89
 CHAPTER THREE: RESOURCES FOR SELF-HELP HOUSING IN NAIROBI, 
KENYA AND JOHANNESBURG, SOUTH AFRICA
 3.1. Introduction 
The aim of this chapter is to explore the resources that are used towards self-help 
housing in Nairobi, Kenya and Johannesburg, South Africa. The resources have been 
grouped into four broad categories, namely: land; finance; labour, materials and 
technology; and infrastructure and services. 
With regard to land, patterns of access to land created by colonialism in Kenya and 
apartheid in South Africa remain intact and serve the politico-economic elites in both 
countries. While in Kenya this has been attributed to the interest of the political elite 
and an outdated constitutional order that vests absolute powers over public land to the 
presidency, in South Africa a comprehensive legal framework has failed, in its first 
ten years, to significantly reverse patterns of land distribution entrenched by 40 years 
of apartheid rule. The token achievements of land redistribution in South Africa are 
generally absent from Kenya.
 Next, I explore different ways through which low-income housing is financed in both 
countries. Through exploration of financial mechanisms of the public sector, private 
sector and grassroots initiatives, community initiatives through savings can contribute 
significantly to housing finance. Any low-income housing policy can only succeed 
with state funding, as the private sector interests together with inappropriateness of 
mortgage finance for the low-income limit the efficacy of these. 
Thirdly, I look at labour, materials and technology in Nairobi and Johannesburg. I 
show that self-build by the actual households is not the dominant trend in both cities, 
due to availability of cheap skilled and semi-skilled labour, as a result of high levels 
of joblessness. Self-management is a more dominant approach to self-help housing. 
Even more importantly, my finding is that local construction initiatives in South 
Africa are not working in spite of legislative frameworks like the Preferential 
Procurement Act. This is an area where Nairobi offers good lessons, although this 
90
 happened by default, when the state and big construction companies withdrew from 
low-income housing. Even though alternative materials and technology are allowed in 
both cities, their uptake is low, because of the conflicting interests of the state and the 
market. 
Lastly, I discuss access to infrastructure and services. The central concern is the 
privatisation approaches that are favoured by the cities of Johannesburg and Nairobi. 
These do not only subvert the intentions towards equity in service provision, but they 
are not necessarily more efficient than public sector approaches. 
3.2. Access to land by low-income groups in Kenya and South Africa
 The first resource towards housing in Nairobi and Johannesburg that I discuss is land. 
I will take the position of various commentators that colonial segregation with regard 
to land-use in Nairobi has been perpetuated 40 years after independence through the 
politico-economic elite?s interests. In South Africa the race-based access to land 
entrenched during apartheid, has been sustained through class-based access to land 
despite land redistribution and subsidised housing programmes. There are very 
limited means through which the poor can access any land, let alone well-located 
land. In Nairobi, the cooperative model and Community Land Trusts have been tried 
as alternatives for the poor to access land. However, the scale of delivery through 
these is negligible. The cooperative model works in favour of middle-income groups, 
rather than the poor. In Johannesburg, the state?s housing subsidy programme is the 
dominant way of accessing land. But this has entrenched the status quo, as the most 
poorly located land is often the easiest to access on a willing seller willing buyer 
basis. Creative attempts by the local authority to deal with apartheid geography, e.g. 
the Greater Johannesburg?s Rapid Land Release Programme in 1999/2000, failed 
because of unresolved tensions within the programme and structural issues around 
access to land in post-apartheid South Africa that have not been addressed, e.g. the 
issue of litigations. Therefore, informal access to land and invasions remain the two 
main means through which the poor access land in both cities. However, the successes 
of the initiatives are limited. In Nairobi a conservative legal framework and an 
outdated constitutional order defeat these invasions. In Johannesburg, conservative 
91
 interpretation of a transformative constitution and legislative instruments achieves 
similar results. The consequence is that in Nairobi access to land is dominated by 
clientelist relationships and political abuse. In South Africa, increase in informal 
access to land and slow titling processes - where the poor are able to access land 
formally, demonstrate the incapacity of the formal system to meet the demand for 
formal access to land by low-income groups. Another consequence in both contexts is 
the entrenchment of the historical inequality in access to urban land, in favour of 
upper income groups and the growth of urban informality.
 In this section I expound on three broad themes, namely: failure of current 
mechanisms of access to land to redress urban segregation; the limited options of 
access to land by the poor, e.g. through government?s low-income housing 
programmes and land invasions; the consequences of these, e.g. entrenchment of 
informality and illegality, and increase in insecurity of tenure.
 3.2.1. Problems with legal and legislative frameworks for access to land in 
Nairobi and Johannesburg
 There is recognition, both in Kenya and South Africa that the means through which 
land is accessed are inadequate. The response in South Africa has been to develop a 
comprehensive legislative framework that attempts to incorporate all the conflicting 
interests, without resolving them adequately. In Kenya, land reform has been going on 
for the last few years in response to the land problem, which has been interpreted, 
narrowly, to mean land regularisation, with little space for engagement with broader 
issues like equity in access to land.
