of the interview, most houses were completed, although there were a few which were not. Due to the nature of this study, only the household heads whose plots were completed were interviewed. Construction time ranged from one year to thirteen years with the average construction time being five years. According to Figure 13 the plot construction was progressive and the plot owners built when they could. This was one of the objectives of the project. A progressive construction process was allowed, to enable the plot owners without available funds to build when they could, which made the project accessible to low-income households. Almost one-third of the plot owners finished the total construction in two years. This was the amount of time in which the plot owners were expected to construct two rooms. The respondents finishing within two years were those with funds readily available at the time of allocation. Within six years, 70 percent of all the respondents had finished their plots, while the remaining 30 percent took seven to thirteen years. Most of those who took the longest time to build stated that finance was the major constraint. Some of the respondents did not do the actual construction but bought the plots after they had been
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Fi gu
re
13 Pl
ot
C on
st ru
ct io
n T
im e
1 4 5
CO
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
S o
u rc
e:
Fi el
d D
a ta
1 46
constructed. These accounted for one percent of the plot owners.
Construction of the plot was 'one by both skilled and unskilled people, as indicated by Table 4.6. As can be seen in Table 4.6, only one respondent constructed his own house while one other had the house constructed by his son. A total of 50 percent of the respondents used subcontractors to do construction. These subcontractors are commonly known as "Fundis* and are not usually linked with the formally-trained contractors who get the big contracts. The subcontractors played a significant part in the construction in Dandora. A high percentage, 45 percent of the plot owners, supplied unskilled labour for construction, while the Fundis did the skilled labour. This provided the expected cheaper labour which was hoped to reduce the cost of construction. A small percentage of two employed formal contractors to do their construction, so they only provided the finances. The contractor bought all the materials and did construction while most of those who helped the Fundis had to buy the building materials and help the Fundis with the unskilled work. The families of the plot owners also helped to provide unskilled labour. Thus 33 percent had their families involved in doing some unskilled work while the remaining 67 percent did not have any of their families helping with construction.
This project generated work for both skilled and unskilled people. The actual number employed varied from time to time since the development was progressive. The most common
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
147
Table 4.6 People Involved in the Construction
Process
Involvement No. from Sample Percent
Self 7 5 Subcontactor 75 48 Self and Subcontactor 68 43 Contactor 3 2 Son 1 1 Bought 2 1 Total 156 100
Source: Field Data
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1 4 8
number of skilled workers employed was one (see Tables 4.7 and 4.8). Thus 60 percent of the respondents employed one skilled person, 34 percent employed two, while 5 percent employed three to five people. The plot owners usually had one person working for them because that is all they could afford. Since construction of different units took place at the same time, the Fundis could commit themselves to different plot owners and work whenever the owners had funds.
Although the respondents were expected to provide unskilled labour, only 9 percent of the respondents provided all the required unskilled labour. Most of the respondents provided labour only on weekends because they had full-time jobs. Those working in the informal sector did not have the weekends free, because this is their busiest time. They were therefore forced to employ unskilled labour. Thus 76 percent had hired two unskilled workers at a time, and the remaining 15 percent had from three to six unskilled workers. The plot owners employing one unskilled person sometimes provided labour because at different times of construction more unskilled workers were needed. The involvement of the plot owners in the construction process enabled them to provide cheap or free labour, and thus made the project more affordable for those available to do some work.
There were four sources of finance for construction. These included personal savings, loans from the City Council, loans from one's place of work and help from family and friends. The major sources of finance for this project was
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 4.7 Average Number of Skilled People
Employed
No. Employed No. Sampled Percent
0 8 5 1 84 56 2 51 34 3 4 3 4 2 1 5 1 1
Total 150 100
Source: Field Data
Table 4.8 Avarage Number of Unskilled Workers
No. Employed No. Sampled Percent
0 13 9 1 66 44 2 48 32 3 7 4 4 10 7 5-6 6 4 Total 150 100
Source: Field Data
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1 5 0
supposed to be the plot owners' savings. A small materials loan was given by the Nairobi City Council to help the plot owners. Those benefiting from this loan consisted of 86 percent of all the plot owners. The materials loan was later increased but this was after most of the respondents had finished construction. The remaining 14 percent of all the plot owners did not take loans from the City Council. Most of those not taking loans had either bought the plots and paid back the Nairobi City Council loans or had declined to take the loan right from the beginning because they did not need it. Other sources of finance included personal savings, loans from cooperatives or money given by family and friends. The major source of finance was personal savings, utilized by 98 percent of all the plot owners. Only two percent solely depended on other sources of finance and did not use their savings. Provision of finances by family or friends was not significant, and benefited a small percentage (25 percent) of plot owners. Another source of finance for those who had jobs in the formal sector, was loans from cooperatives at the place of work. Those getting loans from their place of work comprised 32 percent. Except for two percent, the plot owners did not get finances from one sources only but used several sources.
The estimated total cost of the plots ranged from Kshs. 20,000 to 300,000 as indicated in Table 4.9. The construction cost was higher than expected, with the average construction cost being Kshs. 17,031 per room. Construction cost for a
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1 5 1
two-roomed unit was expected to be Kshs. 24,000 at the time of the implementation of the project. This was much lower than in Dandora where the average cost for two rooms, excluding services, was Kshs. 34,062. The reasons for the high cost, among others, included the increasing cost of building materials and high building standards. Although Grade II By-laws were used, they were not low enough to significantly redoce the building cost. Electricity connection was said to ba o. of the major costs. Most plot owners having electricity paid more than they expected. Electricity connection was also unavailable to 43 percent of the respondents, as illustrated in Table 4.10. A total of 11 percent of those included in Table 4.10 only had wiring done at certain stages while 46 percent haa electricity connected on the plots.
The above discussion shows a higher housing construction cost than intended. The reasons for this include the high building standard, the increasing cost of materials and the fact that most plot owners had to use hired labour. The assumption that the respondents would provide cheap labour ignored the realities of beneficiaries, most of whom had full time jobs and were only available on weekends. Some of those working in the informal sector did not have the weekends off, which forced them to use hired labour. The building groups expected to help reduce construction also faced the same problem, because the respondents were not available to join these groups. The few that did faced many problems and
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
l b :i
Table 4.9 Estimated Total Construction Cost of the
Dandora Plots
Construction Cost No. from sample Percent
20000-50000 18 12 50001-100000 78 54 100001-150000 32 22 150001-200000 15 10 200001-300000 4 2 Total 147 100
Source: Field Data
Table 4.10 Estimated Total Electricity Connection Cost
in Dandora
Connection Cost No. from Sample Percent
0 62 43 900-4000 23 15 4001-6000 18 12 6000-8000 13 9 8001-10000 19 13 10001-15000 7 5 15001-20000 3 2 20000+ 1 1 Total 146 100
Source: Field Data
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
disbanded after the first room was constructed. Most of the respondents involved in building groups did not do the actual labour but contributed money to each other. This meant that they had to hire labour as well, and thus did not reduce the cost of construction. The project, however, created jobs for many people in the City. This was especially true of both skilled and unskilled workers who had jobs readily available to them due to this project.
4.4 Housing Conditions in the Dandora Project
This section discusses the number of rooms, the level of services and the number of adults and children using these services in Dandora. The number of rooms include those originally meant to be kitchen facilities. These are included because most of the respondents who had rented kitchen facilities did not have any other rooms but used the rented room for all purposes. The households that had kitchen facilities did not use them exclusively for cooking but had part of the family sleeping there as well. Few of the respondents had extra rooms to use as living rooms. Those who had one room shared it with their children, slept, cooked and ate in the same room.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
15 4
4.4.1 Number of Rooms Occupied by the Respondents in Dandora
The number of rooms occupied by respondents ranged from one to seven. According to Figure 14, a high percentage of respondents occupied one room. Those living in one room comprised 54 percent of all the respondents while 20 percent had two rooms, 2 2 percent had three to four rooms and the remaining 4 percent had five to seven rooms. The plot owners had more rooms than tenants. Thus, the number of plot owners with one room accounted for 24 percentage of all the plot owners. Those with two rooms comprised 27 percent while 42 percent had three to four rooms. The remaining 9 percent had five to seven rooms. Unlike the plot owners, almost all the tenants had rented one room. Thus 84 percent rented one room, 13 percent had two rooms, while 3 percent had three to four rooms. The number of households living in one room have predominated the low-income housing sector in Nairobi since, before independence, it was the policy to build one-room dwelling units. Although this housing policy was abolished after independence, the few housing units constructed for low income households and the increasing number of households has made this the only alternative left for them. Their low income has meant that most of them can only afford to rent one room.
The number of persons per household ranged from one to twenty-one. The average number of persons was 4.9, which is
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
R eproduced
with perm
ission of
the copyright
ow ner.
F urther
reproduction prohibited
w ithout
perm issio
n .
Figure 14
Number of Rooms Occupied by the Respondents
Percent 100
80
60
40
20
0 sm- 3
Number of Rooms
111 5-7
General Owners Tenants
Source: Field Data i— *
ci w
1 5U
the same as the one estimated by the City Council for the City for 1970s. The households with one to two persons include 31 percent of the households. Those with three to five persons include 32 percent of the households. A high 29 percent had six to ten persons while the remaining 8 percent had eleven to twenty one persons (see Table 4.11). A total of 54 percent of the respondents had one room and the number of persons living in that one room ranged from one to thirteen. A total of 2 5 percent of all the respondents with one room had one to two persons; 2 3 percent had three to six persons; 4 percent had seven to ten persons while the remaining one percent had eleven to thirteen persons. The respondents with two rooms comprised of 2 0 percent of all the respondents. These households had 1-12 persons. Thus, 4 percent had one to two persons; 14 percent had three to twelve persons. The households with 3-4 rooms accounted for 22 percent and these were occupied by one to thirteen persons. Those with 5-7 rooms had 5-12 persons. The above figures show high occupancy rates for all the rooms. Crowding was defined by respondents as a situation where more than one person occupied a room. The exception was the situation where the married respondent lived with their spouse in the urban areas. An assessment of all those married living with their spouse in the urban areas showed that all had more than two people per room except 4 percent. Given the low percentage of those living with their wives having enough room, this study is using one person per room as adequate room. A total of 2 4 percent of all the
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1 5 7
Table 4.11 Number of Persons per Room in Dandora
No. of No. from Number of No. from Rooms Sample Percent Persons Sample :Percent
1 161 54 1-2 77 25 3-6 69 23 7-10 13 4 11-15 2 1
2 59 20 1-2 13 4 3-6 32 10 7-10 10 3 11-21 4 2
3 36 12 1-2 ■% 1 3-6 14 4 7-10 15 5 11-15 4 2
4 30 10 5-6 6 2 7-10 15 5 11-21 9 3
5-7 13 4 1-6 4 2 7-10 5 2 11-21 4 2
Total 299 100 299 100
Source: Fielr' Data
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1 5 8
respondents in Dandora had adequate room. This accounted for 21 percent of respondents who had one room per person while the remaining 3 percent had two to six rooms.
4.4.2 Available Services in the Dandora Project
The Dandora project provided the respondents with shared facilities. Each plot had water, toilet and shower facilities which were shared by an average number of thirteen persons. The other services discussed here include electricity and kitchen facilities. Although electricity was brought to the site, each individual plot owner had to get electricity connected to their plots. This has proven to be expensive, and not all the respondents can afford it. Thus, 46 percent of the respondents had electricity while 54 percent did not. Kitchen facilities were provided on the plot but these may have been rented out. If the tenants wanted a kitchen, they had to rent it separately. This explains the reason why most people did not have this facility. For the tenant, it meant an added expense while it meant extra income for the plot owners. A high percentage of 79 percent had no kitchen facilities while 21 percent did. Most of the respondents with kitchen facilities were plot owners. Thus, 36 percent of all the plot owners had kitchens while only 6 percent of all the tenants did. The lack of kitchen facilities meant that people cooked m the room they slept in. This is especially true given the fact that mos'i' of the respondents occupied one room.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1 5 9
Thus, 94 percent of all the tenants cooked in the room where they slept, compared with 64 percent of the plot owners. However, it is worth noting that those who had kitchen facilities did not use them exclusively for cooking purposes. Most of these were used as bedrooms for children or other members of the household. Although the plot owners had more rooms, they opted to sublet them so as to get extra income to enable them to repay their loans to the City Council. Except for 8 percent, all the plot owners had sublet some of their rooms. According to Table 4.12, the plot owners were subletting a high percentage of their rooms in Dandora. The average number of rooms which the plot owners were subletting was 3.4, with 17 percent subletting one to two rooms and using the rest of the rooms for the family. A high 54 percent had sublet three to four rooms while 21 percent had sublet five to eight rooms. Those subletting more than five rooms usually occupied one room or had built more rooms than originally allowed.
The income earned from subletting ranged from Kshs. 2 50- 2,950, as indicated in Table 4.13. These figures show a significant amount of income from subletting. The amount of rent charged depended on the level of finish, the number of rooms rented and whether the plot had electricity or not. The plots with proper gates charged more because they were said to be more secure. The location of the plot also influenced the amount of rent charged. The plots in phase 1
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
160
Table 4.12 Total Number of Rooms the Plot Owners
were Subletting
Number of Rooms Sample Percent
0 12 8 1-2 26 17 3-4 81 54 5-6 25 17 7-8 6 4 Total 150 100
Source: Field
Total Income
Data
Table 4.13 from Subletting in Kshs.
Total Income Sample Percent
0 12 8 300 or less 2 1 301-500 5 3 501-750 24 16 751-1000 31 21 1001-1500 48 32 1501-2000 19 13 2001-2950 9 6 Total 150 100
Source: Field Data
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
16 1
were cheaper because they were not as well developed as the ones in phase 2.
4.4.3 Monthly Housing Cost in the Dandora Project
The housing expenses paid by the respondents ranged from those paying less than Kshs. 100 to those paying Kshs. 1,500. Rents paid by the tenants included the cost of services while the plot owners included repayments to the Nairobi City Council and the cost of services (see Table 4.14). Housing cost for most respondents were high compared to the City Council housirg. Except for two plot owners who had bought their plct from the original owners, all the others had to make monthly repayments to the City Council. These repayments ranged from Kshs. 100-434.
All the Dandora residents had water, toilet and bathroom facilities. These services were shared, so there is only one of each per plot. Although every plot had water, the plot owners had to pay for this every month. The water bills were the high 1 _n the City; this was mainly due to the high number of persons per plot, an average of thirteen persons. The monthly payments for water ranged from Kshs. 40-740. The average payment per month was Kshs. 285, which is higher than the average plot charges which was Kshs. 205.
The monthly payment for electricity was paid by the plot owners, although this was part of the rent charged per room. This payment ranged from Kshs. 40-300 with 25 percent paying
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 4.14 Total Housing Cost Incurred by Dandora
Residents
Housing Cost No. from Sample Percent
100 or less 1 0.3 101-200 10 3 201-300 50 17 301-400 81 27 401-500 62 20 501-1000 91 31 1001-1500 5 1.7 Total 300 100
Source: Field Data
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
16 3
Kshs. 4 0-100, 42 percent Kshs. 101-200, while one percent paid Kshs.300. The average monthly payment were Kshs. 51. As can be seen from Figure 15, the housing expenses for the plot owners increased significantly when all the other housing expenses were included. Thus, while 56 percent paid Kshs. 200 or lass, the percentage decreased to one percent when all the other expenses were included. A total of 38 percent of the plot owners paid Kshs. 201-300 to the City Council. This percentage decreased to 3 percent after the other expenses were included. Only 5 percent of the \ lot owners had plot repayments of over Kshs. 300. This compares with 95 percent when all the other expenses are included. One thing worth noting here is the fact that many of the plot owners were still repaying the loans they had accrued from other sources
The tenants' housing cost, which included services, ranged from Kshs. 100-1,500. The average rent paid was Kshs. 399. The tenants paying up to Kshs. 300 included 35 percent ot the tenants. The highest percentage of 53 percent paid K^hs. 301-500 and the remaining 12 percent paid rent which ranged from Kshs. 501-1,500. *4ost of the tenants paying Kshs. 2 50 or less had rented a kitchen whose rent was lower than the rest of the rooms. The housing expenses of the tenants were lower than those of plot owners (see Figure 16).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
R eproduced
with perm
ission of
the copyright
ow ner.
F urther
reproduction prohibited
w ithout
perm issio
n .
Figure 15 Housing Expenses for the Ow ners
e r c e n t 100
80
60
40
20
0 1 0 0 o r l e s s 1 0 1 - 2 0 0 2 0 1 - 3 0 0 3 0 1 - 4 0 0 4 0 1 - 5 0 0 5 0 1 - 1 0 0 0 1 0 0 1 - 1 5 0 0
P a y m e n t in K s h s .
R epayment to NCC J E l e c t r i c i t y Cost Cost of Water
S o u r c e : F i e l d D a t a
CD
R eproduced
with perm
ission of
the copyright
ow ner.
F urther
reproduction prohibited
w ithout
perm issio
n .
Figure 16 Comparison of O w n e rs ’ and Tenants’
Housing Expenses
P e r c e n t 60 50 40 30 20 10
0 i _ _n 100 or less 101- 200
I 7.1 2 0 1 - 3 0 0 3 0 1 - 4 0 0 4 0 1 - 5 0 0
H o u s i n g E x p e n s e s in K s h s .
Hi O w n e r s L I T e n a n t s
5 0 1 - 1 0 0 0 1 0 0 1 - 1 5 0 0
S o u r c e : F i e l d D a t a
165
166
4.4.4 Reasons for Moving to the Dandora Project
The major reason for moving to Dandora for most of the landlords was the allocation of the plot. A total of 44 percent of all the households stated this as the reason why they moved to Dandora (see Table 4.15). The project provided them with much better housing than they had before. This was also cheaper in the sense that they did not need to pay rent, except the City Council repayment, which they could get from subletting part of the plot. For the majority of them, it as a source of income now and in old age. The second reason given by 17 percent of the respondents, was the fact that Dandora provided cheap housing compared to other available standard housing. This was especially true given the fact that many had shared their accommodation with others before. The Dandora project provided the needed single rooms that they could r^nt on their own. The third reason given by 14 percent of the respondents, was the availability of rooms to rent in Dandora. Some of the respondents stated that one could find a room in Dandora in a day. This was not the case anywhere else in the City.
Another reason was privacy and the need for more space (6 percent), since most had shared their accommodation with other people before moving to Dandora. A total of 6 percent of the respondents moved because of job or business opportunities in the Dandora. The establishment of such a large settlement created jobs for many people in the informal
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 4.15 Reasons for Moving to Dandora Project
Reason No. from Sample Percent
Got a Plot 128 43 Cheaper Housing 50 17 Housing Available 42 14 Needed their own Place 19 6 Job Opportunities 17 5 Transportation Convenience 11 4 Better Housing 8 3 Other 25 8 Total 300 100
Source: Field Data
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
16 8
sector. This is evident in the project as one finds all kinds of self-employed people in the settlement. Also, five market stores were opened up in this area, but due t'-* the high rents required by the City Council, most of the businesses are outside these designated markets. Many other businesses, like dry cleaning facilities, tailoring, bars, vegetable kiosks, shops etc. have been opened in the settlement. Some of the respondents moved to Dandora because the project provided better housing (2 percent) and transportation convenience (4 percent). For some, the project was either near their place of work, or they could easily get transportation to and from their place of work. Given the high concentration of people in this settlement, public transportation is more frequent than in many othnr settlements in the City. This does not imply that there is ample transportation in Dandora but that there are more buses and Matatus in this settlement than in any other part of the City. The high concentration of people also led many of the employers to provide free transportation for their employees. Other reasons for moving included formation of new households (one percent) and safety (one percent). Dandora plots had large iron gates and locks, unlike the places that the respondents had lived in before. Also, with so many people per plot, some were bound to be home during the day when most of the thefts took place.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1 6 9
4.5 Respondents' Housing Conditions Before Moving to Dandora
Respondents' housing conditions prior to the move to Dandora are described in this section and compare^ in the next section in order to assess who benefited from these projects and the degree to which housing conditions improved as a result of the move. The variables discussed include the places they moved from, the house types, the nimber of persons per household, the ni. ber of rooms and the level of services.
The respondents in Dandora moved either from different locations in Nairobi or from the rural areas. A small proportion, 13 percent, moved from rural areas, while a majority of 87 percent had lived in different parts of Nairobi (see Table 4.16). The highest concentration of households was in Mathare Valley, one of the oldest squatter settlements in Nairobi. This accounted for 13 percent of all the respondents. This settlement is located in the eastern part of Nairobi and is near the City centre. The settlement is said to be the largest squatter settlement in Nairobi. It consists of nine separate villages and covers approximately 157 hectares, stretching some four kilometres along the Nairobi riverbank (Obudho 1937, 104). At present, there are a number of re-development projects, but these only cover a small area. The area lacks basic infrastructure and most facilities have been provided on self-help basis. It is
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
17 0
Table 4.16 Location of Housing Estates the Residents Lived
in Before Moving to Dandora
Location Sample Percent
Mathare Valley 38 13 Kariobangi 21 7 Eastleigh 18 6 Huruma 15 5 Outskirt of the City 13 5 Buruburu 10 3 Muthurwa/Landmawe 10 3 Jericho 9 3 Bahati 9 3 Pangani/Ngara 9 3 Kwa Michael 8 3 Kibera 8 3 Makadara 7 2 Korokosho 7 2 Umoja 7 2 Dandora/Ofafa Maring 6 2 Ofafa Maringo 6 2 Other Locations 61 20 Other Urban 8 3 Rural Areas 30 10 Total 300 100
Source: Field Data
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
17 1
estimated that 7 0,000 to 100,000 people live in Mathare Valley.
The next highest concentration was in Kariobangi, where 7 percent of the respondents lived before moving to Dandora. The area is near Mathare Valley and has many different types of housing. There is both permanent and temporary housing. A total of 6 percent moved from Eastleigh. The history of this settlement dates back to around 1920 when the land was privately bought and developed for Asian families without much infrastructure. In 1925 it was partly bought by the government to control the infrastructure development. The settlement received its final shape around 1925. This included street patterns and court-yard dwelling type for extended Asian families. Now these are mostly private tenements rented out per room to families and bachelors who find work locally in small industry and commerce or the nearby industrial area ana the Central Business District. This area has shown the highest rate of room occupancy in Nairobi, due originally to the size of low-income Asian households. The gradual shift of these families away from Eastleigh has resulted in the extensive subdivision of the exiting housing, and the infill structures. Deteriorating conditions are due to the fact that no additional services or facilities have accompanied the increased number of dwellings.
The next highest percent was from Huruma, where 5 percent of the respondents had moved from. This settlement had both formal and informal housing. One of the City Council’s new
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
estates is also in this settlement. The remaining 69 percent were spread in smaller percentages all over the City. Thus 4 percent lived at the outskirts of the City, 3 percent lived in Buru Buru (private middle income estate) and Muthurwa and Land Mawe (housing provided by the Railway Administration constructed before independence), 9 percent lived in Jericho, Bahati, Pangani and Ngara, 6 percent lived in Kibera and Kwa Michael, while 7 percent lived in the Umoja, Makadara and Ngomongo/Korokosho area. A total of 12 percent lived in Majengo/ Pumwani/California, Ofafa Maringo/Shauri Moyo, elsewhere in Dandora, the Kileleshwa area, Kenyatta, River Road, Westlands, Embakasi and in different hostels in the City; 2 percent lived in Jerusalem, Dolnholm, Starehe and Riruta Satellite, 3 percent lived in Kawangware, Kayole, Lower Kabete, Langata, Madaraka, Gikomba, Uhuru, Grogon Road, Highridge, and Pioneer Estate while 7 percent lived in Makongeni, Kaloleni, Mbotela, Kangemi, City Centre and South B/Nairobi South. Most of these are City Council estates and were constructed just before Independence (see Figure fc) . The remaining 12 percent had come from either the rural areas (10 percent) or other urban areas 3̂ percent).
The housing units respondents lived in before moving to Dandora were developed by public, formal private and informal private sectors. Public housing was developed by the Nairobi City Council and the Central Government and was built with permanent materials. Formal private developers include both individuals, housing finance corporations and other employer-
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
provided housing which was built according to prescribed building standards. Housing in the informal private sector was developed by individuals and housing companies. This housing was substandard and was mainly constructed with temporary building materials. The materials used varied from one unit to another, with some of the units being constructed with permanent materials but still considered temporary because they were not approved by the City authorities. According to 1987 estimates, one-third of Nairobi's population live in this type of housing. Most of the housing in the rural areas is in this category because local materials were used for construction. From Table 4.17, 26 percent of all the households lived in public and employer-provided housing. The highest percentage, 3 6 percent, came from formal private housing while 40 percent either moved from squatter settlements (27 percent), rural areas (10 percent), or other urban areas (3 percent). The housing units the respondents lived in before moving to Dandora were built with both permanent and temporary materials. That is, 65 percent of all the units discussed above were built with permanent materials, while 35 percent were built with temporary materials.
The units from which the households moved were either rented, owned, or provided by the employer. The highest percentage of respondents lived in rented accommodation. Those living in rented accommodation accounted for 58 percent of all the households. A total of 14 percent owned the housing units they had occupied before moving to Dandora and
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
17 1
Table 4.17 House Types the Residents Lived
to Dandora in Before the Move
House Type Sample Percent
Nairobi City Council 38 13 Government Housing 18 6 Employer-Provided 21 7 Privately Rented 105 35 Squatter Settlement 80 27 Rural Areas 31 10 Other Urban 4 3 Total 297 100
Source: Field Data
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
175
these were mainly from the rural ar*. s. The remaining 20 percent lived with other people before moving to Dandora and did not know whether the housing was owned or rented. The average number of persons per household was 5, which is the same as the current one in Dandora. The total number of persons per household ranged from one to twenty-one. The number of persons per household was distributed as follows: 14 percent had one person, 19 percent had two persons, 17 percent had three to four persons, 21 percent had five to six persons, 2 3 percent had seven to ten persons while the remaining 6 percent had eleven to twenty-one persons.
The number of rooms the respondents occupied before the move to Dandora ranged from one to seven rooms. The highest percentage of 61 percent lived in one room before moving to Dandora (see Table 4.18). This was followed by 17 percent of residents who had two rooms. Those with three to four rooms accounted for 18 percent. The remaining 3 percent had five to seven rooms. A total of 19 percent of all the respondents had adequate space before moving to Dandora. Some people, however, lived in housing constructed with temporary materials. Most of those not crowded had lived in one-roomed units as well. These accounted for 13 percent of all respondents witn adequate housing. The highest percentage (81 percent) of all the respondents were crowded. The level of crowding was highest among the respondents with one room (61 percent). Thus, 2 9 percent had one to two persons per room, 2 3 percent had three to six persons f *r room while the
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1 7 6
Table 4.18 Number of Persons per Room Prior to the Move to Dandora
No. of No. from No. of No. from Rooms Sample Percent Persons Sample Percent
1 183 61 1-2 85 29 3-6 69 23 7-10 21 7 11-15 8 3
2 52 17 1-2 10 3 3-6 22 7 7 -10 17 6 11-21 3 1
3 37 13 2-6 18 6 7-10 18 6 11-15 1 0.4
4 18 6 1-6 5 1.6 7-10 9 3 11-20 4 1
5-7 8 3 5-6 3 1 7-10 2 1 11-15 3 1
Total 298 100 2^3 100
Source: Field Data
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1 7 7
remaining 10 percent had seven to fourteen persons per room. The respondents with two rooms (18 percent) were less crowded. Thus 3 percent had one to two persons, 7 percent had three to six persons while remaining 6 percent had seven to twelve persons. The remaining 22 percent of all the respondents had three to seven rooms. A total of 3 percent of these were not crowded and had one to five persons. The remaining 18 percent were crowded and some of the households had as many as twenty persons. The above figures show high level of crowding before the move to Dandora. This crowding was experienced in both standard and squatter housing. The number of rooms increased slightly with the move to Dandora but the size of households have been increasing at a faster rate, thus making crowding levels high in Dandora as well. The difference in Dandora is the fact that all the respondents had permanent housing and had the required minimum services while some of the respondents did not nave these facilities before moving to Dandora.
Housing cost before moving to Dandora was minimal with a high 37 percent paying no rent. This was because they either lived with relatives or friends or moved directly from the rural areas (see Table 4.19). The average rent paid per month was Kshs. 184 with the percentage of 3'> paying no rent while 43 percent paid Kshs. 15-300. This relatively low rent is explained by the fact that these were paid a few years ago, and some of the respondents had moved from the squatter areas which have the lowest rents in the City. A total of 19
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
17 8
percent paid Kshs. 300-2,700. The housing rent in most cases included payment for services like water and electricity.
Not all the respondents had the basic services before moving to Dandora. As can be seen from Table 4.20 the present Dandora respondents had most of the services before moving to Dandora. Only 7 percent had no water at all while 5 percent had no toi’.et facilities. A high 30 percent had no showers, but stated that the toilet was usually used for both purposes. Kitchen facilities were not available to 61 j crcent of the respondents, while electricity was not available to 56 percent (see Table 4.20).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1 7 9
Table 4.19 Rent Paid per Month
Rent Sample Percent
0 111 37 15-10 58 19
101-20 4 * 15 201-30 29 9 301-40 20 7 401-50 12 4 501-10 20 7 1001-2 3 2 Total 297 100
Source : Field
Table
Data
4 .20 Level of Services Before Moving to Dandora
Service Yes
SamplePercent No Sample Percent Outside Inside
Water 278 93 22 7 73 17 Toilet 284 95 16 5 76 19 Shower 210 70 90 30 52 18 Kitchen 117 39 183 6.1 Electricity 133 44 167 56
Source: Field Data
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1 8 0
4.6 Evaluation of Bousing Conditions as a Rssult of tha Move to Pandora
This section discusses the four different measures which were used as the housing quality index to assess improvement in housing conditions for both the tenants and the plot owners. The four measures are: the number of persons per room, the housing cost, the building materials and level of services. These four measures were included in the questionnaire in which the respondents were asked to rank them according to the order of importance. The ranking was from one to seven, with seven representing the most important ranking while one represented the least important. Using Likert scales, the residents responses were recorded. These were added to determine what the most important components of housing were, according to the respondents (see Table 4.21). The respondents ranked water and toilet facilities as the most important housing components, followed by shower facilities. These were followed by a building constructed with permanent materials followed by electricity and a kitchen.
To determine the acceptable number of persons per room, the respondents ic’antified one room per person as adequate space for single households. Where the respondents were married, the adequate number of persons per room was identified as two. The survey revealed a small number of respondents who lived with their spouses in the urban areas and had adequate room. This comprised 4 percent of all the
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
18 1
Table 4.21 Ranking of Housing Components
Least Important Most Important Component 1 2 3 4 5 6 7
Water 0 1 2 3 2 0 292 Toilet 1 1 2 3 2 0 291 Shower 5 1 1 3 4 * 1 285 Permanent Stucture 4 1 0 41 6 2 246 Electricity 15 1 1 0 61 0 216 Kitchen 16 0 4 63 5 2 210
Source: Field Data
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1
respondents. This is mainly because most of the respondents had children or relatives living with them, and they mainly lived in one room. The respondents would also have preferred to have a separate room where they could do all the other household functions. However, this was not considered in deciding what adequate room was. This is mainly because most respondents live in one room in the City. Those living in one room at the time of independence comprised 80 percent of all the City residents. The situation has not improved significantly due to the few number of housing units constructed since independence. This is especially true among low-income housing, where housing production has been slow in spite of the rapidly increasing number of low-income households. Given all these aspects, one room per person was chosen as adequate housing. Thus, all the households with more than one person per room will be regarded as crowded. Housing quality was reflected by the building materials used for construction. Almost all the respondents living in unauthorized housing live in temporary or semi-permanent housing. The quality of these units differs and the amount of rent depends on the construction materials used. The building materials used in Dandora and before moving to Dandora were different. While all the plots in Dandora were built with permanent materials, only 65 percent were built with permanent materials before the move to Dandora. Thus, 35 percent of the respondents who lived in temporary housing before the project had housing of a higher standard than
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1 8 3
before. Those living mainly in temporary housing included respondents who had either moved from the squatter settlements or the rural areas. Most of the housing that respondents moved from was rented. Ownership was only limited to 14 percent of the respondents. Most of those owning housing before moving to Dandora were from the rural areas. Their previous housing was mainly substandard and lacked the basic services. The move to Dandora enabled 50 percent of all those interviewed to own permanent housing with all the basic services.
All respondents had water, toilet and shower facilities in Dandora. These were, however, shared by all the people on the plot and were outside the rooms. Although not all respondents had these facilities before moving to Dandora, some of those who had the above services did not share with other households. The respondents had the services inside their room, unlike in Dandora. Those with services inside accounted for 19 percent with water and toilet facilities while 18 percent had shower facilities. The move to Dandora for these respondents meant a lack of privacy, as they had to share these facilities with others who were not part of their household. Most of the respondents had water and toilet facilities before moving to Dandora. As indicated in Figure 17, only 7 percent of all the respondents did not have water before moving to Dc dora, this compares with 6 percent of the respondents who did not have toilet facilities. The project was of major benefit to these respondents because it provided
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
R eproduced
with perm
ission of
the copyright
ow ner.
F urther
reproduction prohibited
w ithout
perm issio
n .
Figure 17 Level of Services Before and After
Moving to Dandora
P e r c e n t 120
100
80 60 40 20 0
W a t e r T o ile t S h o w e r S e r v i c e s
K i t c h e n E l e c t r i c i t y
S e r v i c e s B e f o r e S e r v i c e s in D a n d o r a
Source.- F ie ld D a t a 184
1 8 5
them with the most important services which they had lacked before. Except for 30 percent of the respondents, all the others had shower facilities before moving to Dandora. Kitchen and electricity facilities were the least available to the respondents, before and after moving to Dandora. Kitchen facilities are almost non-existent among the low- income households because they represent an added expense. The room-based, low-income housing market rarely provides housing with kitchen facilities. Those witheut kitchen facilities before moving to Dandora comprised 61 percent. A total of 39 percent had kitchen facilities before moving to Dandora. In Dandora, only 21 percent of all the respondents had kitchen facilities. This accounted for 9 percent who had these facilities before and 12 percent who did not. A total of 30 percent of all those with kitchen facilities before moving to Dandora did not have these in Dandora. The high percentage of those lacking kitchen facilities was due to the fact that the tenants could not afford to rent them. Although each plot had a kitchen, the plot owners would sublet it and earn extra money. Many preferred to sublet the kitchen than use it, because they felt that they could cook and sleep in the same room. The respondents had ranked kitchen facilities as the least important. When asked whether they would prefer an extra room or a kitchen many said that they would prefer a room because it was bigger. Using one room for cooking and sleeping purposes did not seem to be a problem with most people. The apparent lack of enough room for those in the low
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1 86
income areas made availability of kitchen facilities a luxury they could not afford. Electricity ranked second last. It was a convenience many would have liked to have but realised that they could not afford. A total of 56 of the residents had no electricity before moving to Dandora, compared with 54 percent in Dandora. High electricity connection charges made it too expensive for plot owners in Dandora.
The number of rooms per household ranged from one to seven. The number of rooms before Dandora were however constructed using both permanent and temporary materials. The number of rooms per household in Dandora increased slightly as shown in Table 4.22. Before moving to Dandora, 61 percent of all the households had one room. This decreased to 54 percent giving 8 percent more rooms in Dandora. The added rooms meant more space for the respondents and better housing than they had before. Those with two rooms increased by 2 percent; those with three to four rooms by 4 percent while those with five rooms increased by 2 percent. Thus 15 percent of all the respondents had more rooms in Dandora than before.
The number of persons per household ranged from one to twenty-one with the average household size being 4.9 persons before and after moving to Dandora. The number of households with one person increased ir Dandora by 8 percent. Although the number of households with one person increased with the move to Dandora, the number of persons per households did not change significantly. The number of persons per household are
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1 8 7
compared to the number of rooms to assess whether these households had more rooms available to them in any of the locations in Table 4.23. A total of 61 percent had one room before moving to Dandora compared to 54 in Dandora. The single rooms with one to two persons before moving to Dandora comprised 29 percent. This compares with 26 percent in Dandora. Those with three to four persons accounted for 12 percent compared to 17 in Dandora. The remaining 21 percent had five or more persons before moving to Dandora. This compares with 18 percent in Dandora who had five or more persons per room. The one-roomed units were more crowded in Dandora than before the move to Dandora.
Those with two rooms comprised 17 percent before Dandora and 20 percent in Dandora. A total of 6 percent had one to four persons before moving to Dandora, while 8 percent had the same number of persons in Dandora. The remaining 11 percent before Dandora and in Dandora had more than five persons for two rooms. The households with three rooms had 6 percent of the rooms with three to six persons before moving to Dandora. These decreased to 5 percent of the households with the same number of persons in Dandora. The remaining 6 percent had 7- 15 persons before moving to Dandora. These had the number of persons decrease to 6 percent. The households with four rooms or more experienced very slight changes with the move to Dandora.
A total of 19 percent had adequate room before moving to Dandora. This accounted for 16 percent who had one room per
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1 Bd
Table 4.22 Number of Rooms Before and After Moving to Dandora
Number Before Dandora Dandora o f Rooms No . Sampled Percent No. Sampled Percent
1 183 61 161 54 2 52 18 60 20 3-4 55 18 66 22 5-7 9 3 13 4 Total 299 100 300 100
Source: Field Data
Table 4. 23 Comparison of the Number of Persons per Room
and After the Move to Dandora Before
Before Dandora Dandora Persons No . Sampled Percent No. Sampled Percent
1 43 14 68 23 2 57 19 25 8 3-4 50 17 66 22 5-6 62 21 59 20 7-10 67 23 58 19 11-21 19 6 24 8 Total 298 100 300 100
Source: Field Data
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1
person while 3 percent had two to three per room. Those with adequate room in Dandora comprised 26 percent of all the respondents. A total of 23 percent of these had one room, while 3 percent had two rooms with the remaining one percent having three to six rooms. According to these figures, 7 percent of the respondents benefited by moving to Dandora where they had adequate room, unlike in their previous residence. It is worth noting that 35 percent of all the housing before moving to Dandora was constructed with temporary materials unlike in Dandora where all the housing was constructed with permanent materials.
Housing costs included the monthly rent and the cost of services. The move to Dandora made the housing costs increase significantly as shown in Figure 18. A total of 37 percent of all the respondents who paid no rent before moving to Dandora paid rents ranging from less than Kshs. 100 to Kshs. 1,500. The increase was due to the increased housing cost for plot owners who not only had to make repayments to the City Council, but also had to pay for water ar»d electricity. Since most people rented one room, the number of households living in one plot increased significantly, thus making the cost of services high. Most of the respondents had lived with relatives or friends and did not pay any rent before moving to Dandora. Only household heads were interviewed in this survey, and all had to pay for their housing in Dandora. Some of the respondents also had moved to Dandora from the rural areas, and no rents are paid for
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
with perm
ission of
the copyright
ow ner.
F urther
reproduction prohibited
w ithout
perm issio
n .
CD ■o-5O Q.C oCD Q.
Percent 40
30
20
10
Figure 18 Rents Before and After the
Move to Dandora
|» R l 15-100 1 0 1 -2 0 0 2 0 1 -3 0 0 3 0 1 -4 0 1 4 0 1 -5 0 0 5 0 1 -1 0 0 0 1 0 0 1 -2 7 0 0
B e f o r e M o v e C l J A f te r M o v e
S o u r c e : F ie ld D a t a
191
housing in the rural areas. Those paying up to Kshs.300 comprised 44 percent before Dandora and 20 percent in Dandora. The low rents before moving to Dandora are explained by the fact that most of those paying the low rent had either lived in squatter settlements or shared their rent with others. Those paying over Kshs. 300 accounted for 19 percent before Dandora and a high percentage of 80 percent in Dandora. This clearly indicates increased housing costs for those who moved to Dandora, where all the respondents had high housing expenses either in the form of rents or repayment to the City Council. The expenses analyzed here do not include the repayments for construction other loans borrowed from cooperatives or other sources. If loan repayments were included the housing costs would be much higher that the figures displayed above.
4.7 Conclusions
The beneficiaries of the Dandora project included both male and female household heads, most of whom were married and lived with their spouse in Dandora. A significant number, however, had left their wives and children in the rural areas because they either could not afford housing in Nairobi or did not have access to family housing. Most of the respondents worked in both the formal and the informal sector and had stable incomes. Most of those employed in the formal
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
192
sector worked in the City centre, industrial area or further away. These mainly used public and employer-provided transportation to get to their place of work. Host of the respondents living in Dandora or one to five kilometres away from Dandora walked to their place of work. Only a small percentage had their own cars. Most of the beneficiaries earned higher incomes than specified by the project. This left the 30th percentile of low-income urban housing without the benefit of low cost housing as had been intended.
The high building costs made housing inaccessible to the low-income groups who could not afford the cost of construction. The assumptions made on the availability of owners to provide labour and reduce housing costs ignored the realities of low-income households who spend most of their time working in the informal sector and would therefore not have time to work on the plots. The project however allowed progressive development; that provision helped some people who would not have benefited from other types of housing. The project also created many jobs for both skilled and unskilled workers. The main beneficiaries of this project were the tenants, most of whom were paying high rent due to the scarcity of standard housing in the City. The project thus provided rooms for rent, which is the major type of housing required by low and low-middle income households. The project also made services available to respondents who did not have these services before. It did, however, increase the housing cost for all the respondents. The level of crowding seems
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1 9 3
to have been least affected, although the respondents had more rooms in Dandora. The increasing size of households in both the City and the country is reflected in this settlement. The large households also show the ever-increasing housing need, which forces many to share their rooms with relatives or friends or live in sguatter settlements.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER 5
EVALUATION OF HOUSING BENEFITS AS A RESULT OF A DIRECT MOVE TO DANDORA FROM THE SQUATTER SETTLEMENTS
5.l introduction
This chapter examines whether the beneficiaries of the Dandora project came directly from the squatter settlements or from the rural areas. The project was intended for households which were living in the squatter settlements. It was hoped that the implementation of such a project would directly minimize the further spread of the squatter settlements. This was because the housing provided so far had not benefited most of the urban dwellers and the squatter settlements were increasing at a rapid rate. The. survey showed that those benefiting directly from the squatter settlements included a small percentage of 27 percent. This chapter also discusses the households that moved directly from the rural areas to Dandora. These comprised 13 percent of all the respondents and were not among the intended beneficiaries of this project. Chapter 6 assesses whether the project drew groups from better housing outside of the squatter settlements. In cases where the project drew people from better housing, the study assessed whether, through the workings of the vacancy chain, housing was left behind that could have benefited residents from squatter settlements,
1 94 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1 9 5
inferring from the migration patterns of the current Dandora residents, the degree to which the second tier moves from the squatter settlements have accrued.
The different moves made by the respondents are shown in Figure 19. A total of 27 percent of all the respondents living in Dandora had lived in the squatter settlements immediately prior to the move to Dandora. Some of the household heads had moved from the rural areas to the squatter settlements and then to Dandora, while others had moved from the rural areas, to standard housing, and then to squatter settlements before moving to Dandora. A total of 52 percent1 of all those moving from squatter settlements to Dandora had initially lived in the rural areas. Those moving from the rural areas to standard housing before coming to the squatter settlements comprised 32 percent of all the respondents moving from squatter settlements. The remaining 16 percent had been born in the squatter settlements. The low number of those born in Nairobi is due to the restrictions placed on permanent residency of African populations which was enforced by the colonial government prior to independence in 1963. Most people came to the City to work, and eventually moved back to the rural areas. Although this has changed somewhat, most people maintain close ties with their rural homes and do not intend to retire in the City even though they have worked
1 These percentages refer only to the respondents who had moved directly from the squatter settlements (a total of 27 percent of all those moving to Dandora).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1 9
Figure 19
MIGRATION PATTERNS OF THE DANDORA RESIDENTS
Rural Areas
27 f. Squatter Areas
DANDORA
Standard Housing
This represents the move prior to Dandora
This represents secondary moves
A total of 7% were not included in these calculations because they were born in Nairobi
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1 97
there all their lives. Other respondents who moved to Dandora directly had come
from the rural areas. A total of 13 percent had moved directly from the rural areas to Dandora. As will be shown later, most of these were tenants, and had moved to the project long after it was implemented.
Most of the beneficiaries had moved from standard housing to Dandora. This accounted for 60 percent of all the respondents moving to Dandora. A total of 43 percent of these respondents had moved from the rural areas to standard housing and then to Dandora. Those moving from the squatter settlements to standard housing and then to Dandora consisted of 10 percent. The remaining 42 percent had moved from the rural areas to standard housing in different locations, and then to Dandora. These households are discussed in Chapter 6.
The housing condition of the respondents who moved directly from the squatter settlements to Dandora is shown in Figure 20. Those moving from the squatter settlements comprised 27 percent of all the respondents, compared to 13 percent who moved from the rural areas. While a direct move from the squatter settlements was one of the stated objectives of those initiating the Dandora project, a direct move from the rural areas was not among the original objectives2.
One of the major eligibility criteria was that the applicant should have lived in the City for a period of not less than two years.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
FlflU?s20
IMG RATION HISTORIES OF THE RESIDENTS MOVMQ FROM THE SQUATTER SETTLEMENTS TO DANDORA
100% (80 rssidants) Dandora
I6S ( 1 3 )
Standard Housing
Bom In tha Squattar Sattlam ants
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1 9 9
The study established that the respondents moving directly to Dandora from both the squatter settlements and the rural areas had their housing improve significantly in terms of having permanent housing and better services, other changes experienced as a result of the move were increased cost of housing and increased number of persons per room, thus raising crowding levels.
5.2 Socio-economic Characteristics of the Respondents Moving from the Squatter Settlements to Dandora
Those moving directly from the squatter settlements made up 27 percent of all the respondents. As can be seen from Figure 20, a total of 16 percent of all the respondents moving from squatter settlements to Dandora had been born in squatter settlements. The highest percent (42 percent), however, had moved from the rural areas to squatter settlements and then to Dandora. This situation represents moves made by the majority of new migrants, most of whom live in the squatter areas when they first move to the City. The remaining 32 percent of all the respondents moving from squatter settlements to Dandora had lived in standard housing when they first moved to Nairobi and then moved to squatter settlements before moving to Dandora. Except for 8 percent, all the respondents moving from standard housing to squatter settlements had lived with friends or relatives in standard
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
20 0
housing. This is a common practice in Nairobi, where on arrival, most people live with relatives and friends until they can find their own place to live in. It is not surprising that most of these people would move to the squatter areas where housing is affordable. Standard housing is not only expensive but hard to find, thus only those who have lived in the City for a while are able to get access to the standard housing that they can afford.
The respondents who had moved from the squatter settlements to Dandora had an average age of 45 years. A total of 17 percent were 30 years or younger while the remaining 84 percent were 31-90 years old. The low percentage of those aged 30 years or less, indicates that the project benefited an older squatter population. Most of these respondents had lived in the squatter areas for a long time. A total of 59 percent of all the respondents moving from the squatter settlements to Dandora had lived in Nairobi for more than 20 years. The new migrants to squatter settlements were not among the major beneficiaries of this project. Most of the new migrants are younger, as indicated by the age of those who had moved from rural areas to Dandora. A total of 56 percent of all moving from rural areas were 30 years old or less. These respondents had lived in the City for relatively shorter periods, with 50 percent of all those moving from the rural areas, having lived in Nairobi for five years or less. Thus, most of the respondents moving to Dandora from the
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
20 1
squatter settlements included older household heads, while those moving from the rural areas were younger.