 South Africa developed very detailed legislation around the issue of access to land 
during the transition to the post-apartheid policy environment. This included: 
Abolition of Racially Based Land Measures Act 108 of 1991, which repealed Group 
Areas Act of 1966, one of the apartheid?s instruments for restricting black access to 
urban land. The other legislation was the Distribution and Transfer of Certain State 
Land Act, 199 of 1993 that would enable the state to use some of its land to redress 
past inequalities. The Development Facilitation Act (DFA) 67 of 1995 and the Less 
Formal Township Establishment (LEFTE) Act 113 of 1991 were aimed at fast 
92
 tracking development in the ?black? areas of the previously ?white cities?.  Restitution 
of Land Rights Act, 22 of 1994 was intended to help those disposed of ancestral land 
in the apartheid regime to regain rights to these. Upgrading of Tenure Rights Act 112 
of 1991 was to deal with those who had been occupying land informally, but had not 
acquired legal rights. The Prevention of Illegal Eviction From and Unlawful 
Occupation of Land (PIE) Act 19 of 1998, was to repeal the apartheid?s Illegal 
Squatting Act, through recognition of tenure rights of persons who had been 
occupying land for more than six months, while simultaneously recognising the 
ownership rights of the landlord. Through this legislation a court order is required to 
evict the squatter. 
In spite of this legislation, urban land problems in South Africa have remained almost 
intact (Berrisford, 1999). There is very little redistribution of urban land. Rural related 
legislation, e.g. the Extension of Security of Tenure Act (ESTA) 62 of 1997 and 
Interim Protection of Informal Land Rights Act 31 of 1996 have been relatively more 
effective in securing tenure for farm workers. The white landowners are being joined 
by black elites who are more concerned about property values, than about social 
justice (Berrisford, 1999). 
Corruption and inadequate land legislation are the two main issues with regard to 
access to land in Nairobi. Corruption in land transfers and ownership, and land 
grabbing by upper income groups, enabled through political linkages and patronage, 
are widely reported in academic writing and public documents (see for example 
Majale, 2002; Syagga, Mitullah & Karirah-Gitau, 2001; Konyimbi, 2001; Republic of 
Kenya, 2002b). The legal framework for management of land in Kenya is not merely 
complicated but almost ?unworkable? (see Konyimbi, 2001). The draconian legislative 
framework inherited from the colonial government, merely transferred power over 
land from the queen of England to the President of Kenya, without much 
transformation in the way land was to be accessed by citizens. This is reflected in the 
Government Lands Act Cap 280 and explained by Syagga, Mitullah & Karirah-Gitau 
(2001)
 ?The president?assumed all authority to ?make grants or dispositions of any 
estates, interests or rights in or over unalienated government land?. Neither the 
93
 constitution nor the Government Lands Act imposes limitations on the 
president on his powers to allocate public land??
 In exercise of this power, the former president of Kenya, Daniel Arap Moi, instituted 
a commission to review the legal framework for management of land. In 1999 the 
Commission of Inquiry into the Land Law System in Kenya ? ?the Njonjo 
Commission? ? produced a report titled: ?Report of the Commission of Inquiry into the 
Land Law System of Kenya on Principles of a National Land Policy Framework, 
Constitutional Position of Land and New Institutional Framework for Land 
Administration? (Republic of Kenya, 2002b). According to the commission (Republic 
of Kenya, 2002b: 32) the following are some of the current challenges around land in 
addition to those caused as a result of colonisation: rapid population growth; impacts 
of HIV/Aids; deterioration of production of land; breakdown in land administrative 
system; rapid urbanisation; uncontrolled development; desertification; poverty; 
gender; disparities in access to land and involvement of unauthorised persons in land 
matters. The commission failed to deal with the fact that government land was vested 
in the person of the president who had hitherto abused this privilege and caused some 
of the current problems. Corrupt allocation of land, on the other hand, became the 
subject of another presidential enquiry in 2003: the Presidential Commission of 
Inquiry into Illegal Allocation of Land (the Ndungu Commission). A list of land 
grabbers was produced by this report, but the government is yet to act on it. 
Meanwhile, the Land Reform Process is progressing, and the extent to which it will 
deal with some of these ills remains to be seen, especially when land management is
 its focus.
 The inadequacy of engagement with the land question both in Kenya and South 
Africa has resulted in weakness in urban reforms, perpetuation of urban segregation, 
and limitation of access to urban land by the poor, which I discuss next.
94
 3.2.2. Failure to redress racial and economic class-based land access 
Colonial and apartheid racial segregation, in Kenya and South Africa respectively, set 
the basis for post-colonial and post-apartheid economic class-based access to land. 
The racial and class patterns remain intact in Nairobi and Johannesburg with little 
effect of new policies to redress them.
 Spatial segregation dominates land-use patterns in both cities. In South Africa there 
are clear intentions to redress: 
??ineffective and inequitable cities: the geographic segmentation of living 
areas according to race and class, urban sprawl, and disparate levels of service 
provision and access to amenities in different areas [which] make South 
Africa?s cities very inequitable, inefficient and relatively expensive to manage 
and maintain? (Department of Housing, 1994:12, in Huchzermeyer, 2001b: 4).
 In Nairobi, engagement with segregation is not a central issue. In both cities, the 
market controls land redistribution in favour of the upper-income. This has led to a 
situation where geographical locations of new settlements for the poor in both cities 
are relatively predictable, i.e. in the outskirts of the cities, with limited access to 
infrastructure, services and urban opportunities. This is enforced through a strong 
Not-In-My-Backyard (NIMBY) factor. An example of NIMBY in Nairobi was the 
removal of the Soweto slums from the upmarket Hill View/Loresho area in Nairobi 
early in 19941. In Johannesburg the numerous litigations against the City of 
Johannesburg?s Rapid Land Development Programme (RLDP) exemplified this 
NIMBY problem (Bremner, 2000: 98). The RLDP was intended to identify and 
develop well-located land for the low-income, rapidly, through the provisions of 
Development Facilitation Act (DFA) and Establishment of Less Formal Township 
Act (LEFTE) (Bremner, 2000). However, there was opposition from the host 
communities, forcing the City of Johannesburg to abandon the programme. The 
visions within the government were conflicting; with the Provincial Housing Board 
not sharing in the City of Johannesburg?s commitment to social and urban integration. 
  
1 I was involved in attempts to help the squatters through an initiative that was being driven by the 
Architectural Firm I was working for at the time, Planning Systems Services. Our offices were then in 
the same neighbourhood as the squatter settlement, in Hill View Estate, Nairobi. The businessman in 
question gave us the go-ahead to help upgrade the settlement, only to refuse when the settlement was 
mysteriously gutted down by fire. See also Republic of Kenya (1999) on Soweto relocations.
95
 The former preferred instead to keep funding projects in traditional (peripheral) 
locations (Bremner, 2000: 99). Bremner (2000: 99) talks of the upper-income 
hysteria, fuelled by inaccurate and sensational media reports, racial tension and 
distrust, as some of the factors that caused the programme to fail. Thus in 
Johannesburg, race-based locations are still predominant, while in Nairobi 
segregation, which was initially race-based, has now taken economic class lines 
(Olima, 2003) ? even as the racial substructure remains intact - reinforced by 
capitalist, non-redistributive approaches the post-colonial Kenyan governments have 
adopted. 
There has been no meaningful land reform in either of the two cities (see Berrisford, 
1999 on the South African case; see Olima and Kreibich, 2002; Olima and Rukwaro, 
2003 on the Nairobi case. See Republic of Kenya, 2002b on Land Reforms in Kenya).
 Land reforms remain illusive in Johannesburg in spite of various instruments to deal 
with them, e.g. the Urban Development Framework; the Development Facilitation Act 
(DFA) 1995, requiring Land Development Objectives (LDOs); and the White Paper 
on Local Government, calling for integrated Development Plans (IDPs) 
(Huchzermeyer, 2001b: 34). The good intentions of the government are thwarted by 
the government?s own developer driven housing programme, which tends to avoid 
good locations, but where they may have to overcome the sort of opposition that the 
RLDP encountered (Huchzermeyer, 2001b: 9). A legislative system and a 
constitutional order that recognises the rights of the previously disadvantaged, while 
almost falling short of giving absolute rights to private property (Section 25 of the 
Constitution of South Africa) creates a situation where court decisions on rights of the 
poor to access land often work in favour of the rich (see the Bredell case in 
Huchzermeyer, 2003).
 Lastly, there are limited alternatives for the poor to access land in both cities. The 
South African Department of Housing?s new Housing Plan, Breaking New Ground
 (Department, 2004: 5) recognises that ?identification, acquisition, assembly and 
release of state owned and private land in terms of the revised procurement 
framework has proved to be a slow and complex process?. The situation in Nairobi is 
no better. One way this obstacle has been responded to, especially among the middle 
income groups in Nairobi, is the emergence of land buying companies, which do 
96
 wholesale buying of land and subdivide into plots for their members (see Olima and 
Rukwaro, 2003: 144). Most members are linked horizontally, i.e. they would belong 
to more or less similar economic classes. They tend to be formally employed. This is 
also enabled by a strong culture of savings through cooperatives (Mwaura, 2002). 