Half of the respondents moving from the squatter settlements were married. This compares with the same percentage of those moving from the rural areas. In contrast, a higher percentage of those moving from standard housing were married. A total of 33 percent of all those married moving from the squatter settlements lived with their spouse in Dandora while the remaining 17 percent had left their spouse in the rural areas. In addition, 91 percent of these respondents had children ranging from one to ten. Almost 26 percent of all the respondents did not live with their children on the plot. The children had either been left in the rural areas, or had moved out to live on their own. The high number of household heads living with their spouse ^nd children put a lot of pressure on the number of rooms available. This was because most of the respondents lived in one room and shared that one room with the rest of the family. The lack of he using for families in Nairobi has meant that most of the households will continue to occupy one room in spite of the increasing number of families moving to the City. This is especially true of the low-income earners, most of whom are unable to afford any other kind of housing. Those leaving their spouse and children in the rural areas did so because of a lack of adequate housing and low incomes.
Except for one respondent, all the household heads moving from the squatter settlements had some source of income. A
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2 0 2
total of 29 percent depended wholly on Income from subletting, 38 percent were self employed and 13 percent were employed office support staff, while the remaining 20 percent were spread throughout smaller sectors (see Table 5.1).
A total of 69 percent worked in both the formal and the informal sectors3. Those employed in the informal sector comprised 46 percent, while the informal sector had 23 percent. Except for those who have lived in the City for a long time and have established jobs, informal sector employment provides little income, thus making housing too expensive except in the squatter areas. This explains why initially most of these households would have stayed in the squatter areas. However, many of the respondents had lived in the City for a long time and were established in the informal sector, thus having stable incomes. This enabled them to have the required funds to construct their plots in Dandora. Except for 31 percent, all the respondents worked five or less kilometres away. Of these, all but 38 percent walked to their place of work. This consisted of 32 percent who took public transportation, the one percent who had their own cars and the 5 percent who used enployer-provided transportation. The highest number of respondents (64 percent) did not pay for transportation, while 32 percent paid Kshs. 24-300. The high percentage of respondents not paying for transportation is explained by the fact that most of them
3 See definition in Chapter 1
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
aoa
Table 5.1 Jobs Held by the Residents who Moved Directly
from the Squatter Settlements to Dandora
Occupation No. from Sample Percent
Self-employed 31 38 Landlord 23 29 Office Support Staff 10 13 Building Trades 4 5 Dr iver/Conductor 4 5 Casual Worker 2 3 Machine Operator 2 3 Hotel Services 1 1 Electrical Mechanic 1 1 Factory Workers 1 1 Unemployed 1 1 Total 80 100
Source: Field Data
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2 0 4
worked within the settlement, thus walked to their place of work. Except for one-third of the respondents, the 12 kilometre distance to the project did not create a problem for respondents because they worked within walking distance.
The incomes earned by most of the residents moving from the squatter settlements were in the middle income range (see Figure 21). Those earning Kshs. 1,000 or less comprised 29 percent, while the remaining 71 percent had higher incomes. Those earning less than Kshs. 1,0000 worked in different sectors. The highest percentage were landlords, who comprised 13 percent of all those earning less than Kshs. 1,000 from the squatter settlements. The landlords included in this category were those who did «ot have any other sources of income, except from suble'-ting their plots. This compares with 3 percent of the landlords who had moved from standard housing. The disparity is explained by the fact that most of the plot owners who had moved from standard housing had other sources of income. Only a small percentage had the plots as their only source of income. This was not the case with those moving from the squatter settlements, where many depended wholly on income from subletting.
The next highest percentage of respondents earning Kshs. less than 1,000 were self-employed. This consisted of 8 percent of all the residents who had moved from the squatter settlements. Self employment in the informal sector in low- income settlements docs not usually provide stable incomes. This is because of the nature of the businesses undertaken by
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
R eproduced
with perm
ission of
the copyright
ow ner.
F urther
reproduction prohibited
w ithout
perm issio
n .
Figure 21 Monthly Incomes of Residents Moving from
the Squatter Settlem ents to Dandora
1501-2000
2000+ 40%
S o u r c e : F ie ld D a t a 205
2 0 6
those self-employed. Most of these are vegetable hawkers, tailors, dry cleaners and shopkeepers. The market for most of these items are residents from the settlements who are not able to pay much for these goods and services. Others earning Kshs. 1,000 consisted of 3 percent office support staff, 3 percent construction workers, one percent worked in hotels while the remaining one percent were unemployed.
A total of 40 percent of all the respondents moving from squatter settlements, earned incomes in the middle income range. The respondents moving from the squatter settlement had lived in the City for a long time. Most of these had stable jobs in the informal sector and earned high incomes which enabled them to afford housing construction in Dandora. This is shown by the incomes earned by those self-employed in the informal sector. This comprised 38 percent of those who were self employed. A total of 30 percent of all the respondents self-employed earned more than Kshs. 1,000. Only a small percentage (9 percent) of those self-employed in the informal sector earned Kshs. 1,000 or less. This project thus gave households that were self-employed in the informal sector an opportunity to own property. This would not have been the case with the conventional housing finance agencies who base their decisions on the type of occupation and income earned. The above figures show that the project benefited squatters in the middle income bracket. Thus, the majority of the squatters who earned Kshs. 1,000 or less could not afford to move to this project because of their low incomes. This
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2 0 7
explains why the project met the housing needs of only a small number of households from the squatter areas. This is especially true given the fact that most of those moving from squatter settlements to Dandora were plot owners and not tenants. This meant that they were required to have higher incomes to be able to construct their rooms, whose cost was beyond the reach of the low-income group as shown in Chapter 4. The high construction cost has in turn resulted in high and inadequate benefits to the intended beneficiaries.
5.3 Evaluation of Housing Conditions as a Result of the Move from the Squatter Settlements to Dandora
All the respondents moving from the squatter settlements to Dandora had moved from different squatter areas in the City. The highest percentage of 45 percent had moved from Mathare Valley, the largest squatter area in Nairobi (see Figure 22), and proximate to Dandora. The second highest percentage of 14 percent had moved from Kariobangi which is next to Mathare Valley. This was followed by those who moved from the Korokosho/Ngomongo area (5 percent) and those moving from the outskirts of the City (5 percent) . Most of these origins were however within a five mile radius from Dandora (see Table 5.2).
A total of 73 percent of all the respondents lived in rented accommodation, 19 percent owned the units, while 9 percent lived with other people. A total of 90 percent of
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
R eproduced
with perm
ission of
the copyright
ow ner.
F urther
reproduction prohibited
w ithout
perm issio
n .
F igure 22
SQUATTER SETTLEMENTS ORIGIN AREAS OF THE DANDORA RESIDENTS
D andora P roject
Km Source. MA2INGIRA 1983, hr
1 M athare V a lle / 2 K ariobangi 3 K orogosho 4 Pum w ani 5 Ktbera 6 K angem i
45%19% 8% 4% 4% 4%
H C ity C entre
M O QD
2 0 9
Table 5.2 A List of the Different Squatter Areas the Residents
Lived in Before Moving to Dandora
No. from Place Lived Sample Percent
Mathare Valley 36 45 Kariobangi 15 19 Korokosho/Ngomongo 6 7 Outskirts of the City 8 10 Maj engo/Pumwani 3 4 Kibera 3 4 Kangemi 3 4 Dandora 3 4 Kawangware 1 1 Makadara 1 1 Gikomba 1 1 Total 80 100
Source: Field Data
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2 10
all these units in the squatter settlements were constructed with temporary materials. Only 10 percent had moved from housing built with permanent materials. Thus the move to Dandora, where all the housing was built with permanent materials, significantly improved the housing situation of those moving from the squatter settlements.
The number of persons per household ranged from one to twelve in the squatter settlements, compared with one to nineteen in Dandora. The average household size in the squatter settlements was 5 persons compared to 6 persons in Dandora. These are large household sizes compared to the number of rooms available per household. Thus, although the number of rooms increased with the move to Dandora, this did not increase at the same rate as the number and size of households.
The number of rooms per household increased with the move to Dandora as indicated in Table 5.3. Those with one room in the squatter settlements comprised 69 percent of all the respondents moving from the squatter settlements to Dandora. The number of respondents with one room were reduced to 42 percent in Dandora. This meant that a total of 27 percent of the respondents with one room in the squatter settlements had more rooms in Dandora. Those with two rooms in the squatter settlements comprised 19 percent, and this increased to 22 percent. Similarly, 10 percent of the respondents had three to four rooms before moving to Dandora. This increased to 30 percent in Dandora. Those having five rooms in Dandora
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2 11
Table 5.3 Number of Rooms in the Squatter Settlement
and in Dandora
Squatter Settlements Dandora No. from No. from
Rooms Sample Percent Sample Percent
1 55 69 31 39
2 15 18 16 22
3 5 6 11 13
4 3 4 12 14
5 0 5 6
6-7 2 3 5 6
Total 80 100 80 100
Source: Field Data
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2 12
comprised 6 percent. No households had five room before moving to Dandora, but 3 percent had six to seven rooms. The number of respondents per room ranged from one to twelve in the squatter settlements compared to one to nineteen in Dandora (see Table 5.4). The high percentage of respondents with more rooms in Dandora is explained by the high percentage of household heads who owned the plots. A total of 75 percent of all those moving from squatter settlements were plot owners while 25 percent were tenants. As discussed in Chapter 4, most of the tenants had one room while the plot owners had more rooms.
The number of persons in the single room units changed significantly after the move to Dandora. Thus, those with one to two persons per room consisted of 34 percent of all the respondents in the squatter settlements, compared to 19 percent in Dandora. Thus, 12 percent of the households had more persons in Dandora than in squatter settlements. Those with 3-10 persons in the squatter settlements decreased from 28 percent in the squatter settlements to 18 percent in Dandora, while those with more than ten persons accounted for one percent in both locations. According to these figures, one-roomed units were crowded, with the highest level of crowding being experienced in Dandora. Thus, the move to Dandora did not reduce the level of crowding for those in one- roomed units.
The respondents with two rooms accounted for .19 percent in the squatter settlements, compared to 22 percent in
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2 13
Table 5.4 Number of Persons per Room in the Squatter Settlement
and in Dandora
Squatter Settlement Dandora
Rooms Persons No. from Sample Percent
No. from Sample Percent
1 1-2 27 34 15 19 3-6 18 24 12 15 7-10 9 10 3 4 11-15 1 1 1 1
2 1-2 0 0 6 8 3-6 6 8 8 11 7-10 8 10 1 1 11-19 1 1 2 2
3 2-6 1 1 3 4 7-10 4 5 6 8 11-15 0 0 1 1
4 3-6 0 0 2 2 7-10 3 4 5 6 11-20 0 0 5 6
5 7-10 0 0 3 4 11-15 0 0 2 2
6-7 3-6 1 1 5 6 7-10 1 1 0 0
Total 80 100 80 100
Source: Field Data
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2 14
Dandora. A small percentage of 3 of the respondents with two rooms in the squatter settlements had more rooms in Dandora. The rooms occupied by one to two persons consisted of 8 percent in Dandora and none in the squatter settlements. The units with three to ten persons consisted of 18 percent in the squatter settlements, compared to 12 percent in Dandora. Those with more than ten persons per household accounted for one percent in the squatter settlements, compared to 3 percent in Dandora. These figures represents a slightly lower level of crowding in Dandora.
The respondents with three rooms comprised of 6 percent in the squatter settlements, compared to 15 percent in Dandora. A total of 9 percent of the respondents had more rooms as a result of the move to Dandora. The rooms occupied by two to six persons comprised 4 percent in Dandora, compared to one percent in the squatter settlements. The units with seven to ten persons comprised 5 percent in the squatter settlements, compared to 9 percent in Dandora.
None of the households with three rooms had more than ten persons in the squatte*' ctlements while one percent had eleven to fifteen persons in Dandora. A total of 4 percent had four rooms in the squatter settlements compared to 15 percent in Dandora. Those with two to six persons accounted for 9 percent of the respondents in Dandora and none in the squatter settlements. A total of 4 percent had seven to ten persons in both locations while 3 percent had eleven to fifteen persons in Dandora. This shows a significant
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2 15
improvement in terms of the number of rooms available to the respondents as a result of the move to Dandora. More households not only had the number of rooms increase, but also had fewer persons per room. None of the respondents had five rooms in the squatter settlements, while 6 percent did in Dandora. These respondents had seven to fifteen persons. A total of 3 percent of the respondents in the squatter settlements had six to seven rooms. These rooms were occupied by three to seven persons. Thus, the move to Dandora not only increased the number of rooms for respondents moving from the squatter settlements, but also reduced the number of occupants per room. In addition, the rooms were constructed with permanent materials in Dandora, while 90 percent of the rooms in the squatter settlements were constructed with temporary materials.
The housing costs discussed in this section include the actual rent paid and cost of services such as water and electricity. The housing costs for the respondents moving from the squatter settlement increased significantly with the move to Dandora (see Figure 23). The high cost was explained by the better housing available to the respondents in Dandora. All the plots had water, toilet and shower facilities in Dandora, unlike in some cases in the squatter settlements. A total of 15 percent of all the respondents paid no rent in the squatter settlements, while all the respondents in Dandora paid rent. Those paying no rent included those respondents who lived with relatives or friends before moving to Dandora.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
P e r c e n t 50
Figure 23 Housing Expenses in the Squatter
Settlem ents and in Dandora
40
30
20
10
0 ■ i:z\ 1-100 1 0 1 -2 0 0 2 0 1 -3 0 0 3 0 1 -4 0 0 4 0 1 -5 0 0 5 0 1 -1 0 0 0 1001-1250
C ost in Kshs. p.m.
S q u a t t e r A r e a s i I D a n d o r a
S o u r c e : F i e l d D a t a
2 17
Those paying Kshs. 200 or less comprised 73 percent in the squatter settlement compared to 3 percent in Dandora. A total of 10 percent paid Kshs. 201-400 in the squatter settlements compared to 31 percent in Dandora. The remaining 3 percent of all the respondents moving from the squatter settlements paid Kshs. 401-728, compared to 53 percent of the respondents who moved to Dandora.
The maximum rent paid in the squatter settlements was Kshs. 728, compared to Kshs. 1,250 in Dandora. A total of 22 percent paid Kshs. 728-1,250. The average rent paid in the squatter settlements excluding those who did not pay rent was Kshs. 62, compared to Kshs. 506 in Dandora. Using the rule of thumb where one is expected not to spend more than 25 percent of their income on housing, the squatter settlement housed those earning as little as Kshs. 248. On the other hand, the Dandora project would house those earning Kshs. 2,024. The low rent in the squatter settlements explains why these settlements continue to house the lcw-income households in Nairobi. These settlements are also increasing at a rapid rate because they house most of the new migrants to the city. The high housing expenses in Dandora explain in part why the project benefited only a small percentage of squatters. A small percentage of the respondents from squatter settlements also earned high incomes. This explains why low-income households will not continue to directly benefit from these projects. Until the rents are reduced to the level of that in squatter settlements, these projects will continue to be
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2 1 8
out of reach of most of the squatters who cannot afford to pay more than they are paying in the squatter settlements.
The respondents from the squatter settlements benefited significantly from this project because they not only obtained permanent housing, but also had basic services available to them. The Dandora project provided all the respondents with w?':er, toilet and shower facilities (see Figure 24) . This was not the case in the squatter settlements, where 10 percent of the respondents did not have water. Most of those with water in squatter settlements depended on water bought from communal water taps. These were within a walking distance for most people but not on the plot like in Dandora. Water was paid for separately according to quantity, while payment was included in the monthly rent in Dandora. Water was thus much more accessible in Dandora, where the cost was fixed and paid for once a month. In addition to not having water, a total of 18 percent did not have toilet facilities in the squatter settlements. Those with toilet facilities either used pit latrines or communal toilet facilities. These were of a lower quality than those in Dandora. Thus, the Dandora project not only provided these services, but they were of a better quality. Shower facilities were not accessible to 80 percent of the respondents in the squatter settlements. All these services were available on each plot in Dandora, thus improving the level of services for all those moving to Dandora.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
R eproduced
with perm
ission of
the copyright
ow ner.
F urther
reproduction prohibited
w ithout
perm issio
n .
Figure 24 Available Services in the Squatter
Settlem ents and in Dandora
P e r c e n t 120
100
80
60
40
20 0
W a te r T o ile t
mm..
S h o w e r S e r v i c e s
K it c h e n E l e c t r i c i t y
S e r v i c e s B e fo r e [ 1 S e r v i c e s A f te r
S o u rc e : F ie ld D a t a 21 9
2 2 0
The other two services discussed included electricity and kitchen facilities. These facilities were not present in most of the squatter settlements. Thus, only three respondents had kitchen facilities, compared to six who had electricity in squatter settlements. The move to Dandora increased the number of those with kitchen facilities to 18 households and those with electricity to 36 households.
To assess the benefits of the project as perceived by the respondents, they were asked why they moved to Dandora. The major reason why the respondents moved to Dandora was the fact that they had their own plot. Many had lived in Nairobi for a long time, and had no hope of getting their own houses given the fact that they did not qualify to get loans which could have enabled them to afford the current housing available in the city. Thus 74 percent of the respondents moved because they got their own plot. Other reasons for moving included job opportunities (9 percent), better housing (8 percent), safety (4 percent) , more space (3 percent), cheaper housing (3 percent) and availability of housing (one percent).
In conclusion, the move to Dandora brought many changes to the households that moved from the squatter settlements. The major changes included better housing quality and a higher level of services. A total of 90 percent living in housing built with temporary materials in the squatter settlements had housing constructed with permanent materials. Similarly, households lacking water, toilet and shower facilities in the
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
22 1
squatter settlements gained these in Dandora. In addition, the services in Dandora were of a better quality than what they had before. Some of the households experienced an increase in the number of rooms with the move to Dandora. The increase in the number of rooms was however accompanied by a higher increase in household size. The increase in the size of households is not peculiar to the Dandora project, but has characterized most households in the country. This increase is more noticeable in the low-income settlements where many have to double up to afford rent. The housing cost however increased significantly with the move to Dandora. The increased housing expenses excluded low-income households, who could not afford to pay the high rents charged in this proj ect.
5.4 Direct Move from the Rural Areas to Dandora
A total of 13 percent of all the respondents moved from the rural areas to Dandora (see Figure 19) . This group of respondents consisted of thirty one tenants and four plot owners. The low number of plot owners was due to the fact that respondents from the rural areas were not among the intended beneficiaries. This was specified in eligibility criteria which required the beneficiaries to have lived in Nairobi for a period of not less than two years. Also, incomes are higher in urban areas, and so most migrants come
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2 2 2
to the City in search of better jobs. It is unlikely that they would have had the required savings to be able to construct the houses in Dandora. The number of migrants to Nairobi is important given the high growth rate of the City at 7.5 percent. Most of the new migrants live in the squatter settlements when they first move to the City. Most of the housing they leave behind in the rural areas is substandard. It is often constructed using temporary materials and lacks basic services. These factors indicate that migrants to the City will continue to benefit directly from these projects as tenants and not as owners, because they usually do not have the finances required for construction.
The respondents moving directly from the rural areas to Dandora included a higher percentage of younger households than those moving directly from the squatter settlements. Thus, while the average age for those moving from the rural areas was 31 years, it was 45 years for those moving from the squatter settlements. A total of 56 were aged between 20-30. This compares with 17 percent of those moving from the squatter settlements. The relatively younger population of respondents moving from rural areas is explained by the fact that most of the migrants who come to the City are young. These respondents had lived in Nairobi for a shorter period, with the average number of years lived in Nairobi being 9 years. A total of 50 percent had lived in Nairobi for periods ranging from 3 months to five years while the rest had stayed for 6-27 years.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2 2 3
The respondents were comprised of 49 percent married and 51 percent single household heads. Of all the married respondents, 26 percent lived with their spouse in the urban areas, while 17 percent left their spouse in the rural areas. In addition, a total of 77 percent had children ranging in number from 1-12. However, not all the respondents with children lived with them in Nairobi. Thus, a total of 32 percent of all the respondents with children left them in the rural areas. This reduced the average number of children per household from three children to one. The reason given for leaving children in the rural areas was low incomes and lack of adequate housing. These figures, however, represent large households when compared to the number of rooms available.
A total of 91 percent of all the respondents who moved to Dandora from the rural areas had jobs in both the formal and the informal sectors. Thus 57 percent had jobs in the formal sector, while 34 percent worked in the informal sector. A total of 54 percent of all those employed had permanent jobs, while the remaining 37 percent had temporary jobs. The high rate of temporary employment is explained by the fact that most migrants first take temporary employment while waiting for permanent jobs which are hard to find. Those not employed accounted for 6 percent, while 3 percent had their plots as their only source of income (see Table 5.5). The highest number of the respondents were self employed. This is followed by those who worked as office support staff, who were employed in both the public and the private sectors.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
224
Table 5.5 Jobs Held by the Residents who had Moved from the
Rural Areas to Dandora
Occupation No. from Sample Percent
Self-employed 11 32 Office Support Staff 6 17 Barmaid 5 14 Machine Operator 4 11 Driver or Conductor 3 8 Electrical Mechanic 1 3 Building Trades 2 6 Landlord 1 3 Unemployed 2 6 Total 35 100
Source: Field Data
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Almost half of all the respondents moving from rural areas to Dandora worked in Dandora. Those who worked on the plot or within the project comprised 43 percent of all those moving from the rural areas to Dandora. A total of 6 percent worked 1-5 kilometres away, 43 percent worked 6-11 kilometres while 9 percent worked twelve kilometres or more. The high percentage of those working more than five kilometres away is because most jobs in the formal sector, where many worked, are located in the City centre or beyond. The means of transportation to work comprised 31 percent who walked, 43 percent who took public transportation and 17 percent who took company transportation. A total of 54 percent did not pay for transportation, while 46 percent paid Kshs. 90-200 per month. The transportation cost represents a significant cost to the respondents, and especially if they had low incomes. If those earning Kshs. 280 were to benefit, they would have spent between 35-71 percent of their income on transportation. If those paying belonged to higher bracket (Kshs. 1,000) they would have spent 20 percent of their incomes on transportation alone. Given the high cost of housing, none of the respondents earning less than Kshs. 1,000 could afford to live in Dandora and pay for transportation. Thus, the distance of the project from employment areas meant that only those with high paying jobs could afford to live in the Dandora project and pay for transportation.
The income earned by the respondents who had moved from the rural areas was lower than that earned by those moving
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2 2 0
from the squatter settlements. Those earning Kshs. 1,000 accounted for 37 percent from the rural areas (see Figure 25) compared to 29 percent from the squatter settlements. Similarly those earning Kshs. 1,001-2,000 comprised 48 percent of those moving from the rural areas, compared to 31 percent of those who had moved from the squatter settlements. Those earning more than Kshs. 2,000 comprised 14 percent of those from the rural areas, compared to 40 percent of those from the squatter settlements. The high incomes for those moving from the squatter settlements is attributed to the fact that most of them had lived in Nairobi for a long time and were established in the informal sector, thus having stable incomes. In addition, most of them had plots and had the added advantage of having extra income from subletting. Unlike them, however, most of the respondents from the rural areas were just starting in new jobs, which do not usually pay well. Also, incomes in the informal sector are low, especially for the newcomer to the City. This is because the new migrant does not usually have enough money to start a business, and usually has to depend on others for employment.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
R eproduced
with perm
ission of
the copyright
ow ner.
F urther
reproduction prohibited
w ithout
perm issio
n .
Figure 25 Incomes of the Residents Moving from
Rural Areas to Dandora
Kshs. 1001-1500 31%
Kshs. 1000 or less 37%
Kshs. 2000+ 14%
Kshs. 1501-2000 17%
Source: Field Data
227
2 2 8
5.4.1 An Evaluation of Housing Conditions as a Result of the Move from the Rural Areas to Dandora
The housing conditions of those moving from the rural areas increased significantly as a result of the move to Dandora. One major improvement was in terms of quality of housing. A total of 66 percent of those moving from the rural areas lived in housing constructed with temporary materials. All of these respondents moved to housing constructed with permanent materials in Dandora. A total of 74 percent of all the households from rural areas owned the housing they moved from compared to 11 percent in Dandora. The high rate of ownership is explained by the fact that housing in the rural areas is not rented/ but built by residents as the need arises.