However, cooperatives do not work for the lowest income groups, even in Kenya 
where the cooperative movement is more pronounced than in South Africa. In South 
Africa, the role of cooperatives as vehicles for access to urban land and housing is still 
relatively weak (Mthweku and Tomlinson, 2000; Department of Housing, 2004). So 
how then do the poor access land in Nairobi and Johannesburg?
 3.2.3. Land subsidies and invasions
 One way of access to land by the low-income in Johannesburg is through the state?s 
housing subsidy programmes. There are occasional projects based on land subsidies 
for developments on public land in Nairobi, e.g. in the Mathare 4A scheme and in 
Kibera Slum Upgrading Scheme, but this is not mainstream policy. 
The other way through which the poor access land in both cities is through land 
invasions. Invasion of well-located land in Nairobi is rare, but not totally absent. 
Cases in point are the Deep Sea settlements in the upmarket Westlands area and 
Mukuru in the middle-class South B area2. In most cases they occur, in close 
proximity to other low-income settlements or in un-developable areas like riparian 
way leaves e.g. expansion of Kiambiu settlement, near Buruburu and Korogocho in 
Nairobi. Huchzermeyer (1999a; 2001b) also noted this pattern of land invasion in 
South Africa. In Nairobi, invasions are enabled by strong clientelism, especially from 
the politicians. Studies I conducted in Gitare Marigo area, in Dandora, as discussed in 
the Chapter Five, confirmed this fact. Invasions in both cities are rarely successful. 
South Africa has a more detailed legislative framework with regard to invasions, as 
discussed next.
   
2 I took part in a mission to Nairobi in July 2003 as part of a fact-finding mission on forced evictions by 
Centre for Housing Rights and Evictions. During this time I had a chance to visit most of these 
informal settlements, in addition to having been brought up in the city of Nairobi. See the COHRE 
report in http://www.cohrekenya.org.
97
 Huchzermeyer (2003: 62) outlines the legal position with regard to land invasions and 
evictions in South Africa. Most of the legislation finds its legal basis from Section 26 
of the Constitutions of the Republic of South Africa, which states that:
 - (1) Everyone has a right to have access to adequate housing;
 - (2) The state must take reasonable legislative and other measures within its 
available resources, to achieve the progressive realisation of this right;
 - (3) No one may be evicted from their home, or have their home demolished, 
without an order of court made after considering all the relevant 
circumstances. No legislation may permit arbitrary eviction.
 Section One and Two give rise to various pieces of legislation and policies to realise 
the right to housing. It is the basis of the Housing Act number 107 of 1997. Section 
Three has been elaborated through the Prevention of Illegal Occupation of and 
Unlawful Eviction From Land Act of 1998 (PIE Act), which criminalizes evictions. 
Incidentally, the landmark case on evictions, ?Grootboom and others versus the 
Western Cape Municipality? was won by the invaders in the Appeal Court on the 
basis of Section 27(1)(c) that confers absolute right of shelter to children. In that case 
the right to shelter for children was interpreted progressively to include the context of 
the communities they live in. The Constitutional Court subsequently gave a different 
interpretation of the right to housing in the Grootboom case, instead focussing on the 
state?s obligation to use its available resources in a reasonable way to assist those in 
desparate need for shelter (Huchzermeyer, 2003). However, many cases do not find 
their way to the courts, with property owners and government managing to evict on 
all manner of bases, including remnants of apartheid legislation, like the Trespass Act 
of 1959 and health regulations. Additionally, Section 25 of the Constitution protects 
existing property rights, making it difficult to realise sub-section 4, which deals with 
equitable access to land. 
In both cities, successful invasions as a means of access to land by the poor, is 
limited. Application of the law to deal with the rights of the poor with regard to land 
yields mixed results and as Huchzermeyer (2003) observed in South Africa, this is 
marked by an ambiguous interpretation of the law.
98
 3.2.4. Entrenchment of informality and illegality in access to land
 Neither the market nor the state?s legislation can guarantee access to land by the poor 
as I have already discussed. The consequences of this are entrenchment of informality 
and illegality with regard to occupation of land in both cities.
 In Kenya only 6% of the total land area has been registered under individual titles 
(Development Plan, 1997-2001, in Konyimbi, 2001: 50). In Nairobi, the local chiefs 
and councillors allocate public land irregularly, thereby giving temporary, verbal and 
insecure tenure. Beneficiaries of this tenure tend to be linked mainly through local 
politico-economic networks. This system has been massively abused and is 
considered to be corrupt, as captured in these quotations below: 
?The procedures for allocating public land have been subjected to abuse and 
violation resulting in corruption in land matters, speculation and improper 
allocation. This is what has come to be branded as ?Land Grabbing? in Kenya, 
which is actually an abuse of land delivery system, with public land being 
disposed of at prices below the market value to the powerful? (Syagga, 
Mitullah & Karirah-Gitau, 2001: 83; See also Olima, 1997: 327-28). 
?That there has been abuse of trust by the Government and the county 
councils, its officials and councillors in the irregular allocation of public and 
community land without following legally laid down procedures that ensure 
appropriateness, transparency and fairness. The abuse has led to massive 
grabbing of land reserved for public use...? (Republic of Kenya, 2002: 91).
 74% of a sample of households in a study conducted by Olima (1997: 328) in a 
section of Nairobi had lost their land through irregular re-allocation to government 
officials and the then ruling party (KANU) officials. Land was reallocated to 
?prominent and influential personalities? (Olima, 1997: 328). These same people 
ignore development conditions attached to the titles (ibid.). 
Irregularity and informality are also on the increase in South Africa. Some legal 
experts argue that there is already a legal framework in place, which allows for 
99
 various forms of non-paper security of tenure (Roux, 2002), a position that is 
contestable based on the evictions that have been happening in Johannesburg. 
Huchzermeyer (2004b: 6) documents that in few circumstances, like in the case of Joe 
Slovo informal settlement, in Port Elizabeth, the state moved in to help the invaders 
acquire land legally. This is definitely not the mainstream policy, as acquisition of the 
land in this case was only enabled through ties between the leadership of the 
Homeless People?s Federation and the People?s Dialogue with the Minister for Land 
Affairs, Derek Hanekom. Additionally, Huchzermeyer (2001d) argues that 
informality should be interpreted as lack of rights instead on mainstream 
interpretation as illegality.
 3.3. Access to finance for low-income housing in Kenya and South Africa
 Effective modes of financing low-cost housing remain elusive in Kenya and South 
Africa. In South Africa, the key source of housing finance is government funding 
through the housing subsidy programme, though it was expected the private sector 
would participate in partnership with the government and communities (see 
Tomlinson, 1999; Rust, 2002). The South African National Department of Housing is 
still trying to make this public/private partnership work for lowest income groups, 
with little success. In Kenya, there is no government subsidy for low-income housing, 
and even the new housing policy has ignored the issue. There are lessons that could be 
learnt from the South African situation, with strong government participation in low-
 cost housing, and where a plethora of bargained structures have been put in place to 
enable the private sector participation, with limited success3. The role of micro-
 finance and individual investment in low-cost housing is not adequately appreciated 
in both countries, hence their potential remains largely un-explored. However, there 
are isolated cases of success amongst individual savings groups, but these are yet to 
influence policy meaningfully.
   
3 The Financial Services Charter is being discussed between the banks and the government as this 
report is being written, to work out modalities of availing funds to the low-income after failure by the 
banks to lend in this sector from 1994 despite the state?s effort to make them do so.
100
 In this section I discuss four broad issues relating to housing finance in Kenya and 
South Africa. In section 4.3.1, I discuss the need for state subsidies for low-income 
housing, and some lessons that Kenya could learn from South Africa. In section 4.3.2 
I discuss the failure of formal housing finance to deliver funds for the low-income in 
Kenya and South Africa, hence the inappropriateness of any low-income housing 
policy hinged on these. The next section 4.3.3 looks at the inappropriateness of 
mortgage finance for the low-income, given the latter?s fluctuating financial positions 
over time. The section also explores exploitation enabled through semi/unregulated 
micro-finance. Section 4.3.4 puts a case for recognition of the roles already being 
played by savings in both countries and strengthening these especially in light of 
diminishing NGO capacity to support them. 
3.3.1. Funding the housing policy: the role of state subsidies
 In South Africa, formal low-income housing is mainly funded through different 
government initiatives, including an extensive subsidy system. In Kenya, it is 
assumed that the government can enable the private sector finance housing for the 
low-income, in spite of local and international evidence to the contrary.
 Funds for housing subsidies in South Africa emanate from the South African Housing 
Fund through an annual allocation of the national revenue. The current revisions to 
the decade old housing subsidy system are reflected in the Breaking New Ground
 housing plan (Department of Housing 2004: 23, 24 & 25). Households earning below 
R3 500.004 qualify for full state housing subsidy as from the 1st April 2005. Those in 
the bracket of R 3 500.00 ? R 7 000.00 will qualify for credit and savings linked 
subsidy. The details of this are being worked in the ?Financial Services Charter?, 
being discussed between the government and the banks, as this thesis is being written. 
It is assumed that the banks will finally be able to avail credit to the low-income for 
housing through new loan models. Other revisions to the subsidy framework are: 
linking the subsidy to inflation to ensure that developers can maintain their profit 
margins (see Department of Housing, 2004: 23). There are debates whether the 
mandatory beneficiary contributions of R 2 475.00 should be removed to avoid down 
  
4  Exchange rates: 1 US Dollar = 6.3 Rand; = 74.4 Kenya Shillings (13th September 2005)
101
 raiding. However, as observed in studies by Mthweku and Tomlinson (1999), and 
Tomlinson (1999: 292) successful housing policies elsewhere, e.g. Singapore, Chile, 
Germany tend to have strong savings components. 