A total of 83 percent did not pay for any housing expenses before moving to Dandora, while 17 percent paid rents ranging from Kshs. 15 to 394. The small number of respondents paying housing expenses was due to the fact that housing in the rural areas is free. All these respondents paid rent in Dandora which ranged from Kshs. 200-800. Thus 34 percent paid Kshs. 200-300, 37 percent paid Kshs. 301-400 while the remaining 29 percent paid Kshs. 401-800. This represents a high expense for all the respondents, since all except 5 households did not pay for housing before Dandora. They however, had access to permanent housing and services that they did not have before. The expenses for the respondents
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2 2 9
moving from the rural areas were lower than those moving from the squatter settlements. This is explained by the fact that most of the respondents moving from the squatter set lements to Dandora were plot owners, and plot owners had higher housing expenses than tenants.
The number of persons per household before moving to Dandora ranged from one to fifteen while they ranged from one to thirteen in Dandora. The average number of persons per household decreased from 6 persons before moving to Dandora, to 4 persons in Dandora (see Table 5.6). The households with one to two persons consisted of 37 of all the households in Dandora compared to 23 percent in the rural areas. The highest number of households in Dandora had three to four persons and consisted of 31 percent. This compares with 14 percent in the rural areas. The remaining 17 percent of the households in Dandora had five to thirteen persons. This compares with 60 percent of all the households moving from the rural areas. While no households in Dandora had more than thirteen persons, one household from the rural areas had fifteen persons. The above figures clearly show very large households in the rural areas compared to those in Dandora. This is true for the rest of the country, where households in the rural areas are much bigger than in the urban areas. The high number of one to two person households in the rural areas is explained by the fact that households do not have to double up in the rural areas because they build houses when they need
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2 3 0
Table 5.6 Number of Persons per Household in the Rural Areas
and in Dandora
Dandora Rural Areas No. from No. from
Persons Sample Percent Sample Percent
1 9 26 6 17 2 4 11 2 6 3-4 11 32 5 14 5-6 5 14 10 28 7-10 3 9 9 26 11-13 3 8 2 6 15 0 0 1 3 Total 35 100 35 100
Source: Field Data
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2 3 1
them, unlike urban areas where this depends on the ability to pay.
The number of rooms available in both Dandora and the rural areas varied significantly. The average number of rooms in Dandora was one, while this was 3 in the rural areas. Those with one room in the rural areas comprised 26 percent, compared to 77 percent in Dandora. Those with two rooms were the same in both Dandora and the rural areas, comprising of 17 percent of all the households. The remaining 6 percent in Dandora had four rooms. This compares with 57 percent who had three to seven rooms in rural areas. The high number of respondents with more rooms in the rural areas is explained by the fact that houses in the rural areas are built as needed and are constructed with local material, thus are not as expensive as in urban areas. Also, housing is not rented in the rural areas, while all housing in urban areas is rented, thus restricting the number of rooms one can rent. The functions of a house are culturally determined in the rural areas, unlike in the urban areas where this is determined by income.
The number of persons per room determined the level of crowding (see Table 5.7). The one-roomed units with one to two persons in the rural areas comprised of 14 percent compared to 34 percent in Dandora. The households with three to six persons included 12 persons in the rural areas compared to 26 percent in Dandora. Consequently those with seven or more persons consisted of 3 percent in both locations. The
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2 3 2
Table 5.7 Number of Persons per Room in the Rural Areas
and in Dandora
Rural Areas Dandora
Rooms Persons No. from Sample Percent
No. from Sample Percent
1 1 3 9 9 25.7 2 1 3 3 8.6 3-6 4 10 12 34 7-10 1 3 2 5.8 13 0 0 1 2.9
2 1-2 2 6 1 2.9 3-6 2 6 1 2.9 7-9 2 6 1 2.9
3 2 1 3 0 0 3-6 7 20 0 0 8-9 4 10 0 0
4 1-6 2 6 0 0 7-9 2 6 2 5.7 11 1 3 2 5.7
5-7 5 1 3 0 0 11-15 1 3 0 0 16-21 1 3 1 2.9
Total 35 100 35 100
Source: Field Data
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2 3 :i
above figures represent a higher level of crowding in one roomed units in Dandora than in the rural areas. Those with two rooms consisted of 17 percent in both Dandora and the rural areas. A total of 6 percent had one to two persons in the rural areas, compared with 3 percent in Dandora. Those with three to six persons accounted for 6 percent in the rural areas, compared to 12 percent in Dandora. The remaining households had seven to nine persons. These comprised 6 percent in the rural areas compared to 3 percent in Dandora. The two-roomed units * Dandora were slightly more crowded than those in the rural areas.
Those with three rooms consisted of the highest percentage in the rural areas (34 percent). Those with two persons accounted for 3 percent while 20 percent had three to six persons. The remaining 11 percent had eight to nine persons. The highest number of rooms in Dandora was four rooms, and only two households (6 percent) had four rooms. One of these households had seven persons, while the other had eleven persons. Those with four to seven rooms before moving to Dandora consisted of 23 percent.
Except for 14 percent, all the respondents moving from rural areas were crowded. A total of 29 percent had adequate room in Dandora and all these respondents had one room per household. This compares with 9 percent before moving to Dandora. Thus, a higher percentage not only lived in permanent housi .g in Dandora, but also had adequate room.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2 3 4
The move to Dandora thus significantly improved the housing conditions of the households moving from the rural areas.
The services available to the respondents moving to Dandora increased significantly. All the respondents in Dandora had water, toilet and shower facilities. This compares with 31 percent of the respondents in the rural areas who did not have water before coming to Dandora. Most people in the rural areas have to walk long distances to fetch water. Except for one household, all the households had toilet facilities before coming to Dandora. These are mainly pit latrines which are of a lower standard compared to those in Dandora. A high 34 percent of all the respondents did not have shower facilities before coming to Dandora. Except for 9 percent, all the households in Dandora lacked kitchen facilities. This compares with 63 percent in the rural areas who had kitchen facilities. Thus 91 percent had no kitchen facilities in Dandora compared to 37 percent in the rural areas. The high percentage of respondents with kitchen facilities in rural areas is due to the fact that individuals built them when they need them. Accessibility to a kitchen in Dandora, however, meant an added expense for the tenants. Electricity was not readily available to most of the respondents in both locations. Thus, only 17 percent had electricity in the rural areas, compared to 37 percent in Dandora. Most places in the rural areas are not supplied with electricity, and where it does exists it is too expensive for the rural households. The move from the rural areas . ot only
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2 3 5
increased the services available, but also provided services of a higher quality than that in the rural areas, where services are not only unavailable but are also of a lower quality.
5.5 Conclusions
The respondents moving directly from the squatter settlements and the rural areas benefited significantly from the Dandora project. The Dandora project benefited most of these households by providing them with permanent housing and better services. Thus, the move enabled 90 percent of all the respondents moving from the squatter settlements to have access to housing constructed with permanent material such as in Dandora. This compares with 63 percent of those from the rural areas who had access to housing built with permanent materials in Dandora. All respondents had water, toilet and shower facilities in Dandora. The housing quality in Dandora was also better than that found in the squatter settlements or the rural areas. On the other hand, the number of rooms available to the respondents only increased for those respondents who were moving from squatter settlements to Dandora, and actually decreased for those moving from rural areas. This is explained by the fact that most of the housing in rural areas is built as needed, and is not usually paid for, while housing in the squatter settlements is constructed
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2 3 6
on expensive urban land. The high land cost influences the rents charged in urban areas, while the rural areas are unaffected.
Household sizes increased significantly in Dandora. This is partly explained by the fact that a number of people have to double up with others so as to be able to afford housing in Dandora. This is not necessarily the case in the squatter settlements, where the rents are cheaper, and in the rural areas where most of the housing is free. The move to Dandora significantly increased the housing expenses for all the households. This is mainly because housing in Dandora was built with permanent materials and had most of the required services. For the owners, the housing expenses were even higher because of the high cost of services due to the large number of households per plot.
The data lead to the conclusion that these kinds of projects directly benefit a small percentage of squatters. Those who benefit belong to a higher income group than the income group targeted for this project. The high housing costs for site and service projects make them too expensive for most squatters, whose incomes are low. The direct benefit of this project to the squatters is minimal. Since the real incomes of the Nairobi residents are said not to have changed since the 1970s when this project was implemented, it is unlikely that these projects will directly benefit a large squatter population. The projects, however, made it possible for those squatters with high incomes to leave the squatter
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2 3 7
settlements and move to better housing in Dandora. Although the percent benefiting from squatter settlements was small, it is significant, given the high competition for home ownership in the City. Also most of the squatters moving to Dandora were older and had lived in the City for a long time. Most of these also worked in the informal sector and would not necessarily retire and return to the rural areas because they may not have land there. This project made it possible for them to move to better housing. If the project had not been implemented, most of them would have to live in the squatter settlements all their lives, because they did not earn incomes or hold jobs that would have made then eligible for other standard housing developments in the City.
The respondents moving from the rural areas were young and had not lived in the City for a long time. These mainly benefited from this project as Tenants. Given the high number of migrants living in the squatter settlements when they move to Nairobi, the project provided housing for respondents who would have had to live in the squatter settlements because of their incomes. Most of these had no contacts in the City, and stated that Dandora provided them with affordable housing. Some of the respondents had moved to this project because they could find a room to rent in Dandora in a day. Given the few number of available standard housing units in the City most of these residents would not have found available standard housing elsewhere in the City. Thus, for 13 percent who moved directly from the rural areas to Dandora, the project provided
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2 3 8
them with the rare opportunity of living in standard housing that they did not even have to share because they could afford the rent. The project thus affected the growth of squatter housing which may have been the inevitable choice for most of these households. Given the rapid increase of the new migrants to the City, these projects will continue to play an important role in providing housing for households who would have no other choice but to live in the squatter settlements.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER 6
THE MIGRATION FROM OTHER HOUSING AND INDIRECT IMPACT OF THE SITE AND SERVICE PROJECTS ON THE SQUATTER SETTLEMENTS
6.1 Introduction
One of the objectives of this study was to assess whether the Dandora project benefited low-income households indirectly through the filtering process. This was done by gathering the migration history of the respondents who moved from standard housing1 to Dandora, and inferring from that history the likely indirect impact of Dandora on housing conditions on other housing sectors of Nairobi. This section discusses the respondents who moved from standard housing to Dandora. The respondents identified the different moves they had made in the City and gave information on their socio-economic characteristics and housing conditions in different locations. These are assessed and compared in order to determine whether the housing left behind benefited low-income households.
The study also assessed whether the housing conditions of those who moved from standard housing improved over time. The households making these moves are represented by "Case three" and "Case four" as shown in Figure 26. "Case three" represents a situation where the respondents moved directly
1 See definition in Chapter 1
239
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2 4 0
Figure 26
MIGRATION HISTORY OF THE RESIOENTS MOVING FROM STANDARD HOUSING TO DANDORA
Rural Areas
Standard Housing
V
10% (18 Residents) S q u a t t e r A r e a s
100% (182 Residents) Standard HousingDANDORA
i/i
CO
Born in Nairobi
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2 4 1
from standard housing to Dandora and comprised 60 percent of all the respondents moving to Dandora. Case four represents the households who had moved from the squatter settlements to standard housing before moving to Dandora and constituted 10 percent2 of all the respondents moving from standard housing to Dandora.
A total of 4 percent of the respondents moving from standard housing had been born in standard housing in Nairobi. This compares with 16 percent of the respondents born in Nairobi who moved from the squatter settlements. Those who had lived in standard housing since moving to Nairobi from rural areas comprised of 44 percent. These households had lived in the same housing unit since moving to the City. A total of 42 percent of those moving from standard housing had, however, moved from standard housing in another location. These households had originally moved from rural areas. The high number originating from rural areas is due to the fact that until 1963, when Kenya gained independence, Africans were regarded as temporary migrants who came to the City to work and would therefore eventually return to the rural areas. Although this policy has been changed, most people coming from the rural areas come to the City to work and eventually move back to the rural areas. This explains the high demand for rental housing. This chapter examines whether the move to
2 This refers to 10 percent of the respondents moving from standard housing to Dandora. When all the respondents are included those moving from the squatter settlements to standard housing and then to Dandora included 6 percent.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2 4 2
Dandora improved the housing conditions of the households moving from standard housing to Dandora, and whether they left housing that could have benefited low-income households. Both socio-economic and housing conditions are discussed in the following sections.
6.2 Socio-economic Characteristics of Residents Moving from Standard Housing to Dandora
The respondents who moved from standard housing to Dandora were much younger than those who moved from the squatter settlements. The mean age for those moving from standard housing was 38 years, compared to 45 years for those moving from the squatter settlements. The age of those from standard housing ranged from 18-80 years. These respondents had lived in the City for periods ranging from 3 months to 59 years. The average number of years lived in Nairobi was 15 years, which compares with 23 years for those from the squatter settlements. The difference in age is explained in part by the large number of tenants moving from standard housing to Da idora. Most of these tenants were young and had not lived in the City for a long time. The respondents moving from the squatter settlements were, however, mainly plot owners, and they had lived in the City for a longer time. Although the respondents from standard housing were younger than those from the squatter settlements, the project did not
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2 13
reach a majority of the City's younger population who are less than 30 years old.
A higher percentage of residents moving from standard housing were married. A total of 60 percent of the respondents moving from standard housing were married, compared to 50 percent of those moving from the squatter settlements. Those living with their spouse in the Dandora project consisted of 36 percent for those from standard housing, compared to 34 percent for those from the squatter settlements. This represents a large number of respondents living with their spouse in urban areas in both cases. As illustrated by the number of rooms available per household, most of them had to live in one-roomed units and thus experienced high levels of crowding. In addition to living with their spouse in urban areas, a total of 85 percent had children ranging in number from one to sixteen. Those unable to live with their children on the plot due to lack of housing and low incomes comprised of 26 percent of all those with children. The remaining 58 percent had children ranging from one tc ten. The large households put a lot of pressure on the limited number of rooms available to the households.
A total of 83 percent of all the respondents moving from standard housing worked in both the formal and the informal sectors3. A total of 29 percent of these worked in the informal sector, while 54 percent worked in the formal sector.
3 See definition on page 13
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2 4 4
This compares with 69 percent of the respondents from the squatter areas who worked in both these sectors. A total of 13 percent of the respondents from the squatter settlements had temporary jobs, compared to 9 percent of those moving from standard housing. Those with permanent jobs comprised 51 percent of the respondents moving from standard housing, compared to 56 percent for those from the squatter settlements. The higher percentage from the squatter settlements was comprised of people mainly working in the informal sector, while those from standard housing were from the formal sector. Jobs in the formal sector are more stable and usually bring in a higher income than those from the informal sector. A total of 6 percent of those moving from the squatter settlements were unemployed, while only one percent of those from the standard housing were unemployed. The different jobs held by the respondents moving from standard housing to Dandora are shown in Table 6.1.
The distance to the place of work for these respondents ranged from those working in the settlements to those working 12 or more kilometres away. A total of 23 percent of the respondents moving from standard housing worked in Dandora, 4 percent worked 1-5 kilometres away, and 53 percent worked 6 kilometres away or more. The high percentage working six kilometres away or more is explained by the fact that most of the respondents worked in the formal sector. The majority of the jobs in the formal sector are located in the City Centre or industrial areas which are situated six or more kilometres
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3 1 5
Table 6.1 Jobs Held by the Residents who had Lived in
Standard Housing Prior to Dandora
Occupation No. from Sample Percent
Office Support Staff 52 29 Self-employed 39 22 Landlord 24 13 Electrician 11 6 Driver or Conductor 10 5 Building Trades 10 5 Hotel Services 9 5 Machine Operator 5 3 Musician 2 1 Barmaid 5 3 Factory Worker 2 1 Teacher 2 1 Unemployed 11 6 Total 182 100
Source: Field Data
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2 16
away. In contrast, those working six or more kilometres away comprised 30 percent of the respondents from the squatter settlements. This lower percentage in the squatter settlements is explained by the high percentage working within Dandora, because many of the respondents were self-employed in the informal sector.
A total of 25 percent of the respondents who had moved from standard housing walked to work, 40 percent took the public transportation system, 14 percent took employer- provided transportation while the remaining one percent had their own personal cars. A total of 53 percent did n pay for transportation, while the remaining 47 percent paid Kshs. 24-640. The high percentage of the resp idents paying for transportation is explained by the fact that most of these had to take public transportation to work. The high transportation cost would not have been affordable by low income earners group who would have to spend most or all of their income on transportation. Those not paying for transportation included a high 14 percent who used employer- provided transportation. Employer-provided transportation of this magnitude was necessitated by the distance of the project from the industrial areas and the fact that most of the respondents were shift workers. If no transportation was provided, production would be affected drastically, because most of these would have to depend on the public transportation systems which operate at times which do not necessarily coincide with the shift workers' schedule.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2 4 7
The incomes of the respondents moving from standard housing to Dandora were similar to those of the respondents moving from the squatter areas. Those earning Kshs. 1,000 or less comprised 27 percent of the respondents moving from standard housing to Dandora (see Figure 27). This compares with 29 percent for the respondents moving from the squatter settlements. The highest percentage of those earning Kshs. 1,000 or less we~e self-employed in the informal sector. This was followed by 3 percent of those moving from standard housing who were landlords. Others within this income group comprised 2 percent working in hotels, 2 percent office support staff, one percent machine operators, one percent barmaids and 3 percent unemployed.
Those earning Kshs.1,001-2,000 consisted of 34 percent of those from standard housing while 31 percent were from the squatter settlements. Similarly, those earning Kshs. 2,000 comprised u39 percent in standard housing and 4 0 percent in the squatter settlements. The income of those moving from the squatter settlements and those moving from standard housing were similar. This is because most of those moving from squatter settlements were plot owners who had high incomes. Also not all the residents in the squatter settlements have low incomes. Most of the residents from the squatter settlements cannot find standard housing to rent and so continue to live in the squatter settlements. This project thus played a significant part in making it possible for those
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Figure 27 Monthly Incomes for the Residents Moving
from Standard Housing to Dandora
Kshs. 1 0 0 M 5 0 0 17% A
Kshs. 1000 or Less k 27%
Kshs. 1501-2000 17%
Kshs. 2 0 0 0 39%
S o u r c e : F i e l d D a t a
2 1 9
from squatter areas to have access to standard housing in Dandora.
6.3 An Evaluation of Improvement in Housing Conditions as a Result of the Move from Standard Housing to Dandora
The respondents had lived in different locations in Nairobi before coining to Dandora. The highest percentage of respondents had lived in Eastleigh and consisted of 10 percent of all the respondents moving from standard housing. This was followed by 8 percent from Huruma, 6 percent from Kariobangi, 5 percent from Jericho and 5 percent from Pangani and Ngara (see Figure 28) . Another 7 percent lived in Ofafa Maringo and Shauri Moyo and Makadara, 11 percent lived in Makongeni, Mbotela and Kaloleni, Umoja, Kibera and the Outskirts of the City. Others were distributed in smaller percentages in different parts of the City (see Table 6.2). Most of the housing estates where the respondents who moved from standard housing had lived were constructed just before independence and were characterized by one-roomed units. They were located near Dandora in the eastern part of the City, the most densely populated low-income area in the City (see Figure 28).
The highest percentage (62 percent) of the respondents had moved from rented housing. A total of 31 percent had lived with relatives, or friends while 6 percent lived in employer-provided housing. None of the respondents moving
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2 5 0
Table 6.2 The Housing Estates the Respondents Lived in
Prior to Dandora
Location No. from Sample Percent
Eastleigh 18 10 Huruma 15 8 Kariobangi 10 6 Muthurwa 10 6 Buru Buru 9 5 Jericho 9 5 Bahati 9 5 Pangani/Ngara 9 5 Makadara 6 3 Ofafa Maringo/ Shauri Moyo 6 3 Makongeni/Mbotela 5 3 Umoja 5 3 Kibera 5 3 Mathare 5 3 Outskirts 5 3 Others 56 29 Total 182 100
Source: Field Data
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
R eproduced
with perm
ission of
the copyright
ow ner.
F urther
reproduction prohibited
w ithout
perm issio
n .
Figure 28
STANDARD HOUSING ORIGIN AREAS THE DANDORA RESIDENTS
Dandora Protect
Source MA/INGIRA 1983. w 1 Eastleigh 10% 10 Manngo 3% 2 Hururma 8% 11 Makongeni 3% 3 Kariobangi 6% 12 Umo|a 3% 4 Muthurwa 6% 13 Kibera 3% 5 Buru Buru 5% 14 Mat hare 3% 6 Jericho 5% 7 Bahati 5% 8 Ngara 5% H Citv Centre 9 Makadara 3%
2 5 2
from standard housing owned the housing they had previously lived in before moving to Dandora. As already stated, most of the housing constructed has benefited middle and high income households. The households moving to Dandora did not earn high enough income to qualify for mortgage housing being provided in the City. The project made it possible for 46 percent of all the respondent who moved from rental standard housing to own housing in Dandora. Except for 4 percent of the dwelling units, all the housing units the respondents lived in before moving to Dandora were built with permanent materials.
The number of rooms the respondents lived in ranged from one to nine before Dandora and one to seven in Dandora (see Figure 29). The number of rooms increased with the move to Dandora. Thus, those with one room decreased from 65 percent in standard housing to 54 percent in Dandora. Those with three to five rooms also increased, while those with seven rooms remained the same in both locations. Thus a total of 21 had more rooms in Dandora than in standard housing. The remaining 73 percent had the same number of rooms in both locations.
The number of persons living in these rooms ranged from one to twenty in standard housing, and from one to twenty-one in Dandora (see Table 6.3). The average household size was almost the same in the two locations (4.7 persons in standard housing and 4.5 persons in Dandora). This contrasts with an increase in the average household size amo.ig those moving from
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
R eproduced
with perm
ission of
the copyright
ow ner.
F urther
reproduction prohibited
w ithout
perm issio
n .
Figure 29 Number of Rooms per Household Before and
After Moving to Dandora
P e r c e n t 70
60
50
40
30
20 10
0 □ 2 3 - 4 5 - 6
N u m b e r o f R o o m s p e r H o u s e h o l d 7 - 9
Source Field Data
S t a n d a r d H o u s i n g I » D a n d o r a
i;
ra
2 5 4
Table 6.3 Number of Persons per Household in
Dandora Standard and in
Before Dandora Dandora No. from No. from
Persons Sample Percent Sample Percent
1 24 13 45 25 2 -11 23 13 7 3-4 34 19 44 24 5-6 35 19 38 21 7-10 33 18 35 19 11-15 13 7 3 2 16-20 2 1 4 2 Total 182 100 182 100
Source: Field Data
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2 5 5
the squatter settlements to Dandora; the average household size in the squatter settlements was 5 persons, compared to 6 persons in Dandora. Thus, while little change was experienced in the size of household with the move to Dandora for those moving from standard housing to Dandora, there was a significant change for those moving from the squatter settlements to Dandora. The number of those having adequate room increased from 16 percent in standard housing to 28 percent in Dandora. Thus an additional 12 percent of the respondents had adequate room as a result of the move to Dandora.
The number of persons per room also changed with the move to Dandora (see Table 6.4). A total of 30 percent of the respondents moving from one-roomed units in standard housing had one to two persons before moving to Dandora. This decreased to 27 percent in Dandora. A total of 35 percent had three or more persons per room. This decreased to 27 percent in Dandora. A total of 3 percent of the households had more persons per room in Dandora, while 8 percent had fewer persons in Dandora. The level of crowding for these 8 percent decreased with the move to Dandora. Ho significant changes were experienced by those having two rooms in both locations. Changes in the number of persons for those with three or more rooms were minimal.