The current funding mechanisms mean that the state is likely to remain the main 
supplier of wholesale finance and guarantor of loans towards housing (Khan, 2003: 
80). The only other major organisation that availed credit to the poor, in South Africa, 
has been the state founded and funded National Housing Finance Corporation 
(NHFC). It lends wholesale finance for onward lending by smaller banks, micro-
 financiers and NGO lenders. However, because it was conceived as a private sector 
organisation, there has been pressure that it operates as one. For example it has been 
suggested that the NHFC should aim at getting high returns on loans for low risk 
(Rust, 2002: 27). It has been criticised for inadequate capital, ?inappropriate 
independence?, failing to impact rising interest rates and in fact failing to create 
meaningful access to finance for the poor (Bond, 2003: 47). Bond (ibid.) further 
accuses the bank of failing to keep afloat and poor targeting of intermediate lenders, 
e.g. choosing to lend to the collapsed Community Bank.
 There are lessons that Kenya could learn from the South African housing subsidy 
system. Most of the housing stock for the low income that has been produced in South 
Africa has been done through direct state funding. In Kenya, housing finance has been 
left almost totally to the private and semi-private state organisations, a fact that 
renders housing finance out of the reach of the low-income. The exchequer allocation 
of funds for housing, including rural housing loans, has collapsed (Republic of Kenya, 
2003: 19). The Housing Policy (Republic of Kenya, 2003) envisages a Slum and 
Low-Cost Housing Infrastructure Development Fund, but it is not clear whether this 
fund will ever be capitalised. There remains no targeted housing fund by the state for 
low-income housing in Kenya. However, there is still a role for direct state funding 
for housing, even as the state tries to enable other actors also to participate.
 3.3.2. Failure of the private sector finance to reach the low-income
 The major players in formal housing finance in Kenya include Housing Finance 
Company of Kenya (HFCK), East African Building Society (EABS), Savings and 
102
 Loans (S&L) and National Housing Corporation (NHC). The first two give long-term 
mortgage, Savings and Loans gives commercial loans, while the NHC was intended to 
channel government funds into housing (see Syagga, Mitullah & Karirah-Gitau, 2001: 
104). The NHC?s development is based on cost recovery through provision of secured 
loans. This has ensured that only the formally employed middle-income groups access 
the units. The NHC had developed 42,340 units of rental, tenant purchase, site and 
services and mortgage housing nationally by 1990 at a cost of US$ 40 million 
(Rimber, 2002: 15; Syagga, Mitullah and Karirah-Gitau, 2001). Their current efforts 
to get back into housing development has hit a snag, with their bond issue which was 
to be backed by the International Finance Corporation (IFC) having been rejected by 
the Treasury (East African Standard, 25th November 2004). There are also funds from 
Building Societies registered in terms of Building Societies Act. These were 26 
nationally in 1993. They declined to only three in 1997. The remaining ones are East 
African Building Society (EABS), Equity Building Society and Family Finance 
Building Society. They fund a very small segment of the formal market.
 Inaccessibility to housing finance for the low-income groups has been recognised by 
the government (Republic of Kenya, 2003:19). However, there is still conviction that 
the private sector can provide the finance, as reflected in the current housing policy 
framework. It was initially expected that the USA based company Overseas Private 
Investment Corporation (OPIC) would fund 150,000 housing units annually. But this 
did not take off because the government and the company could not reach an 
agreement. The current housing policy (Republic of Kenya, 2003) recommends 
amendments of Retirement Benefits Authority Act, to allow 10% statutory and 
pensions funds held by the National Hospital Insurance Fund (NHIF) and the National 
Social Security Fund (NSSF) to be lent for housing. It further recommends that new 
legislation be made to compel banks to lend 5% of their funds for housing. Currently, 
commercial banks and non-banking financial institutions are restricted from investing 
in housing by the provisions of the Banking Act (Republic of Kenya, 2003: 19). Other 
mechanisms for increasing investment in housing include recommendations for: tax 
relief on interest from home ownership; waiver of stamp duty for first time home 
owners; housing development bonds; employers to facilitate loan acquisition by 
employees; and off-shore borrowing by the government to reduce local interest rates 
103
 on loans. It is also suggested that the state sets up a Mortgage Guarantee Scheme to be 
administered by the National Housing Corporation (Republic of Kenya, 2003: 20/21). 
The central problem with most of these mechanisms for housing finance is that they 
are all indirect, aimed at enabling markets work. They require a lot of legislation, 
which it appears the current government, does not have the capacity, let alone the 
commitment, to enact. Housing is still not a constitutional right in Kenya, due to an 
incomplete constitution making process, making any commitments by the state to 
deliver housing difficult to claim.  The introduction of a right to ?accessible and 
adequate housing? in Chapter Six, Section 63, of the draft constitution (Republic of 
Kenya, 2004b) is still caught up in the drawn out constitution making process. 