The housing cost for the respondents moving from standard housing to Dandora increased significantly. A total of 37 percent of the respondents who paid no rent in standard
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2 5 U
Table 6.4 Number of Persons per Room in Standard Housing and in
Dandora
Standard Housing Dandora
Rooms Persons No. from Sample Percent
No. from Sample Percent
1 1-2 54 30 49 27 3-6 47 26 43 23 7-10 10 5.5 7 4 11-15 7 4 0 0
2 1-2 8 4 6 3 3-6 14 8 20 11 7-10 7 4 9 5 11-15 2 1 0
3 2-6 9 5 14 9 7-10 10 6 8 4 11-15 2 0.5 2 1
4 2-6 3 1.5 4 2 7-10 3 1.5 9 5 11-15 3 1.5 0 0 16-20 1 0.5 3 2
5+ 1-6 0 0 4 2 7-10 1 0.5 2 1 11-21 1 0.5 2 1
Total 182 100 182 100
Source: Field Data
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
housing paid for housing in Dandora. Most of these were accommodated by relatives or friends and did not pay for housing before moving to Dandora. The remaining 63 percent paid rents ranging from Kshs. 24-2,700 in standard housing. This compares with all the respondents in Dandora who had high housing expenses (see Figure 30). The rents paid before moving to Dandora were also low compared to those in Dandora. While 21 percent paid rents of Kshs. 200 or less in standard housing, only 4 percent paid such low rents in Dandora. The highest percentage of 66 percent paid Kshs. 201-500 in Dandora, compared to 28 percent in standard housing. The remaining 30 percent of the respondents paid Kshs. 500-1,500 in Dandora, compared to 12 percent in standard housing. None of the respondents paid more than Kshs. 1,500 in Dandora, while 2 percent paid more than Kshs. 1,500 before movj-ig to Dandora. Clearly, the housing cost for most of respondents increased with their move to Dandora. Some of the reasons for the high expenses in Dandora included the fact that this was new housing, while most of the standard housing was old. Also most of the households moving from standard housing had lived in one-roomed units. The housing expenses in Dandora were also high because of the high cost of services and the loan repayments to the City Council.
When compared to rents paid in the squatter settlements, the housing expenses for those moving from standard housing were higher. This is mainly because standard housing is of a higher quality and thus much more expensive than that in the
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
R eproduced
with perm
ission of
the copyright
ow ner.
F urther
reproduction prohibited
w ithout
perm issio
n .
Figure 30 Comparison of Rent Paid in Standard
Housing and in Dandora
P e r c e n t 4 0
30
20
10
0 I 1-100
I I □ 1 0 1 -2 0 0 2 0 1 -3 0 0 3 0 1 -4 0 0 4 0 1 -5 0 0 5 0 1 -1 0 0 0 1 0 0 1 -2 7 0 0
H o u s i n g E x p e n s e s in K s h s .
S t a n d a r d H o u s i n g I. ...I D a n d o r a
S o u rc e : F ie ld D a ta tc Cfl 00
2 5 0
squatter settlements. Housing expenses, however, increased for both those moving from standard housing and squatter settlements. This leads to the conclusion that, irrespective of where the households came from, housing in Dandora was much more expensive than where they had previously stayed.
Standard housing is constructed with the necessarv services (see Figure 31) . All Che respondents except one had water and toilet facilities in standard housing. A small percentage of 7 did not have shower facilities. Most of the respondents without shower facilities used the toilet for showering purposes. The large number of respondents having these facilities is due to the fact that most of the standard housing has the above required facilities. The respondents having kitchen facilities comprised 50 percent before moving to Dandora. This compares with 23 percent who had kitchen facilities in Dandora. Thus, 28 percent who had kitchen facilities before moving to Dandora did not have them in Dandora. Electricity was also available to 66 percent of all the respondents moving from standard housing to Dandora. This was reduced to 50 percent in Dandora. This is explained by the fact that most of the standard housing had electricity installed before occupation, while in r ndora the owners connected electricity to their plots wherever they could afford it.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
R eproduced
with perm
ission of
the copyright
ow ner.
F urther
reproduction prohibited
w ithout
perm issio
n .
Figure 31 Comparison of Available S ervices in
Standard Housing and in Dandora
P e r c e n t 120
100
80
60
40
20
0 W a t e r T o i l e t S h o w e r
S e r v i c e s K i t c h e n E l e c t r i c i t y
S e r v i c e s B e f o r e S e r v i c e s in D a n d o r a
Source: Field Data
260
26 1
6.4 Evaluation of Zndiraot Banafita to Residents of the Squatter settlements as a Rasult of the Nova to Dandora froa standard Housing
This section discusses housing benefit resulting from the filtering process, as this process can be inferred from migration history of the respondents moving from standard housing to Dandora. These moves include households who moved directly from standard housing to Dandora (see Figure 26) . Those moving from standard housing to Dandora comprised of 60 percent of alx the respondents. A total of 10 percent of these had lived in the squatter settlements before moving to standard housing (see Figure 26). The residents moving from the squatter settlements to standard housing represent a small percentage of all the respondents. This leads to the conclusion that the project did not indirectly benefit residents from the squatter settlements to an appreciable extend.
The project directly benefited a higher income group than was targeted for this project. A further examination of the data, however, revealed an indirect benefit of the project to lower income groups through the filtering process. The term "filtering process" is used as defined by Hoyt who defined this process as the " the move up a step leaving the oldest and the cheapest houses to be occupied by the poor families" (122). This definition involves the movement of households to better housing and creation of vacancy chains which in turn benefit poor families. To assess whether this process took
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2 6 2
place three questions were addressed. The questions addressed were as follows: a. Did the residents moving from standard to Dandora begin
a vacancy chain? b. Where did the households entering the vacancy chains come
from? c. What was the indirect impact of this project on the
growth of the squatter settlements?
6.4.1 Did the People Moving from Standard Housing to Dandora Begin a Vacancy Chain?
To determine whether the respondents left the housinr they had previously occupied vacant they were asked whether anybody moved to the housing that they had left behind. The survey showed that most of the housing units were left vacant by the households that moved to Dandora. A total of 85 percent of all the standard housing previously occupied by the current Dandora residents was left vacant for occupation. The remaining 15 percent was occupied by other households (see Figure 32) . The move to Dandora thus not only enabled the households to have access to better housing, but also left housing behind that could have benefited other households. The characteristics of the housing units left behind suggests that the units benefited low-income households. This is indicated by the rent paid for this housing and the location of the housing estates. Most of the dwelling units were located in the eastern part of the City, where low-income housing constructed before and after independence is located.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
R eproduced
with perm
ission of
the copyright
ow ner.
F urther
reproduction prohibited
w ithout
p e
rm issio
n .
Figure 32 Housing Left Behind by Respondents
Moving from Standard Housing
Housing Left Vacant 85%
Replaced by Others 9%
Units Shared Before 6 %
S o u rc e : F ie ld D a ta 263
26 4
A total of 65 percent had also moved from one-roomed units. Most of the one-roomed units were constructed before independence as well. This policy was changed after independence and the minimum dwelling unit had to have two habitable rooms. Also most of this housing was developed for low-income households but the lack of standard housing in the City has meant that these housing units have been occupied by middle-income groups. Dandora project was attractive to these middle-income households because they could afford the type of housing provided by the site and service schemes. Low- income households could not afford housing in the site and service projects because of their low incomes. But the move to the Dandora project left housing behind that could have benefited them. Looking at the rents paid before the move to Dandora, standard housing units left behind were much cheaper than the dwelling units in Dandora. The standard housing was also of a much better quality and had the basic services unlike the housing available in both the squatter settlements and the rural areas. The households that moved had their housing significantly improved. As indicated in the following section most of these housing units are inferred to have indirectly benefited new migrants to the City. Those migrating to the City come from the rural areas where housing is of a lower quality and lacks the basic minimum services.
The move to Dandora benefited the households that moved by reducing the number of persons per households and by
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2 6 5
increasing the number of rooms available. They also had their income stabilised by the added income from subletting. The side effect of their moving was that they left housing that could have benefited low-income families. If it is assumed that the pattern whereby the provision of site and service projects facilitated the movement of low-middle income households who leave behind housing initially meant for low- income households, provision of these projects indirectly benefit low-income households. An increase in these projects is likely to indirectly benefit more low-income households.
A total of 15 percent of all the housing units left behind by the residents moving from standard housing were not left vacant. The move to Dandora either improved the housing conditions of the households left behind or provided room for another person. Given the lack of available housing in Nairobi the person moving to replace thore moving to Dandora would have had no choice but either crowd in the same household or move to the squatter settlements. A total of 6 percent of this housing had been shared with others before the move to Dandora. Since no other persons moved to replace the persons moving to Dandora, the move benefited the households left behind by reducing the number of occupants, thus reducing the level of crowding. Overcrowding is a serious problem in Nairobi and has been for a long time. Given the rapid increase of the population and the slow development of housing, the situation has not improved significantly. Since no other large-scaZe low-income projects had been implemented
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2 6 G
before the Dandora project the crowding levels seem to have remained largely unaffected. Although the percentage of households where crowding was reduced was small, it is significant because of the high crowding levels experienced by most households in the City. If site and service projects are increased, more families are likely to move to these projects and not only improve their housing conditions but indirectly improve the housing conditions of the families left behind who would also be less crowdec
A total wf 9 percent of the households did not experience any change as result of the move to Dandora. This represents situations where the respondents moving to Dandora were replaced by others. Although their move did not create more room for those left behind, it created room for others who would either have had to come to the same unit, crowd with others in another unit or move to the squatter settlements. From this point of view the move to Dandora improved the housing situation of the households who were left behind, and also made it possible for households, who may have had no choice but to live in the squatter settlements, to have access to standard housing.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
267
6.4.2 The Origin of Households Occupying Residences Made Available as a Result of the Move to Dandora
The lack of current data on the occupants of the housing left vacant by the Dandora residents led to the use of migration data gathered from the survey and inferring from that data the origins of the households entering the vacancy chains. The trends in the migration history were assumed to be indicative of the subsequent migration pattern followed by the households, moving as a result of the provision of site and service projects. The data showed that most of these residents had initially moved from the rural areas to standard housing befcre moving to Dandora. Those moving from the rural areas to standard housing and then to Dandora comprised 55 percent of all respondents moving to Dandora. A total of 51 percent of all those moving from standard housing had moved directly from the rural areas to standard housing before Dandora. The remaining 49 percent had moved from the rural areas to two different standard housing estates before moving to Dandora (see Figure 33) . The pattern displayed in this case shows a situation where the move to Dandora indirectly benefited respondents who had moved from the rural areas to standard housing before moving to Dandora. If it is assumed that the same migration pattern is indicative of what has been happening as a result of this project, one can infer that the rural households filled the vacancies that were left in standard housing.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2 6 8
Fig'jr* 33
MIGRATION HISTORY OF THE RESIDENTS MOVING FROM THE RURAL AREAS TO STANDARD HOUSING PRIOR TO DANDORA
Standard Housing
tn
Dandora
Standard Housing
\V, &
A. n
1 131 CD V,>
i ! i
Rural Areas
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2 6 9
This point is further reinforced by an examination of the question of whether the residents knew the person moving to the housing units they had previously occupied. According to the survey, most of the households did not know the origins of the people who moved into the housing unit they had vacated. A total of only 12 percent knew the origins of the households that moved to their previous housing units. A total of 11 percent of these residents were new migrants from the rural areas whilt. one percent had moved from the squatter settlements. If this is assumed to be the pattern followed for the housing left behind, one can infer that the housing left behind mainly benefited new migrants to the City. Thus, the respondents who moved to Dandora from standard housing mainly left vacant housing which, as indicated, mainly benefited the new migrants.
6.4.3 The Indirect Impact of the Dandora Project on the Growth of the Squatter Settlements
As indicated by the analysis in the previous -section, the project indirectly benefited households who had initially moved from the rural areas. The number of migrants from the rural areas to the city is of major concern Decause rural to urban migration has contributed significantly to the growth of urban areas. The change of urban policy after independence allowed free movement to cities. Nairobi being the biggest city with the most job opportunities attracted most of the
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2 7 0
migrants. Given the lack of suitable jobs in the rural areas, the number of migrants to the City has not decreased. This has led to a significant increase in the number of households in Nairobi. As already stated most of the new migrants locate in the squatter settlements when they move to the City. A total of 80-90 percent of new households each year are accommodated in squatter settlements (Andreasen 1987, 98). Most of these new households are migrants to the City (Obudho 1987, 10). As already stated, the housing development in the City has not kept pace with the increase of migrants. If these projects had not been implemented, most of these households would have had to stay in the squatter settlements or crowd in with friends in the existing housing which is already highly crowded. When seen from this point of view, these projects have significantly affected the growth of the squatter settlements. This would lead to an increase of the number of people in the squatter settlements. Since population movements from the rural areas are not controlled (Obudho 1987, 9), these respondents would have come to the City anyway.
The survey showed that most of the respondents moving from standard housing to Dandora had indirectly moved from the rural areas. Thus a total of 55 percent of the beneficiaries of this project had moved to standard housing before moving to Dandora. If it is assumed that these same migration patterns will be followed by the new migrants to the City, one can conclude that these projects, though not
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
27 1
directly meeting the housing need of the target income group the projects, have met the housing need of households who would have had no alternative but to move to the squatter settlements which would facilitate the growth of the squatter settlements. In addition, a total of 13 percent of all the respondents moving to Dandora had directly moved from the rural areas. In total, 68 percent of all the households moving to Dandora had directly or indirectly moved from the ru. al areas. The highest percentage however had indirectly moved from the rural areas. Given the high cost of these projects, one can see that they will not directly affect the growth of the squatter settlements but will indirectly affect the growth of the squatter settlements. Thus, although squatter settlements are increasing, if these projects were not implemented, the number of the squatter settlements would be increasing at a faster rate. This is because most of the respondents would have to live in the squatter settlements or crowd in the existing housing, which is already overcrowded. This strategy may not directly have met the housing need of low-income households, yet if the project had not been implemented, the squatter settlements would either have increased at a faster rate or the existing housing would have been more crowded. Given the large migration of the rural households to Nairobi, if no measures are taken to accommodate some of these households, many will continue to live in the squatter settlements.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2 7 ■>M ( M
In addition to the Dandora project, over 15,000 other site and service plots have been implemented in Nairobi. Given the fact that Dandora was the first large scale site and service project to address low-income households, the impact of this project has significantly affected the growth of the squatter settlements. Thus although the households moving to Dandora may belong to a higher income group, they are releasing housing that would benefit new migrants to the City and thus curbing the rate of growth of the squatter settlements.
6.5 Conclusions
The high number of housing units left vacant is indicative of a pattern whereby the provision of site and service projects created chains of vacancies that were inferred to benefit low-income households. This is especially true given the fact that most of the housing units left vacant were one-roomed units, which would usually be occupied by low- income households. The high cost of site and service projects will continue to make them inaccessible to low-income households.
The migration histories of the respondents showeo a pattern whereby most of the residents moving to the housina left vacant had moved from the rural areas. New migrants to the City usually locate in the squatter settlements when they
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
27
first move to the City because that is the only available and affordable housing for them. The Dandora project created vacancy chains that are inferred to have benefited these households. If no horsing had been released, these households would have had no choice but to either move to the squatter settlements or move to the existing housing which is already overcrowded. Given the fact that other sire and service projects developed along the lines of the Dandora project have been increased, site and service projects have significantly affected the growth of the squatter settlements. The current housing market situation where the demand is greater than the supply leads to a situation where new housing is too expensive for low-income households. The provision of the site and service projects, however, causeu movement of households from cheaper older housing which low-income households could benefit from. The increase of these projects would significantly affect the development of the squatter settlements because so far they represent the quickest way of providing housing to households that are not catered for by the existing housing finance agencies. The families benefiting from site and service projects not only moved to better housing but they also made rental housing available at a fast rate than any other project since independence.
The rapid growth of the cities of the Developing Countries has reached such proportions that governments are unable to meet the housing need of the urban dwellers. The involvement of people in solving their housing problem through
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
remaining 10 percent had seven to fourteen persons per room. The respondents with two rooms (18 percent) were less crowded. Thus 3 percent had one to two persons, 7 percent had three to six persons while remaining 6 percent had seven to twelve persons. The remaining 22 percent of all the respondents had three to seven rooms. A total of 3 percent of these were not crowded and had one to five persons. The remaining 18 percent were crowded and some of the households had as many as twenty persons. The above figures show high level of crowding before the move to Dandora. This crowding was experienced in both standard and squatter housing. The number of rooms increased slightly with the move to Dandora but the size of households have been increasing at a faster rate, thus making crowding levels high in Dandora as well. The difference in Dandora is the fact that all the respondents had permanent housing and had the required minimum services while some of the respondents did not nave these facilities before moving to Dandora.
Housing cost before moving to Dandora was minimal with a high 37 percent paying no rent. This was because they either lived with relatives or friends or moved directly from the rural areas (see Table 4.19). The average rent paid per month was Kshs. 184 with the percentage of 3'/ paying no rent while 43 percent paid Kshs. 15-300. This relatively low rent is explained by the fact that these were paid a few years ago, and some of the respondents had moved from the squatter areas which have the lowest rents in the City. A total of 19
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
/I OF/DE M i .
Reproduced with permission of the copyright owner. Further reproduction prohibited w ithout permission.
27 1
site and service nrojects is so far one of the most significant ways of beginning to deal with the acute housing shortage in the cities of the Developing Countries.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER 7
SUMMARY AMD CONCLUSIONS
7.1 Introduction This research has focused on the site and service
projects as a strategy for meeting the housing needs of the urban poor in Nairobi, Kenya. The study assessed who the beneficiaries of the Dandora site and service scheme were and whether the respondents benefited directly as intended or indirectly through the filtering process. The socio-economic characteristics and the housing conditions of the beneficiaries are also discussed. In this concluding chapter, the key findings from the study are summarized and evaluated in terms of the objectives set above. The socio-economic and the housing characteristics of the beneficiaries are discussed in Chapter 4. Information from Chapter 5 and 6 is utilised to help evaluate the characteristics of those who benefited directly or indirectly through the filtering process.
7.2 summary 7.2.1 Household Characteristics of the Dandora Residents
The beneficiaries of the project included both male and female households heads aged between 18 and 90 years. These had lived in Nairobi for periods ranging from three months to
275 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2 7 0
sixty years. They represented older households heads than those currently in the City where most of the people are 30 years or under.
The beneficiaries of the Dandora project included 56 percent married and 44 percent single household heads as well. A total of 3 3 percent of the married respondents lived with their spouse in the urban areas while 21 percent had left their spouse in the rural areas. This represents an increase in the number of spouses living with their families in the City since independence. In addition, a total of 85 percent had children ranging from one to sixteen. This represents large household si*es in the City. This growth of households has not been accompanied by an increase in the number of rooms; most of the households continue to occupy one room. The increased number of families living in the urban areas, combined with the lack of facilities, is one of the major reasons for overcrowding in the Dandora project. The reasons given for leaving the family in the rural areas included ’.ow income and lack of adequate housing.
The Dandora settlement mainly housed families with stable incomes. A total of 80 percent of the respondents had occupations in the formal and the informal sectors while 16 percent had the plots as their only source of income. Those not employed comprised 4 percent and these had their housing expenses paid by family or friends. Mos«- of the respondents working away from the settlements either walked, used public transportation system, or had employer provided
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2 7 7
transportation. A total of 28 percent walked to their place of work, 38 percent took the public transportation while 12 percent used employer-provided transportation. The highest percentage of the respondents took public transportation. The public transportation system was too expensive for low- income groups, most of whom would have to spend over 50 percent or more of their income on transportation. Given that the housing costs are much higher than the transportation costs, the distance of this project from the mf’n jmployment sectors meant that most of the low-income families could not afford to live here and work where they did.
Most of the respondents earned higher incomes than intended for this project. A total of only 28 percent of all the respondents earned Kshs. 1,000 or less 72 percent earned over Kshs. 1,000 with a high percentage of 37 percent being in the middle income group and earning more than Kshs. 2,000. Thus, this project mainly benefited higher income households, and this left the lower income groups without housing. Low- income groups had not benefited from previous housing provided, because of their low incomes. These households have continued to reside in the squatter settlements where the rents are low.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2 7 8
7.2.2 Housing Conditions as Result of the Move to Dandora
The housing quality and the provision of ser/ices changed as a result of the move to the Dandora project. All the housing units in Dandora were constructed using permanent materials while a total of 35 percent of the respondents had lived in temporary housing before moving to Dandora. The project also provided water, toilet and shower facilities to all the residents. This was not the case before the move to Dandora where a total of 7 percent did not have water and 6 percent did not have toilet facilities. A total of 30 percent did not have shower facilities before the move to Dandora. The move to Dandora made these services available to the respondents. The percentage with electricity was the least affected, with 44 percent having electricity facilities before moving to Dandora compared to 46 percent after the move to Dandora. Kitchen facilities were least affected and actually decreased with the move to Dandora. Kitchen facilities were almost nonexistent to low-income families before independence and this situation has not changed significantly. As discussed later, the urban housing market in Nairobi is characterized by predominantly one-roomed units1. This trend is likely to continue given the fact that the housing units
Although the one-roomed housing policy was changed to a minimum of two rooms, most residents in the City continue to sublet one room from existing housing because they cannot afford to rent the whole housing unit.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2 7 9
are increasing at a slower rate compared to the rapidly growing population.
The number of rooms did not increase significantly with the move to Dandora. Only 15 percent of the households had more rooms in Dandora. The rest had fewer rooms or did not experience any change in the number of rooms as a result of the move to Dandora. All the rooms in Dandora, however, were constructed with permanent materials, while a tota- of 35 percent of the respondents had moved from housing constructed with temporary materials. Whether these had more rooms in Dandora or not, the move was an improvement because most of the temporary houses are of poor quality.
The increase in the number of rooms was mainly felt by the plot owners, while almost all the tenants still lived in one room. In total, 19 percent of the plot owners had adequate room before moving to Dandora, compared to 26 percent in Dandora. Thus a total of 7 percent of the plot owners who did not have adequate room before moving to Dandora now had adequate room. The above figures indicate few households with adequate room in Dandora. The problem of overcrowding is not unique to this settlement, it is experienced throughout the City due to the few housing units available. The crowding is however more acute in the low-income areas. Given the continued lack of low-income housing, this trend is likely to continue. This situation is not only aggravated by low housing production but also by the low income of the households. The number of housing units would have to
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2 8 0
increase at a rapid rate for this situation to change. One big project like Dandora could not make the expected change.
Another change experienced as a result of the move to Dandora was the increased housing costs for both plot owners and tenants. For example, a total of 37 percent of all the respondents did not pay for housing before moving to Dandora while all the households in Dandora paid rents. Those paying Kshs. 300 or less comprised 44 percent before Dandora, compared to 20 percent in Dandora. Those paying Kshs. 300- 500 comprised 13 percent before moving to Dandora. This compares with 28 percent in Dandora. Similarly, those paying more than Kshs. 500 comprised 6 percent before moving to Dandora, and compares wif.h 33 percent of all the respondents in Dandora. The high rents charged by the plot owners were due to the scarcity of standard housing in the City, leaving most of the residents with no choice but to pay the high rents or stay in the squatter settlements where the rents are cheaper. The plot owners' housing expenses were higher than the rents charged. This was not because of the repayment of loans to the City Council, but due to the cost of services. The cost of services, especially water, was higher in Dandora because of the large number of persons per plot. The average number of rooms was six and the average number of persons per plot was 13 persons. The large number of the persons per households has made services much more expensive than was anticipated, thus making this housing above the reach of the intended income group. These high expenses mean that these
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2 8 1
projects will continue to benefit a higher income group than intended.
The households moving directly to Dandora from the squatter settlements comprised 27 percent of all the respondents. Those moving directly from the rural areas accounted for 13 percent of all the respondents. Most of these households lived in housing constructed with temporary materials before the move to Dandora. Thus 90 percent of all the households living in the squatter settlements lived in temporary housing compared to 69 percent from the rural areas. Most of these housing units lacked the basic services like water toilet and shower facilities. The housing expenses in the squatter settlements were cheapest in the City while most of the housing in the rural areas was free.