Funding mechanisms are clearly in favour of middle and upper income groups, 
particularly those in formal employment. The token offer to the poor is the Slum and 
Low-Cost Housing Infrastructure Development Fund, which is yet to be capitalised. 
According to the housing policy (Republic of Kenya, 2003) finance is expensive, 
because the financiers source the money in the open market or through housing 
development bonds for mortgage, which are subject to taxation. However, the 
solutions the policy envisages are far from being adequate. For example the biggest 
slum upgrading initiative in the history of Kenya, the Kenya National Slum 
Upgrading Programme (KENSUP), is to be funded 80% by donors (Kusienya, 2004). 
Despite the initiative having been started in 2000, there have not been any committed 
funds from government to fund it. The government has since pledged to fund 20% of 
the KENSUP programme (Kusienya, 2004: 6). 
Private sector finance for low-income housing has also been problematic in South 
Africa. The financial sector quit the low-cost housing market in South Africa partly 
because of the bond boycotts in the early 1990s (see Tomlinson, 1999: 288). The 
deadlock that sent them out was that they were unable to re-posses bonded houses 
from defaulters for political reasons. The defaulters on the other hand were unhappy 
with the quality of the houses, hence their refusal to repay. The Record of 
Understanding (RoU) between the government and the banks in 1994 was aimed at 
getting the banks to resume lending end-user finance to the low-income (Tomlinson, 
1997: 6; Tomlinson, 1999: 288). Several mechanisms were put into place to get the 
banks to lend to the low-income for housing. I discuss some of those initiatives next.
104
 The Masakhane (let?s build together) initiative was the state?s response to engage the 
communities to continue paying their bonds. This campaign was largely unsuccessful 
(Tomlinson, 1997: 15). The other initiative was the establishment of the Mortgage 
Indemnity Fund (MIF) to cover the banks in case of bond defaults (Tomlinson, 1997: 
7; Bond, 2000; Jenkins, 1999). The MIF gave coverage in areas the banks considered 
risky to lend to (Tomlinson, 1997: 7; Tomlinson, 1999: 288). By agreeing to give 
coverage in these areas, the state was confirming that they were indeed risky areas. 
Thus the MIF ended up acting indirectly as a redlining tool (Bond 2003: 47). The 
other initiatives were Servcon Housing Solutions and Home Builders? Registration 
Council (NHBRC) (see also Bond, 2000; Jenkins, 1999, Tomlinson, 1997: 7). 
Servcon was to assist borrowers of about 18 000 non-performing loans after banks 
failed to reschedule these. They were to offer defaulters alternatives and enter into 
new agreements with them (Tomlinson, 1997: 8). Their role was undermined by lack 
of appropriate stock to right-size borrowers into (Bond, 2003: 47). The NHBRC on 
the other hand was meant to solve the problem of poor quality of house construction. 
It was to achieve this through registration of contractors and offering indemnity in 
case of shoddy workmanship. This organisation has succeeded more in inhibiting self-
 build processes in the cities, which are exempted from registration, but are assumed to 
be of poor quality. On the other hand it has been accused on focusing on collection of 
monies from contractors without protecting the consumer adequately5. There was also 
potential use of the Home Loan Guarantee Company (HLGC) established in 1990 to 
cover short-term loans. 
Apart from the direct initiatives to cover the banks, there have been other initiatives in 
South Africa to avail end user finance to the low-income. There were ideas to raise 
funds in the market place and lend wholesale to the low-income housing sector. The 
Independent Development Trust Finance Corporation (IDTFC) was established to 
realise that objective. There was also exploration of possibilities of lending pensions 
and provident funds to employed individuals who could only afford mortgage 
payment in the long term. Once the mortgage was fully paid, these funds could revert 
to the pension fund (Tomlinson, 1997: 12). The schemes never took off (Tomlinson, 
  
5 This is based on my notes from discussions in the National Review of the PHP, on the 1st and 2nd Dec 
2003, at Burgers Park Hotel, Pretoria, South Africa.
105
 1997: 13), but could find their way back through the Financial Services Charter, 
currently being negotiated. The other experimental model was the Community Bank, 
where communities would be the shareholders, making them bear the risk in case of 
defaults on loan repayments. The bank, launched in 1994, was operating as a 
conventional bank. It was capitalised through soft loans from various sources notably, 
the ABSA Bank. It made operational and strategic errors and could not raise the 50 
000 customer base that it required to break even. However, it did lend to redlined 
areas (Tomlinson, 1997: 15). The government in South Africa has and is still doing a 
lot to get the private sector to make housing finance accessible to the low income, 
with very little success.