The highest percentage of the respondents moved directly from standard housing to Dandora. This comprised 60 percent of all the households moving to Dandora. These households lived in housing constructed with permanent materials and having basic services. The move to Dandora improved their housing situation by increasing the number of rooms available to the households.
7.3 Conclusions
The site and service projects were a response to the housing need of the low-income households. The housing need
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
of these households had not been met by the provision of conventional housing provided since independence. The housing units constructed then were not only too expensive but too few. The site and service housing strategy thus addressed the need for low income housing and the high cost of its construction. It was hoped that the construction costs would be reduced by the involvement of the residents in the construction process. The main objectives, identified in chapter One, are as follows:
Meeting the housing need of low-income households Incremental development to enable low-income households to build whenever they could The reduction of housing cost through the residents' involvement The implementation of affordable services The provision of security of tenure The elimination of future squatter settlements
The dissertation evaluated whether the project met the above objectives and whether the project benefited the residents directly or indirectly through the filtering process. The following section wi'1 discuss how well these objectives were met.
7.3.1 Did the Site and Service Schemes Address the Housing Needs of Low-income Groups?
The survey showed that the project benefited households earning more than Kshs. 1,000 For example, those earning
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Kshs. 1,000 or less comprised 28 percent of all the respondents. The remaining 72 percent earned higher incomes. The high housing expense in Dandora made it impossible for those earning Kshs. 1,000 or less to afford the housing provided. The average housing expenses per month was Kshs. 469. If one was not to spend more than 25 percent of their income on housing as intended, the average income of those living in Dandora should be Kshs. 1,876. In addition, the plot owners would need to earn higher incomes to be able to repay their loans. The high housing expenses are partly attributed to the high cost of construction which was not affected by the use of Grade 2 By-laws as expected.
A further examination of the data however, led to the conclusion that the project indirectly benefited low-income households. This was indicated by the fact that most of the housing units left behind by the Dandora residents were left vacant for occupation by the other households. The nature of the housing units suggests that they could have benefited low-income households. Also the migration pattern of the current Dandora residents show a situation where most of the respondents moved from the rural areas to standard housing before moving to Dandora. Most of those benefiting from the housing left behind are inferred to have moved from the rural areas. Most new migrants usually locate in the squatter settlement since that is the only housing available to them. These households would have had no alternative but to live in the squatter settlements if no other vacant
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2 8 4
housing was available to them. From this perspective the project indirectly met the housing need of households who would have had no alternative but to live in the squatter settlements and in so doing increase the growth of squatter settlements.
7.3.2 Did the Project Allow Incremental Housing Development?
Given that most households could not afford to construct all the rooms at the same time, progressive development was adopted allowing households to construct most of the rooms whenever they could. As discussed in chapter 4, the respondents did their construction whenever funds were available. This enabled many households to benefit from this project. Given the fixed capital, the construction by incremental development benefited considerably more families than would have been possible by financing complete standard dwelling units. This is because the former requires a lower initial investment than traditional conventional housing. Also, construction by incremental development accelerated capital formation by urban low-income families. Incremental development also enabled the families to maintain control over the phasing, the programming and the construction of the dwelling unit, and to combine the monetary and the non monetary resources at their own economic convenience. This is unlike traditional housing, where housing credit
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2 8 5
mechanisms consider only the family monetary resources as a criterion for programme participation. Another complimentary advantage of home ownership and construction by incremental development is the flexibility of the schedule for family investment and expenditures for housing and other goods and services as preferred. This is necessitated by the fluctuating monetary resources of low-income households.
7.3.3 Were Housing Costs Reduced Through Self-help and Residents' Involvement?
By 1970, it was realised that the government had no resources to meet the housing need of the ever-increasing number of households, especially those in the low-income groups. So far, these households have provided themselves with housing in the squatter settlements. The implementation of site and service schemes policy was a realization that if any meaningful increase of housing was to be experienced, the households had to be involved in the housing production. Thus, the purpose of this strategy was to ensure that the eligible participants contribute as much of their own effort, skills and financial resources as possible to their own dwelling. The people's involvement was to mobilize finances and to provide cheap labour.
The involvement of building groups was minimal in the project. Only 7 percent of all the respondents were involved in the building groups. They mainly contributed money and
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2 8 6
not actual labour. Those providing all the required unskilled labour comprised 9 percent of all the respondents. The remaining 91 mainly hired labour and provided minimum labour themselves because they were employed, and thus limited their availability. The assumption that the respondents would provide their own labour ignored the realities of most low- income earners who have to work most of the time including weekends. In addition to the many problems associated with the building groups, most of the plot owners could not have the time to be involved and at the same time have the finances for construction.
The average room cost does not reveal any reduction of cost for the Dandora residents. In the 1970s, the average cost for a two-roomed habitable unit with services was estimated at Kshs. 24,000. This compares unfavourably with the average cost of one-room, excluding services, which was Kshs. 16,918, making the average cost for a unit Kshs. 101,508 (the average number of rooms was 6). The above figures show very little involvement of the respondents in providing the expected cheap labour. Also the housing construction cost was much higher than expected.
One of the problems facing the government was the availability of financing for housing. People's involvement in Dandora enabled the beneficiaries to use their own savings and other sources of finance that would not have been available previously. It has been estimated that the average cost to the government of this project was 15 percent. Thus
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2 8 7
the plot owners mobilised 85 percent of the funds used for construction. This represents Kshs. 12,942,610 for 150 plots. When all the 6,000 plots are included, the average amount of funds provided by the respondents, from the different sources, was over Kshs. 6 billion. This is very significant capital from the respondents and represents the highest investment in one low-income project. As already stated the plot significantly increased the number of rental rooms in the city at a low cost to the Central government and the local authority.
Other than providing funds for housing construction, the plot owners have played a very significant part in making rooms available for tenants. Using the average number of rooms per plot, the project made available approximately 36,000 rooms in Dandora. Given the fact that 80 percent of these are occupied by tenants, a total of 28,800 rooms are rented out to other people. This is the highest number of rooms in this category provided since independence. Most City residents need rental housing, which has been scarce since independence. This project met that need for a significant number of households.
7.3.4 Did the Project Provide Security of Tenure?
The project did provide security of tenure for plot owners, but this security was not for those intended. It had been anticipated that the plot owners would be low-income
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2 8 8
people moving from the squatter settlements who would pool their resources and invest in permanent housing development.
7.3.5 Did the Project Provide Affordable Services?
Although the project provided services to many respondents, they were expensive. Water bills were the highest in the City because of the large number of people per plot (averaging 13 persons). Electricity connec .on to the plots was also costly. Thus, more than half of the respondents did not have electricity connected to their sites.
7.3.6 Did the Project Contribute to the Elimination of Squatter Settlements?
One of the major objectives of this strategy was elimination of squatter settlements. The survey showed that a small percentage of 27 percent benefited directly from this project by moving from the squatter settlements to Dandora. Thus, the project did not directly benefit a significant number of households from the squatter settlements as intended. In addition the households moving from the squatter settlements had higher incomes than those intended for the project. The lack of alternative housing for these households had made it hard for them to move to better housing although they could afford better housing. The Dandora project provided these families with better housing and better
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2 8 9
services than they had before. Most of the household heads from the squatter settlements were plot owners and the move to Dandora enabled them to have access to added income from subletting thus stabilising their incomes. A total of 6 percent of all the respondents benefiting from the Dandora project had indirectly moved from the squatter settlements. These respondents lived in the squatter settlements when they moved to Nairobi, they then moved to standard housing before moving to Dandora. In total, only one-third of all the respondents directly and indirectly moved from the squatter settlements to Dandora. This is not a significant percent given that these were the intended beneficiaries of this project. The high expenses of this project made it inaccessible to most squatters because of their low-income.
A further examination of the data, however, showed that lower income groups benefited indirectly through the filtering process. Most of the households moving to Dandora had lived in standard housing prior to the move to Dandora. An examination of the migration history data showed that these households left most of the dwelling units they had occupied vacant. Thus a total of 85 percent of all the respondents left vacant the housing they had previously occupied. The move to Dandora thus facilitated the freeing of housing that could have benefited low-income households. The implementation of the project made housing available to other households who would have been unable to find standard housing in the City. This is especially true given the fact that most
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2 9 0
of the respondents had been unable to move to other housing units since their move to Nairobi because no suitable housing was available. Most of these households had also occupied housing that had initially been built for low-income families. These dwelling units had been occupied by the middle income group and thus making them unavailable to low-income households. The Dandora project thus not only provided better housing for the families that moved but also freed housing that could have benefited low-income households.
These benefiting from the housing left vacant are inferred to have been rural migrants. The migration history of the respondents indicated that most of those moving from standard housing had originally moved from the rural areas. If the same migration pattern is assumed to be indicative of the patterns followed by those benefiting from this project, one can assume that the project indirectly benefited new migrants to the City. However, there is some considerable time between the second tier moves of the Dandora residents and those who would have moved into their prior residents and this inference is somewhat needing further research. However, this point is reinforced by the fact that the households who knew the persons moving to the unit they had previously occupied, stated that most of these had moved from the rural areas.
As well previous studies have shown that most of the new migrants to the City usually locate in the squatter settlements because that is the type of housing that is
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2 9 1
affordable and available. The number of those moving to the city have significantly increased since independence thus increasing the growth of the squatter settlements. If this project had not been implemented, most of the households moving to the City would either have moved to the squatter settlements thus facilitating their growth, or have moved in to live with their friends in the already highly overcrowded existing standard housing. This leads to the conclusion that the project played a role of diverting the flow of rural to urban migrants from the squatter settlements and thus slowing down the growth of these areas.
In terms of eliminating squatter settlements, the Dandora project was just one of the projects which played a part in curbing the growth of these settlements. The population growth in the city has been faster than the growth of standard housing and it would be unrealistic to expect that the problem of squatter settlements could be eliminated by this strategy. Since the implementation of the Dandora project another 15,000 site and service plots have been implemented. Meanwhile the number of new households are estimated to be 14,800 per year. The site and service projects are clearly not growing at a fast enough rate to stop the growth of squatter settlements. The site and service strategy is however, the first large scale project to benefit people who would have had to live in the squatter settlements and in this, did better than most of the housing developed up to this point which had not reached
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2 9 2
low-income families and had limited effect on the squatter settlements.
The above discussion shows the significant part that the site and service projects have played in housing lower income households. The low rate of housing construction in the city for all income groups, and especially housing for low-income households makes the project worthwhile whether it benefited the squatters or not. Clearly the housing conditions of the households that moved improved and the move also provided rental housing which has housed many households who would have had no standard housing available to them in the city.
7.4 Recommendations
Despite missing some objectives, I would recommend that the number of site and service projects be increased. This will provide needed cheaper rental housing hopefully directly but more likely indirectly. A large increase in these kinds of projects could enable the rents to go down and benefit a lower income group. If the Dandora project had not been implemented, many households would end up living in the squatter settlements because of the low level of standard housing development. From this point of view, these projects will affect the development of squatter settlements as initially intended by the implementors of this project.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2 9 3
The income level should be raised to include those earning incomes that can enable them to afford to construct these kind of plots. Given the income levels in Nairobi, these projects are unable to directly benefit low-income households. To benefit low-income groups directly, low-income rental housing is required. The provision of housing with low rents would mean reducing the housing standards and allowing construction of single room rental units. Although this was abolished at the time of independence, most of the urban residents continue to live in one-room units because that is what they can afford. Allowing construction of one room rental units would allow the City Council to construct rental housing and enable them to charge the stipulated rents. Some major improvements are also required in these settlements. To avoid the bad name associated with these settlements, it is recommended that trees be planted and a higher standard of roads be adopted. The present road network ignored the heavy traffic that has characterized these settlements. These projects will continue to house many residents, and this factor should be addressed in the planning stages.
Given the slow rate at which the site and service projects have been implemented, it is also recommended that the upgrading of the existing squatter settlements be adopted. Although the implementation of squatter upgrading programmes has been stated as one of the policy goals, this has not been implemented successfully in Nairobi. Upgrading of squatter areas has been slow and has not received the necessary support
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2 9 4
from the policy makers and implementors. Some of the projects that have been implemented have faced unnecessary delays due to the land ownership patterns in these settlements. With the political will, the land ownership problems can be solved. This is especially true given the fact that some of the squatter housing is in state owned land.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
appendix I G R A D E II BY-LAWS
L*0»l Nonet NO 16 >131
T H E L O C A L G O V E R N M E N T R E G U L A T IO N S 1963
f L. S. 156 of 19631
IN EXERCISE of the powers conferred by regulation 210 of tbs Local Government Regulations 1963. the Minister for Local Government hereby makes the following Order: —
THE LOCAL G O V E R N M E N T (ADOPTIVE BY-LAWS) .GRADE ir BUILDING) O R D E R 1968
I This Order may be cited as the Local Government (Adoptive By-laws) (Grade II Building) Order 1968.
2. Tbe By-laws set out in the Schedule to this Order shall be .he Adoptive Grade II Building By-laws which any municipal or countv council may adopt
SCHEDULE Psnr I— PkELfWNAitr
1. These By-laws may be cited as the Local Government (Grade II Building) By-laws 1968.
2. In these By-laws, unless the context otherwise requires— "building" means anv structure movable or fixed, of whatsoever
k.nd or any part thereof intended to be used as a dwelling house or shop or anciiliary thereto and includes drainage wo'ks and excavation:
"building line" means a line drawn across a plot so thai no building or permanent structure, except a boundary wall or fence of approved design enclosing the plot, may be erected with.n the area contained between that line and the plot frontage:
"council" means a municipal council or a county council. "clerk" means a clerk to the counc:!: "dwelling house" means a building designed for use exclusively
a one self-contained residence, together with such out-bjiiu.ngs as ire an ancillary thereof.
"habitable room" means a room constructed or adopted as a living or sleeping room.
"plot" means jnv piece or parcel of land whether demarcated by survey or not: and
“residential area" means an area of land which has been approved by the council for use for residential purposes.
3. These By-laws shall apply to all land within tbe council's area of jurisdiction except where otherwise specified by the eouacil after having obtained ihe approval of tbe Commissioner of Lands— and except where tbe Local Government tAdopuve By-laws) t Building) Order has been applied to an area.
295
CiUtton
Interpretation
Application
LN. 15/1969
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Erection ot buildings not to be approved in certain circumstances
Erection of building!.
, ^ e council shall not approve the erectioo of any building which is to be erected in contravention of these By-laws or where—
(u) the land concerned is unsuitable for any reason for the development purposes;
(6) that the plot is located outside tbe boundaries of an existing or proposed municipality, township, trading centre, market or residential area;
(c) the proposal conflicts with the proper planning of the area; (d) the site concerned forms part of an area for which an approved
comprehensive lav out is. in the opinion of council, desirable
5. Every person who proposes to erect a building on any land within the area specified under by-law 3 of these By-laws shall comply with tbe requirements of these By-laws, and for the purposes of these By-laws any of tbe following operations shall be deemed to be the erection of a building after the date on which these By-laws become operative—
(a) the erection of any new building; (b) the erection of any addition to an existing building; (c) the re-erection or alteration of any part of any existing
building; id) the roofing over of any space between walls or buildings; <e) the changing of any purpose or purposes for which a building
or part of a building or appurtenances of a building are used:
</) the using for human habitation of any building or part thereo' which has not been previously used for that purpose:
'?) for using of any building in a manner different from 'hat shown on the plans thereof approved by the council whether before or after the date on which these By-laws become operative, and whether or not it is proposed to execute an> alterations or work in connexion with the proposed change.
Ih) tbe carrying out of any water service or drainage works: (i) any other work which involves the use and assembly of building
materials to a structure in any form whatsoever
Sitingot 6 . No building shall be sited on a plot otherwise than m a building. accordance with the approval of the council.
M inim um areas of plot and buildings ihereon.
7 (1) Except where otherwise approved by the Commissioner of Lands, no plot shall be less than 2.800 sq. ft. in area and not'more than quarter of one plot shall be built upon. In calculating the area of the plot which is built upon, tbe verandah or anv part of the plo> which is not open to the sky shall be included.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2 9 7
(2) No building shall be erected within 5 ft. of a boundary of rbe plot oa which it staods unless the council expressly so authorizes m aay particular case:
Provided that— (i) a latrine may be sited on the line of a back boundary or on a
side boundary of a plot if it forms part of a semi-detached building containing any other latrine on an adjoining plot; and
(ii) buildings constructed of grass or other inflammable material shall be sited not less than 10 ft. from any side boundary.
8. (1) Every dwelling house must be provided with a latrine of a type approved by tbe council.
(2) A pit latrine shall be at least 20 ft. in depth from ground level to the bottom of tbe pit. and shall be provided with a roof the heigh' of which shall be at least 6 ft. 6 in. from the floor to the underside of the roof or ceiling. A pit latrine shall also be provided with a concrete stance and with a fly-proof cover.
(3) A latrine shall be sited in a position approved by the council and shall not be nearer than 30 ft. from any habitable room, or room used for tbe preparation, cooking or storage of food:
Provided that, in exceptional circumstances, the council may, on the advice of the Medical Officer of Health of the council or Chief Health Inspector of tbe council, permit a pit latrine to be constructed within 30 ft. from a habitable room.
9. Every dwelling house shall consist of at least one habitable room in addition to a kitchen, ablution and privy accommodation for the exclusive use of the occupants of the house.
10. (1) Where a ceiling is provided, the average height of a habitable room shall be not less than 7 ft. 9 in. with a minimum height of 7 ft. Where a ceiling is not provided the average height measured to the underside of the roof covering shall be not less than 3 ft. 3 in. with a minimum height of 7 ft.
(2) Every habitable room shall have a superficial area of not less than 75 sq f t . with a minimum width of 6 ft. 6 in. and shall contain a minimum area of 40 sq. ft., for each person accommodated therein:
Provided that m every dwelling of two or more habitable rooms there shall be constructed one habitable room having a superficial floor area of at least 120 sq. ft.
11. The area of the kitchen shall not be less than 25 sq. ft. and not less than 7 ft. in height at any point from the floor to the underside of the roof or ceiling and shall have a satisfactory outlet for smoke and fumes and be lighted and ventilated in accordance with by-laws 13 and 14 of these By-laws.
Latrine.
Dwelling bouse.
Habitable rooms.
Kitchen
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Bathroom
tv mdowi
Ventilation
Surface water drainage.
Fencing.
Relaxation 01 by-law«.
Found* lions.
Walls
Floor*
12. The bathroom shall be at least 2 ft. 6 in. by 4 ft 6 in and if roofed, shall be provided with lighting and ventilatioo in accordance with by-laws 13 and 14 ot these By-laws. The minimum height of any such bathroom, from the floor to the underside of the roof or ceiling shall not be less than 6 ft. 6 in. and adequate provision shall be made for tbe disposal of all waste water by means of a trapped and properly covered soak pit or other method approved by the council.
13. Every habitable room, kitchen, roofed bathroom and latrine shall be provided with a sufficient number of windows opening to the external air so as to provide a clear lighting area equal to at least one-tenth of the floor area of such room, and of which at least one-twentieth of the floor area shall be capable of being opened.
14. Every habitable room, kitchen, roofed bathroom and latrine shall be provided with permanent through or cross ventilation by means of openings which shall give direct access to the external air and tb aggregate area of any such openings shall be equal to at leas: one-hundredth of the floor area of any such room.
15. Surface water drainage shall be provided to tbe satisfaction ot the council.
16. If so required, by the council, the owner of the plot shall cause the plot to be fenced in such manner and by the use of such material as may be required by the council.
17. Where any building is required for a temporary ,-?riod not exceeding six months, tbe council may at its discretic permi: a relaxation in respect of tbe by-laws relating to latrines, ablutions and kitchens.
Part I I — M ethod o f Construction
18. Foundations shall be adequate to support the load transmitted to them and be generally to the satisfaction of the council.
19. No walls shall be constructed to a lower specification than wattle or similar timber adequately framed together and filled and covered with mud. Such walls shall be capable of supporting the roof The covering shall be of adequate thickness and the surface internally and externally shall be sealed and brought to a smooth finish a materials approved by the council and decorated and maintained in • sound and good condition and be redecorated from lime to time as required by the couocil:
Provided that the council may specify the materials to be used in constructing and finishing tbe walls.
20. Every floor shall have a smooth finish and shall be at 'east 6 in. above the surrounding ground level. A floor shall be con structed of concrete, compacted earth or such other materials i: approved by the council.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
21. Every roof shall be of corrugated iron, aluminium, asbestos or jibe* permanent materials or shingles as may be required by council and shall be supported on an adequate frame of poles, timber or similar material. .Any material used shall be in good coo did on and the roof shall be so constructed as to be weatherproof and regular in shape and the pitch of the roof shall conform with the council's requirements:
Provided that— (il where roofs ate to be constructed of corrugated iron or alumi
nium the council may require that ceilings be provided and that corrugated iron roofs be painted and maintained from time to time: and
in) tbe council on tbe advice of the Town Planning Advisor mav set aside either the whole or part ot a Grade II building area where roo s may be permitted to be constructed of grass or other similar material.
22. Frames of doors and shutters shall be constructed in such a way as to be rigid and shall be firmly fixed in the walls.
23. Bathrooms and latrines and each habitable room shall be provided with doors or shall be screened in a manner approved by the councl Such doors shall be at least 2 ft. 3 m. wide and 6 ft. 6 in. high.
24. No person shall construct a well in connexion with any build ing. except with the approval of tbe Medical Officer of Health of the council or the Chief Health Inspector of the council.
Part I I I — Miscellaneous
25. Every person proposing to erect any building in an area to which these By-laws apply shall lodge with the council an application on a form obtainable from the council and three copies of the plan of the proposed building showing its siting and tbe layout of the site together with a front and back elevation of the building and sectioo of the building from the foundations to the uppermost part of the structure to illustrate the construction thereof, with all drawings Jelineated <n a clear and intelligible mar cr and signed by the appli cant or his July authorized agent. The plans shall specify the proposed use of each room and give details of the method of c'istruction and materials to be used:
Provided that where an approved council “type" plan is used the provisions of this bv-law shall be met if the owner signs and deposits three copies of such plan together with the application form duly completed with the council.
26 Applications made under by law 25 of these By-laws shall be accompanied by a fee of Sh 40 for a single dwelling or alterations thereto and an additional fee of Sb. 20 for each additional dwelling included in the application.
Root
Frames ot doors and shutters
Bathrooms and latrine doors
Weli
Applicauons (or erection ot buildings.
Fees
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
300
D e c is io n on a p p lic a tio n s .
Approval by the Council to be void m certain circumstance*.
Erection of buildings »ilhout approval prohibited.
Notice of intention to commence building
27. The council shall approve or disapprove the plan for the erection of a building and it shall signify notice of its < cciMon thereon as soon as practicable after receipt thereof. Such notification shall be given within a maximum period of two months of the receipt of an application in accordauce with these By-laws.
28. The approval of the council of any plans for tbe erection of a building shall be null and void if—
(a) the erection has not been commenced within three months after the date of such approval; or
(b) erection has been commenced but the building has not been completed within a period o;' twelve months from the date of approval, unless the council has agreed to grant an extension of time.
29. (1) No person shall— (a) commence to erect a building without plans thereof having been
approved by the council, or in respect of which the approval of plans has become null and void; or
(b) having obtained the council's appproval to tbe plans for the erection of a building, erect such building otherwise than in accordance with the approved application and plans thereof.
(21 Without prejudice to the liability of any person under para graph (1) of this by-law, the council may serve upon such person or upon the owner of the land upon which tbe building is erected, a notice under the hand of the cleric requiring him within a period of time specified in such notice to do all or any of tbe following things-
(a) to cease the erection of such building; ( b) t o erect such building strictly in accordance with the approved
plans; (c) to execute such work or alterations or additions to such build
ing as may be prescribed in such notice in order to rende; such building safe and sanitary or otherwise conform with the requirements of these By-laws;
(d) to remove or demolish such buildiag. (3) If any person, on whom a notice has been served as aforesaid,
fails to comply with all or any of the requirements of such notice, then the council may, after not less than 48 hours aotice m writing given under the hand of the clerk served upon such person, enter the pre mises and execute such alterations or additions to such bu Iding or remove or demolish tbe building without liability for any loss or damage which may be occasioned thereby and may recover the cost thereof from such person as a civil debt.
30. After plans have been approved by the council the applicant shall give notice to the council in writing of his intention to :om- mence building and shall not commence buildiag until 'be site of the building has beea marked out by the owner and approved by the council.
31. No person shall occupy or permit the occupation of any building to which these By-laws apply until he has obtained from the council a permit in writing authorizing occupation of such premises. Such permit shall not be issued unless the council is satisfied that the building has been erected in accordance with tbe application and approved plans thereof and that the construction is to a standard not lower than is required by these By-laws.
32. ( I ) Any person who contravenes or fails to comply with any of the provisions of these By-laws shall be guilty of an offence and liable—
(a) to a fine not exceeding two thousand shillings <n tespect of a first offence and not exceeding three thousand shillings in respect of a second or subsequent offence, or imprisonment for a period not exceeding six months in respect of a first offence and not exceeding nine months in respect of a second or subsequent offence, or botn such fines and such periods of imprisonment; and
(b) in addition to tbe above penalty, in tbe case of a continuing breach, to a fine not exceeding twenty shillings for each day or part thereof during which such an offence shall continue:
Provided that the aggregate of any fines imposed in the case ot any one continuing offence shall not exceed two thousand shillings.
12) In addition to any penalty as aforesaid, any expenses incurred by the council in consequence of the breach of any of these By-laws or in the execution of any work directed under these By-laws to be executed by any person and not executed by such person, shall be paid by tbe person committing such breach or failing to execute such work
Madr this 29th day of November 1968.
L. G. SAGINI, M inister fo r Local Government.
CPK i::6— lm Bks.— * g?
Occupation ot building without permit pro hibited-
Penalty.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
APPENDIX II
QUESTIONNAIRE
DANDORA SITE AND SERVICE PROJECT
Plot No. ______ Household Head 1. Sex Male
Female
2. Age _____________
3. Marital Status Single _ Married Divorced Widowed
a) Do you have children? Yes __________No b) Total number of children _________________ c) Number living with you on the plot _________ d) No. of wives _____________ e) Residence of spouse(s) Urban Rural f) No. of relatives ____________
3 0 2
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3 0 3
a) What is your occupation? b) Is it permanent ____________ Temporary c) Where do you work? ____________________ d) How far is it from Dandora? Km e) Means of transportation _______ f) Transportation cost per day
per month
6. What is your income? i) Kshs. 300 or less ii) " 301-500 ____ iii) " 501-750 ___ iv) " 751-1000 V ) " 1001-1500 vi) " 1501-2000 vii) " 2000+ ____
7. a) How long have you lived in Nairobi? b) Where did you first live? __________ c) Type of housing
City Council housing _______________ Government housing __________________ Other employer ______________________ Privately rented Squatter settlement
d) Who did you live?
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3 0 4
List all the places you have lived in since then Year Place Rent Rooms Water Toilet Adults Children Sharing
8. a) Where did you live prior to moving to Dandora?
b) Was it in City Council housing ________ ____ Government housing __________________ Other employer ______________________ Privately rented ____________________ Squatter settlement _________________ Rural areas __________________________
Where you renting the place or did you own it? Rental ________________ Owned________________ If renting how much did you pay? Kshs. _______ Was the rent shared? Y e s ____________ No_______ If so how many people? __________________________ How many people lived there? Adults _______ Children_______________ Total number of rooms _____________
b) c )
d) e)
f)
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3 0 5
g) Was the house built with permanent ______ Temporary ____________materials?
10. a) Did you have the following services? Inside Outside Shared Rent p.m
Water ____________________________________________ Toilet ______________________________________________ Shower ______________________________________________ Kitchen ______________________________________________ Electricity ___________________________________________
b) Was the payment for these included in the monthly rent? Yes _________ No___________
c) Did anybody move to that house after you moved out? Yes _______ No____________
d) Do you know the person who moved to the house after you left? Yes ____________ No ____________
e) Do you know where they came from? Yes No __________
11. a) Why did you move to Dandora?
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3 0 6
b) Are you a tenant? _________ An allottee?
RENTERS 12. a) How many rooms do you rent?
b) Payment per month Kshs. __
13 a) Are you sharing the rented room(s)? Yes___________ No___________________
b) If yes, with how many people? ______ c) How much do other contribute towards rent?
Kshs. ____________ d) What is the total number of people living on the plot?
Adults _____________ Children______________ e) In your opinion what is an acceptable number of
Adults ___________ children per room?
14. a) Do you have electricity? Yes _________ No b) Do you pay for electricity or water separately/
Yes __________ No __________ c) Where do you do your cooking?
In the kitchen _________________ In my room _____________________
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3 0 7
15. a) How would you rate the importance of these housing components?
least important most important
1 2 3 4 5 6 7 Permanent structure______ ________________________________ Electricity_______________________________________________ Water ________________________________ Toilet ________________________________ Shower ________________________________ Kitchen ________________________________
b) Which of the following aspects of housing components are most important to you? Permanent Structure _________________ Shower _________________ Water _________________ Toilet _________________ Electricity _________________ Kitchen _________________ c) Would you prefer an extra room or an extra kitchen?
Room ___________ Kitchen_____________ Why? __________________________________________________
d) would you prefer a permanent structure ____________ or the above services ?
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
OWNERS 16. a) How many rooms do you occupy?
b) Do you share these rooms? Yes ______ No c) If yes, with how many people? ___________ d) How much do they pay per month? Kshs. e) What is the total number of people living on the plot
Adults ____________ Children_______________
17. a) How many rooms are subletting? ___________ b) How much income do you get from subletting?
Kshs.____________ c) What is you monthly repayment to the City Council?
Kshs. ___________ d) In your opinion what is an acceptable number of
Adults_________ Children___________ Per room?
SERVICES
18. a) Do you have electricity on your plot? Yes _________ No __________
b) How much did you pay for connection? Kshs. _______ c) How much do you pay for electricity per month?
Kshs. ________________ d) How much do you pay for water per month?
Kshs.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
19.
3 0 9
e) Where do you do your cooking? In the kitchen ___________ In my room _______________
a) How would you rate the importance of these housing components?
least important most important
1 2 3 4 5 6 7 Permanent Structure______ _________________________________ Electricity________________________________________________ Water _________________________________ Toilet _________________________________ Shower _________________________________ Kitchen ___ ____________
b) Which of the following aspects of housing components are most important to you? Permanent Structure _________________ Shower________________________________ Water _________________ Toilet _________________ Electricity_________ _________________ Kitchen _________________ c) Would you prefer an extra room or an extra kitchen?
Room ____________ Kitchen_____________ Why? __________________________________________________
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3 10
Why?
d) would you prefer a permanent structure ____________ or the above services ______________ ?
METHOD OF CONSTRUCTION
20. a) What is the total number of rooms in your plot?
b) How long did it take you to finish construction?
c) Who did the construction? Self_____ Subcontractor/Fundi _______
d) Did your family help with construction? Yes _____________ No__________________
e) How? Skilled Unskilled
21. a) How many people were employed during construction? Skilled ____________ Unskilled _______________
b) How much did you pay them? Total in Kshs. ________ c) What was the total construction cost? Kshs.
22. a) Were you involved in a building group?
Yes No b) What part did the group play in construction?
Skilled Unskilled
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3 1 1
e) What were the disadvantages of the group?
f) Did these groups help reduce the construction cost? Yes ________________ No ____________________
If yes specify
23. What was the major source of finance for construction? Savings __________________ Loans ____________________ Family and friends ______ Others (specify) ________
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
APPENDIX III
PLOT OWNERS
Socio-economic Characteristics
The plot owners were the intended beneficiaries of the Dandora project. The respondents included a higher percentage of female household heads who comprised 56 percent of all the plot owners. The remaining 44 percent were male household heads. The age of the household heads ranged from 20-90 years (see Table Al). Those benefiting as plot owners were older than most households heads in Nairobi. These had also lived in the City for a long time (see Table A2).
A total of 52 percent were married while the remaining 48 percent were single. A total of 32 percent of those married lived with their spouse on the project while 21 percent left their spouse in the rural areas. Except for 3 percent, all the respondents had children ranging in number from 1-16. A total of 21 percent with children however, did not live with their children on the plot. The number of children on the plot ranged from one to ten.
The respondents worked in both the formal and the informal sectors (see Table A3). Those employed in the informal sector comprised 33 percent while 34 percent were in the formal sector. A total of 62 percent of all those
312 Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3 13
Table A1 Age Distribution of the plot Owners
Age No. from Sample Percent
20-30 5 3 31-40 49 33 41-50 45 31 51-90 49 33 Total 148 100
Source: Field Data
Table A2 Number of Years Lived in Nairobi
No. from General Years Sample Percent
6-10 13 9 11-20 54 38 21-30 47 31 31-4 0 24 16 41-60 9 6
Total 147 100
Source: Field Data
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3 1 1
Table A3 Jobs Held by the Plot Owners
Occupation No. from Sample Percent
Landlords 49 33 Self-employed 44 29 Office Support Staff 40 27 Driver/Conductor 8 5 Electrical Mechanic 6 4 Building Trades 3 2 Total 150 100
Source: Field Data
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3 15
employed had permanent jobs while 15 percent had temporary jobs. Most of the plot owners earned higher incomes than that intended for this project. Thus, 16 percent met the income criteria and earned Kshs. 1,000 or less. A total of 28 percent earned Kshs. 1,000-2,000 while 56 percent earned more than Kshs. 2,000. This shows very high incomes for the plot owners. The added income from subletting had made the incomes of the plot owners higher than those of the tenants.
Plot owners Housing Conditions Before and After the Move to Dandora
The plot owners lived in different locations in the City before the move to Dandora (see Table A4) . The highest percentage of the plot owners came from Mathare Valley to Dandora. This was followed by those who came from Kariobangi. These settlements are next to each other end have a large portion of Nairobi's squatter population.
The plot owners lived in both public and private housing before the move to Dandora. Thus, 16 percent came from housing owned by the Nairobi City Council, 3 percent came from Government housing, 7 percent came from employer-provided housing, 30 percent came privately rented housing and 41 percent came from squatter housing. Majority of the plot owner lived in rented housing. Thus, 75 percent lived in rented accommodation while 12 percent owned the housing units. A total of 5 percent were housed by their employers while 8 percent lived with other people.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3 10
Table A4 Location of Housing Estates the Plot Owners Lived
in Before Moving to Dandora
Location No. from Sample Percent
Mathare Valley 31 21 Kariobangi 13 9 Outskirts 9 5 Huruma 7 5 Jericho 7 5 Makadara 7 5 Eastleigh 5 3 Pangani/Ngara 5 3 Bahati 5 3 Kwa Micheal 5 3 Muthurwa/Land Mawe 4 3 Kibera 4 3 Ngomongo/Korokosho 4 3 City Centre 4 3 Other Locations 40 26 Total 150 100
Source: Field Data
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Housing units were constructed with both permanent and temporary materials before the move to Dandora. Thus, 57 percent of the houses were constructed with permanent materials while 43 percent were constructed with temporary materials.
The rent paid for the accommodation ranged from those paying nothing (27 percent) to those paying Kshs. 1200 (see Table A5). The number of rooms ranged from one to six before moving to Dandora and one to four in Dandora (see Table A6). The number of persons occupying these rooms ranged in number from 1-21 before and after the move to Dandora (see Table A7). The services available to plot owners included water, toilet, shower, electricity and kitchen. These services were however not available to all plot owners in their residence prior to moving to Dandora (see Table A8 and A9).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3 18
Table A5 Plot Owners' Housing Expenses Before and After the
Move to Dandora
Rent per Month
Before Dandora In No. from Sample Percent
Dandora No. from Sample
Percent
0 28 19. 4 0 0 100 or less 47 32 . 6 0 0 101-200 30 20.8 2 1 201-300 19 13.2 6 4 301-400 10 7 23 16 401-700 10 7 92 62 701-1250 0 0 26 17 Total 144 100 150 100
Source: Field Data
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3 19
Table A6 Number of Rooms Before and After Moving to Dandora
(Plot Owners)
Number Before Dandora Dandora
of Rooms No. from Sample Percent
No. from Sample Percent
1 98 65 36 24 2 29 19 39 26 3-4 21 14 62 42 5-7 2 2 13 8 Total 150 100 150 100
Source: Field Data
Number Table A7
of Persons per Household Before and the Move Dandora (Plot Owners)
After
Persons
Before Dandora No. from Sample Percent
Dandora No. from Sample Percent
1-2 44 29 30 20 3-6 54 36 54 36 7-10 40 27 44 30 11-21 12 8 22 14 Total 150 100 150 100
Source: Field Data
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3 2 0
Table A8 Available Services to Plot Owners Before
Moving to Dandora
Yes No No. from
Service Sample Percent No. from Sample Percent
Water 144 96 6 4 Toilet 138 92 12 8 Shower 92 61 58 39 Kitchen 48 32 102 68 Electricity 57 38 93 62
Source: Field Data
Table A9 Available Services to Plot Owners in Dandora
Yes No No. from
Service Sample Percent No. from Sample Percent
'vater 150 100 0 0 Toilet 150 100 0 0 Shower 150 100 0 0 Kitchen 53 35 97 65 Electricity 69 46 81 54
Source: Field Data
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3 2 1
TENANTS
Socio-economic characteristics
Most of the tenants were male household heads. The female household heads included a small percentage of included 28 percent of all the tenant household heads. The remaining 72 percent were male household heads. Most of the respondents were young and were 30 years old or younger. Thus, a total of 62 percent were aged 18-30 years (see Table A10) . The respondents had lived in Nairobi for a relatively shorter period when compared to the plot owners. A total of 42 percent had lived in the Nairobi for five years or less while none of the plot owners had lived in the City for less than six years. A total of 27.3 percent had lived in the City for 6-10 years (see Table All).
A total of 59 percent were married while 41 percent were single. Those living with their spouse in Dandora included 35 percent while the remaining 23 percent had left their spouse in the rural areas. Except for 27 percent all the tenants had children. The number of children ranged in number from 1-12. A total of 31 percent of the respondents had left them in the rural areas. This reduced the highest number of children on the plot from 12 to 8 children.
A total of 9 percent were unemployed while the remaining 91 percent worked in both the formal and the informal sectors. Those working in the formal sector included 31 percent while
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table A10 Age Distribution of the Tenants
Age No. from Sample Percent
18-30 91 63 31-40 44 27 41-50 10 7 51-90 P 3 Total 150 100
Source: Field Data
Table All Number of Years Lived in Nairobi
No. from General Years Sample Percent
. 3-10 104 69 11-20 30 20 21-30 10 7 31-40 4 3 41-47 2 1
Total 150 100
Source: Field Data
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3 2 3
those working in the formal sector included 57 percent. A total of 76 percent of all those working had permanent jobs while the remaining 16 percent had temporary jobs (see Table A12 )
The incomes earned by tenants were much lower than those earned by the plot owners. Thus a higher proportion of tenants earned incomes within the target income group for the Dandora project. A total of 41 of the tenants earned Kshs. 1,000, 24 percent earned Kshs. 1,001-1,500, 18 percent earned Kshs. 1,500-2,000 while the remaining 17 percent were in the middle income bracket and earned more than Kshs. 2,000.
Housing conditions Before and After the Move to Dandora
The tenants lived in different parts of the City before moving to Dandora. Some however had moved directly from rural areas or other urban areas (see Table A13). The highest number of the tenants lived in privately rented housing. A total of 40 percent of the tenants lived privately rented housing, 13 percent moved from squatter housing, 9 percent moved from City Council housing while 9 percent moved from Central Government housing. The remaining 22 percent had come directly from rural areas or other urban areas. A total of 41 percent had rented their previous housing while 17 percent owned the housing units. All those who lived in their own houses had mainly come from the rural areas where housing is
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3 2 4
Table A12 Jobs Held by the Tenants
Occupation No. from Sample Percent
Self-employed 40 27 Office Support Staff 33 22 Hotel Services 11 7 Building Trades 15 10 Machine Operator 10 7 Driver/Conductor 9 6 Electrical Mechanic 8 5 Barmaid 7 4 Unemployed 14 10 Others 3 2 Total 150 100
Source: Field Data
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3 2 5
mainly owned by the people living in it. As already said most of this housing is constructed using temporary materials and lacks basic facilities.
A total of 74 percent lived in housing units constructed with permanent material while 26 percent lived in houses constructed with temporary materials. The highest number of those in temporary houses had moved from rural areas where 13 percent of the houses were build with temporary materials. The next highest number of housing build with temporary raterials was in squatter settlement where 11 percent of the housing was constructed with temporal materials.
The rooms ranged from one to seven before the move to Dandora and 1-4 in Dandora (see Table A14) . The number of persons per room ranged in number from 1-17 before the move to Dandora and 1-12 in Dandora (see Table A15) . The rent paid for different housing types ranged from Kshs. 15-2,200 before the move to Dandora and Kshs. 180-1,250 in Dandora (see Table A16) . The services available to the tenants before and after moving to Dandora included water, toilet shower among others (see Table A17 and A18).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3 2 0
Table A13 Location o£ Housing Estates the Tenants Lived
in Before Moving to Dandora
Location No. from Sample Percent
Eastleigh 13 9 Buru Buru 8 5 Kariobangi 8 5 Mathare Valley 7 5 Muthurwa/Land Mawe 6 4 Umoja 5 3 Bahati 4 3 Kibera 4 3 Kangemi 4 3 Pangani/Ngara 4 3 Makongeni 3 2 Kwa Michael 3 2 Westlands 3 2 Embakasi 3 2 Outskirts 3 2 Rural Areas 29 18 Other Urban 5 3 Other Locations 38 26 Total 150 100
Source: Field Data
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table A14 Number of Rooms Before and After Moving to Dandora
(Tenants)
Number Before Dandora Dandora
of Rooms No. from Sample Percent
No. from Sample Percent
1 86 57 126 84 2 23 15 20 13 3-4 34 23 4 3 5-7 4 5 Total 147 100 150 100
Source: Field Data
Table A15 Number of Persons per Household Before and
the Move Dandora (Tenants) After
Before Dandora Dandora
Persons No. from Sample Percent
No. from Sample Percent
1-2 56 38 63 42 3-6 58 39 71 47 7-10 27 18 14 9 11-17 7 5 2 2 Total 148 100 150 100
Source: Field Data
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table A16 Tenants' Housing Expenses Before and After the
Move to Dandora
Rent per Month
Before No. from Sample
Dandora
Percent
In Dandora No. from Percent Sample
0 83 58 0 0 100 or less 11 8 0 0 101-200 14 10 9 6 201-300 10 7 44 29 301-400 10 7 58 39 401-700 14 10 30 20 701-1250 0 0 9 6 Total 142 100 150 100
Source: Field Data
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3 2 9
Table A17 Available Services to Tenants Before Moving to
Moving to Dandora
Yes No No. from
Service Sample Percent No. from Sample Percent
Water 134 89 16 11 Toilet 146 97 4 3 Shower 118 79 32 21 Kitchen 69 46 81 54 Electricity 76 51 74 49
Source: Field Data
Table COH<
Available Services to Tenants in Dandora
Yes No No. from
Service Sample Percent No. from Sample Percent
Water 150 0 0 0 Toilet 150 0 0 0 Shower 150 0 0 0 Kitchen 9 6 94 6 Electricity 67 45 83 55
Source: Field Data
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
BIBLIOGRAPHY Abrams, Charles. 1964. Mans Struggle for Shelter in an
Urbanizing World. Cambridge; Mass. MIT Press. Aggarawal, Ramgopal. 1983. Planning in Developing
Countries: Lessons and Experiences. Washington D.C.: IBRD.
Alonso, William. 1980. "The Population Factor and Urban Structure." The Prospective Citv. Economic. Population. Energy and Environmental Development. Solomon, A., ed. Mass.: MIT Press. 32-51.
. 1964. Location and Land Use: Towards a General Theory of Land Rent. Cambridge: Harvard University Press.
Amis, Philip Henry. 198 3. A Shantv of Tenants. The Commercialization of Unauthorized Housing in Nairobi 1960-1980. Dissertation. Urban and Regional Studies, University of Kent at Canterbury.
1984. "Squat.ers and Tenants: The Commercialization of Unauthorized Housing in Nairobi." World Development. 12. 1: 88-86.
Andreasen, J., J. Kryger., N. Ndung'u. 1987. Case Study of Site and Service Schemes in Kenva: Lessons from Nairobi and Thika. Nairobi: UNCHS (Habitat)
Anderson, T.R., J.A. Egeland. "Spatial Aspects of Social Analysis." American Sociological Review. 26: 392-398
Ast, G. 1979. Space Standard for Urban Low Cost Housing in Kenva. A Proposed Method for Definition of Living Space. Dwelling Size. Plot Size and Plot Use. Heights and Densities. Nairobi: Housing Research and Development Unit. University of Nairobi.
Ball, M.J., R. M. Kirwan. 1977. "Accessibility and Supply Constraints in the Urban Housing Markets," Urban Studies. 14: 11-32.
Bassett, K. , J.R. Short. 1980. Housing and Residential Structure: Alternative Approaches. London: Routledge & K. Paul.
Bloomberg, L.N., C. Abrams. 1965. United Nations Missions to Kenva. A Paper Prepared for the Government of Kenya. Nairobi: Government Printer.
Breese, W. Gerald. 1985. Urbanization in Newly Developed Countries. Englewood Cliffs: N.J.: Prentice-Hall.
3 3 0
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3 3 1
Bourne, L. 1976. "Housing Supply and Housing Market Behaviour in Residential Development." Social Areas in Cities. Spatial Processes and Form. Herbert, D., R.J. Johnston., eds. Vol. 1. London: John Wiley. 111-158. 1982. The Geography of Housing. New York: Wiley.
Bourne, L., J.R. Hitchock. 1978. Urban Housing Markets: Recent Directions in Research and Policy. Toronto: University of Toronto Press.
Bruggeman, W.F. 1974. "Federal Housing Policies Subsidized Housing Filtration and Objectives: A Reply." Land Economics. 50: 317-320.
Brunn, Stanley D.,William F. Jack., eds. 1983. Cities of the World: World Regional Urban Development. New York: Harper & Row.
Burgess, R. 1978. "Poverty Commodity Housing or Dweller Control? A Critique of John Turners's Views on Housing Policy." World Development. 1.9/10: 1105-1133.
Caminos, Horacio., R. Goethert. 1978. Urbanization: Primer for the Design of Sites and Services Projects. Mass.: MIT Press.
Caminos, Horacio., P. Patel. 1973. "A Progressive Development Proposal: Dandora, Nairobi." Ekistics. 36. 214: 205-213.
Chadwick, Bruce A. 1984. Social Science Research Methods. Englewood Cliffs NJ: Prentice-Hall.
Chana, T.S. 1984. "Nairobi: Dandora and Other Projects." Low Income Housing in the Developing World The Role of Site and Services and Squatter Upgrading. Payne, G., ed. Chichester; New York: Wiley. 1982. Background Paper: Human Settlements and Housing
in Eastern Africa. Kenyan and Tanzanian Experiences. Nairobi: United Nations Follow-up Workshop on Housing.
. 1979. Urban Design and Standards in Site and Service Projects. Nairobi: Ministry of Housing and Social Services.
. 1980. Urban Housing Projects in Kenva: An Overview and .Assessment with Special Reference to site and Service Projects. Nairobi: Housing Research and Development Unit. University of Nairobi.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3 3 :i
Chana, T.S., H. Morrison. 1973. "Housing System in the Low-income Sector of Nairobi, Kenya." Ekistics. 34. 34: 214-222.
Chana, T.S., et al. 1979. Evaluation of Site and Service Programs in Kenva. Nairobi: Housing Research and Development Unit. University of Nairobi.
Clark, W.A.V. 1965. "Markov Chain Analysis in Geography: An Application to the Movement of Rental Housing Area." Ann. Assoc. Amer. Geoa. 55: 351-359.
Clark, W. A., P.L. Hosking. 1986. Statistical Methods for Geographers. New York: John Wiley.
Coopers and Lybrand Associates Limited. 1975. Study of the Council1s Housing Operations. Progress Report. Nairobi: Nairobi City Council.
Davies, G. 1976. "Theoretical Approaches to Filtering in the Urban Housing Market." Urban Housing Markets. Bourne,L.S., J. Hitchock., eds. Toronto: University of Toronto Press.
Dennis, M . , Susan Fish. 1972. Programs in Search of a Policy: Low Income Housing in Canada. Toronto: Hakkert.
Dixon, C.J. 1977. Questionnaire and Interviews in Geographical Research. Norwich Geo. Abstracts.
Down, Antony. 1971. Urban Problems and Prospects. Chicago: Markham Pub, Co.
Drakakis-Smith, D.W. 1981. Urbanization Housing and the Development Process. London: Croom Helm. 1987. Third World Citv. London: Methuen.
Duncan, S.S. 1977. "The Housing Question and the Structure of the Housing Market." Journal of Social Policy. 6. 4: 385-412.
Dwyer, D. J. 1975. People and Housing in Third World Cities: Perspectives on the Problems of Spontaneous Settlements. London: Longman.
Ebdon, D. 1985. Statistics in Geography. New York: Blackwell.
El-Shakhs, Sabh. 1974. Urbanization. National Development anti Regional Planning in Africa. New York: Praeger Publishers.
Reproduced with permission of the copyright owner. Further reproduction prohibited w ithout permission.
Fisher, E.M., L.B. Winnick. 1951. "A Reformulation of the Filtering Concept." Journal of Social Issues. 7: 47- 63.
Galtung, Johan. 1967. Theory and Methods of Social Research. Oslo Universtitets for Laget; New York: Columbia University Press.
Gilbert, A., P. Ward. 1985. Housing, the State and the Poor. Cambridge: Cambridge University Press.
Gilbert, A., J. Gugler. 1982. Cities. Poverty and Development: Urbanization in Third World. Oxford: Oxford University Press.
Grebler, Leo. 1952. Housing Market Behaviour in a Declining Area: Long-term Changes in Inventory and Utilization of Housing on New York’s Lower Side. New York: Columbia University Press.
Grigsby, W.A. 1963. Housing Markets and Public Policy Philadelphia: University of Pennsylvania Press.
Grimes, Orville F. 1976. Housing for Low-income Families Economics and Policy in the Developing World. Baltimore: Published for World Bank (by) John Hopkins University Press.
Grosof, Miriam Schapiro. 1985. A Research Primer for Social and Behaviourial Sciences. Toronto: Academic Press.
Hake, Andrew. 1977. African Metropolis: Nairobi's Self- help City. London: Sussex University Press.
Hance, William Adams. 1970. Population. Migration and Urbanization in Africa. New York: Columbia University Press.
Hanna, William John. 1981. Urban Dynamics in Black Africa Hawthorne; New York: Aldine Publ. Co.
Hanushek, E.A., J.M. Quigley. 1978. "The Dynamics of Housing Market: A Stock Adjustment Model of Housing Construction." Journal of Urban Economics. 5: 411-429
Hardoy, Jorge, E., David Satterthwaite. 1981. Shelter Needs and Response; Housing. Land and Settlement Politics in Seventeen Third World Nations. New York: John Wiley and Sons.
Harioe, M., ed. 1977. Captive Cities: Studies in the Political Economy of cities and Regions. London: Wiley.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3 3 1
Harris, J. 1969. "Some Thoughts on Housing Policy for Nairobi." Discussion Paper 78. Nairobi: Institute for Development studies. University of Nairobi.
Harris, Nigel. 1989. "Aid and Urbanization: An Overview." Cities: The International Quarterly on Urban Policy. Butterworths. 6.3: 174-185.
Herbert, J.D. 1979. Urban Development in the Third World: Policy Guidelines . New York: Praeger.
Herbert, J.D., R.J. Johnston., eds. 1976. Social Areas in Cities, spatial Processes and Form. Vol. l. London: John Wiley.
Hoyt, H. 1939. The Structure and Growth of Residential Neighbourhoods in American Cities. Washington D.C.: U.S. GPO.
Housing Finance Company of Kenya. Annual Reports and statement of Accounts 1970-86. Nairobi: Housing Finance Company of Kenya.
Hutton, J. 1969. Urban Challenge in East Africa. Nairobi: East Africa Publishing House.
Houlberg, P. and N. Jorgenson, et al. 1977. Site and Service: An Analysis and Report. Nairobi: Housing Research and Development Unit. University of Nairobi.
Jorgenson, N. 1970. The Urban Housing Market in Kenva. Nairobi: Housing Research and Development Unit. University of Nairobi.
Kabagambe, D. and C. Moughtin. 1983. "Housing the Poor: A Case Study of Nairobi." Third World Planning Review. 5. 3: 227-243.
Kahimbaara, J.A. 1986. "The Population Density Gradient and the Spatial Structure of a Third World City." Urban Studies. 24. 4: 307-322.
Kayongo-Male, D. 1988. "Slum and Squatter Development in Kenya: Housing Problems and Planning Possibilities." Slum and Squatter Settlement in Sub-Saharan Africa: Towards a Planning Strategy. Obudho R.A., and C. Mhlanga., eds. New York: Preager. 133-144.
Keare, Douglas H., P. Scott. 1982. Evaluation of Shelter Programs for the Urban Poor. Washington D.C.: The World Bank.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3 3 5
Kesvani, C. 1979. Nairobi Citv Council Financial Operations and Procedures in Relation to the First and Second Urban World Bank Projects. Seminar for World Bank Implementation in Kenya. Nairobi: Nairobi City Council.
Kiamba, M. 1987. "Private Developers on the Housing Scene in Kenya." A Paper Presented at the International Workshop on Housing. Nairobi.
Kingoriah, G.K. 1980. Policy Impacts on Land Use Patterns in Nairobi. Kenva: 1899-1979. Dissertation. Indiana State University, Terre Haute, U.S.A. 1983. "The Causes of Nairobi's City Structure."
Ekistics. 50. 301: 246-254. Kirby, A.M. 1975. "Housing Market Studies: A Critical
Review." Transaction Institute of British Geographers. 1: 2-7.
Kristof, F.S. 1965. "Housing Policy Goals and the Turnover of Housing." Journal of American Institute of Planners. 31: 232-245. 1973. " Federal Housing Policies Subsidized
Production, Filtration and Objectives,". Part 1. Land Economics. 48: 309-320. , Part II. Land Economics. 49: 163-174.
1974. "Federal Housing, Subsidized Housing, Filtration and Objectives Rejoinder." Land Economics. 50: 320-322.
Lambert, J.P., B. Blackby. 1978. Housing Policy and the State Allocation and Control. London: Macmi11an.
Lansing, J.B. 1969. New House and the Poor: A Study of Chains of Moves. Ann Arbor: University of Michigan, Inst. Soc. Res.
Laquian, A.A. 1977. "Whither Site and Services." Habitat 2: 291-301.
Levanon, z . 1981. Dandora Demonstration Project. Nairobi: UNCHS (Habitat).
Loeckx, A. 1984. The Case of Nairobi. Kenva. Learning from Nairobi, the citv as an Housing Project. Leuven: Post Graduate Centre for Human Settlements. KU-Leuven.
Li, Peter S. 1981. Social Research Methods. Toronto: Brtterworths.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
33<J
Linn, J.F. 1983. Cities in the Developing World: Policies for their Equitable and Efficient Growth. Toronto: Published for the World Bank (by) Oxford University Press.
Lowry, Ira. 1960. "Filtering and Housing Standard: A Conceptual Analysis." Land Economics. 36: 362-370.
Mabogunje, A.L. 1977. Shelter Provision in Developing Countries: The influence of Standards. Chichester; New York: Wiley.
MacJrtnes, G. 1987. Nairobi. Two Cities. Two Economies: An Ecological Analysis of a Site and Service Project. Nairobi: Dandora Catholic Parish.
Macharia, K. 1985. "Housing Policy in Kenya: The View from the Bottom." International Journal of Urban and Regional Research. 9-3: 405-419. Edward Arnold.
Maher, C.A. 1974. "Spatial Patterns in Urban Housing Markets: Filtering in Toronto 1953-1971." Canadian Geographer. 18: 108-124.
Malpezzi, Stephen, S. Mayo, David J. Gross. 1985. Housing Demand in Developing Countries. World Bank Staff Working Papers. No. 733. Washington D.C.:IBRD.
Mangin, w., ed. 1970. Peasants in Cities: Readings in Anthropology of Urbanization. Boston: Houghton Mifflin.
Matthews, J.A. 1981. Quantitative and Statistical Approaches to Geography; A practical Manual. Oxford: Pergamon Press.
Mayo, Stephen., S. Malpezzi., D. Gross. 1986. Shelter Strategies in Developing Countries. World Bank Research Observer. 1.2. Washington D.C.: IBRD.
Mazingira Institute. 1983. Formal and Informal Financing in Site and Service Project in Kenva: A Case Study. Nairobi: UNCHS (Habitat).
. 1982. Community Participation in Low income Urban Settlements in Africa. A Case Study Prepared for UNCHS (Habitat). Nairobi: UNCHS (Habitat)
Mbithi, P., C. Barnes. 1975. Spontaneous Settlement Problems in Kenva. Nairobi: East African Literature Bureau.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3 3 7
Mbogua, J. 1974. Dandora Community Development Project." Housing-summary Project Submission to IBRD. Nairobi: Nairobi City Council.
McAuslan, P. 1985. Urban Land and Shelter for the Poor. London: In-ernational Institute for Environment and Development.
McCallum, D., S. Benjamin. 1985. "Low-income Urban Housing in the Third World: Broadening the Economic Perspective." Urban Studies. 22.4: 277-288.
McClain, J., ed. 1981. "Is Government Home-Ownership Assistance the Way to Go?" Proceedings from a Symposium Held in Calgary, Sept. 18-20. Ottawa: Canadian Council of Social Development.
McGee, T.A. 1976. Urbanization Process in Third World: Explorations in Search of a Theory. London: Bell.
Michelson, W.M. 1977. Environmental Choice. Human Behaviour, and Residential Satisfaction. Toronto: Oxford University Press.
Miller, D.C. 1983. Handbook of Research Design and Social Measurejnentj_ A _Text and_Reference for Social Behaviourial Sciences. New York: Longman.
Morgan, W.T.W., ed. 1967. Nairobi: Citv and Region. Nairobi: Oxford University Press.
Morrison, H. 1974. "Popular Housing Systems in Mombasa and Nairobi, Kenya." Ekistics. 38. 227: 277-280.
Morrison, M.K.C., P.C.W. Gutkind. 1982. Housing the Urban Poor in Africa. Syracuse, N.Y.: Maxwell School of Citizenship and Public Affairs, Syracuse University.
Murie, a . 1983. Housing Inequality and Deprivation. London: Heinemann Education. 1976. Housing Policy and the Housing System. London:
Allen & Unwin. Murison, H.S., J.P. Lea., eds. 1979. Housing in Third
World Countries: Perspectives on Policy and Practice. London: Macmillan.
Mutiso Menezes International. 1983. Umoia II-Project Identification for Nairobi: Nairobi Citv Commission and USAID. Nairobi: Nairobi City Council.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3 3 8
. 1978. Pandora Community Development Project Phase II. Preliminary Reports. Planning and Infrastructure. Provisions for Commercial Areas and Corner Shops. Nairobi: Nairobi City Council.
Mutuku, M. 1980. Community Participation in the Development Process. Nairobi: UNCHS (Habitat).
Muwonge, J.w. 1982. "Intra-urban Mobility and Low Income Housing The case of Nairobi , Kenya." Housing the Urban Poor in Africa. Morrison, M.K.C., P.C.W. Gutkind., eds. Syracuse; N.Y.: Maxwell School of Citizenship & Public Affairs, Syracuse University.
Mwaura, J. 1986. State of Adequacy end Performance of Public and Private Mass-Housing Projects in Urban Kenva. An Evaluation of Community Architectural Services. Leuven: Post Graduate Centre for Human Settlements. KU Leuven.
Nairobi Urban Study Group. 1972. Nairobi Metropolitan Growth Strategy. 2 Vols. Nairobi City Council.
National Housing Corporation. 1967-87. Annual Reports. Nairobi: National Housing Corporation
Njau, G.J. 1981. Nairobi: Evolution of a Housing Policy for Sites and Services and for Squatter Upgrading Schemes. Nairobi: UNCHS (Habitat). 1981. "Mutual Self-help Building Groups in Dandora
Projects." Problems in Search of Solutions. Blair, T., ed. London: School of Environment. 59-66.
Njihia, S.M. 1982. Problems of Housing Finance for Low- income Groups: A Case Study of the Pandora Site and Service Project in Nairobi. Dissertation. University of Nairobi.
Nourse, H.O. 1973. The Effect of Public Policy on Housing Market. Mass.: Lexington Books.
Obudho, R.A. 1983. Urbanization in Kenva A Bottom-up Approach to Development Planning. Lanham; M.D: University Press America.
. 1987. "Shelter and Services for the Poor in Nairobi, Kenya." A Paper Presented at the Experts Meeting in Shelter Services for the Poor in Metropolitan Region, in Nagoya, Japan 12-16 January, 1987.
Obudho, R.A., and Mhlanga., eds. 1988. Slum and Squatter Settlement in Sub-Saharan Africa: Towards a Planning Strategy. New York: Praeger.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3 3 9
Ohls, J.C. 1975. NPublic Policy Towards Low Income Housing and Filtering in Housing Markets," Journal of Urban Economics. 2. 2: 144-171
Ojany, F.F., R.B. Ogendo. 1973. 1935- Kenva. A Study in Physical and Human Geography. Nairobi: Longman.
Ondiege, Peter O. 1981. Housing Policy with Respect to Affordability:__The Kenyan Case. Dissertation. University of Tsukuba, Japan.
. 1984. An Analysis of Implicit Subsidies in Public Housing in Nairobi: A Hedonic Estimation. Dissertation. University of Tsukuba, Japan.
Oppenheim, A.N. 1966. Questionnaire Design and Attitude Measurement. New York: Basic Books Inc.
Payne, G.K., ed. 1984. Low Income Housing in the Developing World: The Role of Sites and Services and Settlement Upgrading. Chichester; New York: Wiley.
Payne, G.K. 1977. Urban Housing in the Third World. London: Routledge and K.Paul.
Peattie, L. 1982. "Some Second Thoughts on Site and Services." Habitat International. 6 1/2: 131-139. 1979. "Housing Policy in Developing Countries: Two
Puzzles." World Development. 7: 1017-1022. Great Britain: Pergamon Press.
Preston, S.H. 1979. "Urban Growth in Developing Countries A Demographic Reappraisal." Population and Development Review. 5. 2: 195-216.
Qadeer, Mohammad. 1983. Urban Development in the Third World Internal Dynamics of Labour. New York: Praeger.
Ratcliff, R.U. 1945. "Filtering Down and the Elimination of Substandard Housing." Land Economics. 21. 4: 322- 330.
Republic of Kenya. 1968. Building Code. Nairobi: Government Printer.
. 1966. Housing Policy for Kenva. Sessional Paper No. 5 of 1966/67. Nairobi: Government Printer
Republic of Kenya. 1964. Kenya's National Development Plan 1964-1970. Nairobi: Government Printer.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3 4 0
. 1970 Kenya's National Development Plan 1970-1974. Nairobi: Government Printer.
. 1974. Kenya's National Development Plan 1974-1978. Nairobi: Government Printer.
. 1979. Kenya's National Development Plan 1979-1983. Nairobi: Government Printer. Ministry of Lands and Settlements. 1978. Human
Settlements in Kenva: A strategy for Urban and Rural Development. Nairobi: Government Printer.
. Ministry of Finance and Planning. Central Bureau of Statistics. 1963-1987. Economic Survey. Nairobi: Government. Ministry of Works, Housing and Physical Planning.
1985. Report on 1981 and 1982 Rent Survey in Urban Areas of Kenva. Nairobi: Ministry of Housing and Physical Planning.
. . 1986. Kenva Low Income By-law Review. Nairobi: Ministry of Housing and Physical Planning.
. . 1986. Urban Housing Survey 1983 Basic Report. Nairobi: Government Printer.
. . 1987. National Housing Strategy for Kenva 1987- 2000. Nairobi: Ministry of Housing and Physical Planning.
Rihal, B. September 1977. "National Housing Policy and the Role of the Private sector. Build. 20-22. Nairobi.
Rodwin, Lloyd and S. Bishwarpriya. 1986. Shelter Settlement and Development: An Overview. Nairobi: UNCHS (Habitat).
. 1987. Shelter. Settlement and Development. Boston: Allen and Unwin.
Rourk, P. and A. Roscoe. 1984. An Assessment of National Housing Needs and Affordability in Kenva 1983-2000. Washington D.C.: United States Aid for International Development.
Sands, G . , L.L. Bower. 1976. Vacancy Chains in the Local Housing Market. Ithaca; N.Y.: Centre for Urban Development Research, Cornell University.
Schnare, A.B. and R.J. Struyk. 1976. "Segmentation in Urban Housing Market." Journal of Urban Economics. 3: 146-166.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3 4 1
Senga, Ndeti and Associates. 1971-1980. Monitoring and Evaluation Study of Dandora Community Project for the Government of Kenva (Medis 1-8). Nairobi: Nairobi City Council.
Sharpe, C.A. 1978. Vacancy Chains and Housing Market. Research notes 3. Dept, of Geography Memorial University of Newfoundland. St. John's.
Skinner, R.J., M. Rodell. 1983. People and Shelter Problems Qf gelizlrelp Housing, in the Third world. London: Methuen.
Simon, Julian L. 1969. Basic Research Methods in Social Science: The Art of Empirical Investigation. New York: Random House.
soja, Edward w. 1968. Geography of Modernization in Kenva: A spatial Analysis of Social Economic and Political Change. Syracuse; N.Y.: Syracuse University Press.
Soni, P. 1981. "Self-help Planning, Construction and Management in a Site and Service Project in Nairobi, Kenya." Ekistics. 48. 286: 50-64.
Stren, Richard. E. 1978. Housing the Urban Poor in Africa: Policies, Politics and Bureaucracy in Mombasa. Berkeley Institute of International Studies, University of California.
Stren Richard. E. 1983. Housing the Urban Poor in Africa. Berkeley Institute of International Studies, University of California.
Struyk, R.J. A. Marshall, L.J. Ozame. 1978. Housing Policies for the Urban Poor. Washington D.C.: Washington Urban Institute.
Springer, J. 1984. Housing Price Change. Residential Mobility and Filtering Metropolitan Toronto 1970-1979. Dissertation. University of Western Ontario.
Syagga, P.M., J.M. Kiamba. 1985. "Urbanization and Housing in Africa." A Paper Presented at the I.C.W.S. Africa Regional Conference, Zimbabwe. 1987. Housing Policy and Systems in Kenva: Nairobi's
Experience. International Workshop on Housing UNCHS/PGCHS, KU-Lueven. Nairobi.
Temple, F. 1973. Politics. Planning and Housing in Nairobi. Dissertation. MIT
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3 4 :j
Tidswell, William V., S.N. Baker. 1971. Quantitative Methods: An Approach to Socio-economic Geography. London: University Tutorial Press.
Tribe, M.A. 1972. "Some Aspects of Urban Housing Development in Kenya." Developmental Trends in Kenva Edinburg.
Turner, A. 1980. Cities of the Poor: Settlement Planning in Developing Counties. London: Cromwell.
Turner, J.F.C. 1978. "Housing in Three Dimensions: Terms of Reference for the Questions Redefined. World Development. Vol. 6, No 9/10: 1135-1145.
. 1976. Housing bv People: Towards Autonomy in Building Environments. London: Marion Boyars.
. 1968. "Housing Priorities, Settlement Patterns Urban Development in Modernizing Countries." American Institute of Planners Journal. 34: 354-363
United Nations Centre For Human Settlements (Habitat) Survey of Slum and Souatter Settlements. Dublin: Publisr ed for UNCHS (Habitat) by Tycooly International Pub: 1982.
. 1985. The Reformulation of Building Act. Regulations and Codes in African Countries. (Annex 1: Outlines of Discussion Papers for the Reformulation of Building Codes and Regulations in Kenya). Nairobi: UNCHS (Habitat).
*-- . 1987. Global Report on Human settlements. Oxford: Oxford University Press.
. 1987. "Workshop on Housing, Nairobi Jan. 18th-31st, 1987. Report and Proceedings." Nairobi: UNCHS (Habitat).
United States Agency for Intern'tional Development. 1982. Umoia II: Project Definition Report. Nairobi: USAID.
. 1979. Kenva Shelter Sector Study and AIDS Experience. Washington D.C.: USAID.
. 1976. Implications of Providing Single-Familv Versus Multi-Familv Housing in Nairobi. Kenva. Nairobi: USAID.
Van Huyck A. 1987. "New Directions for National Shelter Policies." Shelter Settlement and Development. Rodwin Lloyd., ed. Boston: Allen & Unwin. 339-359.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3 43
Van straten, J.J. 1976. Site and Service Schemes; Guidelines for Administrative Procedure. Nairobi: National Housing Corporation.
Vernez, Georges. 1976. "A Housing Services Policy for Low- income Urban Families in Developing Countries." Ekistics. 41. 242: 8-13.
Ward, P.1982. Self-help Housing; A Critique. London: Mansell.
Wanjohi, I., T.S. Chana. 1977. "Strategies for Housing the Lower-Income Groups in the Dandora Community Project, Nairobi: A Case Study." A Paper Presented at the Housing Research and Development Unit. Seminar on Housing for the Lower-Income Groups. Nairobi.
Watscn, C.J. 1974. "Vacancy Chains, Filtering and the Public Sector." Journal of American Institute of Planners. 40: 346-352.
White, H.C. 1971. "Multipliers, Vacancy Chains, and Filtering in Housing." Journal of American Institute of Planners. 37: 88-94.
World Bank. 1974. Site and Service Projects. World Bank Paper. Washington D.C.: IBRD. 1975. Appraisal of a Site and Service Project. Kenva.
Washington D.C.: IBRD. Yahya, S. 1986. Kenva Low Cost Bv-Iaws Review. Nairobi:
Ministry of Works Housing and Physical Planning.
Reproduced with permission of the copyright owner. Further reproduction prohibited w ithout permission.
VITA
NAME: BIRTH DATE: POST-SECONDARY EDUCATION AND DEGREES:
HONOURS AND AWARDS:
RELATED WORK EXPERIENCE
PUBLICATIONS:
MALOMBE, Joyce Mwende November 26, 1953
University of Nairobi 1975-78 B.A. Hon. University of Nairobi (Urban and Regional 1979-81 M.A. (Planning) Institute for Housing Studies Rotterdam (Advanced Diploma in Housing Planning and Building) - 1983 The University of Western Ontario (Geography) 1985-1990 Ph.D.
Canadian International Development Agency, 1985-1990 Special University Scholarship 1986-1988 Netherlands Government Scholarship, 1983 Germany Government Scholarship (DAAD), 1979-81
Research Fellow University of Nairobi, Housing Research and Development Unit, 1981 to present Teaching Assistant University of Western Ontario, Department of Geography, 1986-87, 1988-89
Planning for Housing in Irrigation Settlement Schemes in Kenya (Ahero and West Kano Schemes) November 1906
3 4 4
Reproduced with permission of the copyright owner. Further reproduction prohibited w ithout permission.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission
2 7 4
site and service orojects is so tar one of the most significant ways of beginning to deal with the acute housing shortage in the cities of the Developing Countries.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