 The exploration of the attempts by the governments in Kenya and South Africa to get 
the private sector to finance low-income housing with limited success makes me 
conclude that any housing policy hinged primarily on private sector finance is likely 
to fail. The private sector exists to make profits. As long as low-income housing is not 
the most profitable area for investments, they are unlikely to lend there wilfully. In 
Kenya, heavy state borrowing from the local markets made it unattractive to lend even 
to the middle-income for housing. In South Africa, the banks have been making 
enough profits by lending in comfortable traditional areas. Why are they expected to 
suddenly change course? This is further complicated by the fact that most low-income 
black households tend to be first time borrowers as opposed to the white middle class 
that the banks were used to lending to (Tomlinson, 1997: 10). Other issues that make 
the banks not lend to the poor include: the costs of origination of small loans; the 
structure and operations of formal banking institutions; lack of appropriate security ?
 often banks cannot sell the low-cost house they would be giving loans for (Tomlinson 
1997; 1999: 289). It would be interesting to see how the self-initiated (by the banks) 
?Financial Services Charter? that is currently being discussed benefits the poor.
 3.3.3. Inadequacy and inappropriateness of mortgage and micro-finance for low-
 income housing
 Most of the funding mechanisms from the private sector discussed in the previous 
section are structured as mortgages. In this section, I discuss the failure of mortgage 
finance for housing in Kenya and South Africa, taking the position that in the first 
106
 instance, mortgage finance is largely inappropriate for the low-income, while micro-
 finance is largely exploitative. 
In Kenya there are not very many private sector mortgage institutions. The two major 
players in long-term mortgage finance are the Housing Finance Company of Kenya 
(HFCK) and East African Building Society (EABS). In the Kenyan context 
prohibitive loan requirements and high interest rates, between 20-30% (Syagga, 
Mitullah and Karirah-Gitau, 2001: 104) exclude the low-income and most of the 
middle income from accessing mortgage finance. Other factors include requirements 
for acceptable security, down payment, salaried income, etc. 
The Mortgage Indemnity Fund, in South Africa was intended to deliver at least 50 
000 mortgages in the 1st year of its initiation in 1994. This did not work (Tomlinson, 
1999: 288). The question is whether this was the appropriate finance mechanism for 
low-income housing in the first place. Tomlinson (1999) argues it wasn?t. She 
questions the appropriateness of mortgage finance for the low-income, in the context 
of long-term debts vis-?-vis unreliable incomes. The issue is not merely availability of 
finance, but also the structure of repayment (Tomlinson, 1999: 289). It has been 
observed that low-income people are likely to suffer when exposed to mortgage 
lending as their incomes tend to be unpredictable over time, hence the likelihood of 
default when the economic situation changes (see also Tomlinson, 1999: 289). The 
poor afford less and less when inflation increases, weakening their capacity to manage 
long-term financial commitments. Volatile interest rate is another factor that makes 
mortgages unpredictable (see also Tomlinson, 1999: 289). The products in the market 
are very complicated, even for educated people, making them all the more 
incomprehensible to the poor who often also have limited education (see also 
Tomlinson, 1999: 289). In case of default in payment the poor tend to lose their assets 
as there are very few housing products in the market that they could afford in the 
event of the sale of their houses (Tomlinson, 1999: 289; Bond and Tait, 1997:27). 
Tomlinson (1999: 20) concludes that: ?Research has revealed that it is cheaper to take 
three small micro loans over two years, each at 46 per cent interest, than (to) take one 
mortgage loan for 20 years at 20 percent interest? (Tomlinson, 1997: 20).
107
 Are micro-loans really the answer? Rust (2002) argues that in South Africa, they are 
not wholly effective as the Usury Act regulates interest on loans larger than R 10 
000.00. This implies that micro-lenders would rather give smaller loans where they 
can charge higher interests. Thus larger loans beyond R 10 000 are virtually 
impossible to get. The exploitation that takes place through micro-finance is covered 
in literature. Mthweku and Tomlinson (1999: 11) report that the ?umashonisha?
 charge up to 300% interest on their loans. The micro-finance arena is packed by many 
uncouth operators. Boaden and Karam (2000: 20) observed that taxi owners, drug-
 lords and ?shebeen? operators in Cape Town with ready cash, were coercing those 
they had lent money to sell them their houses. Baumann (2003: 95) also records this 
phenomenon. Thus while some opportunities exist in micro-finance, a lot of work 
need to be done before it can be a real alternative source of finance for low-income 
housing. This leaves savings as one of the means through which housing finance is 
accessed in Kenya and South Africa. I discuss this next.
 3.3.4. Low-income housing delivery through savings: cases from Kenya and 
South Africa
 There are robust savings groups both in Kenya and South Africa. I review the extent 
to which they have been vehicles for access to housing finance and draw some 
lessons. The groups vary in scale and scope. One thing that remains clear is that they 
have not managed to influence policy. The support they get from the state (in South 
Africa) is mainly on ad-hoc basis. This limits their potential contribution to housing 
finance.
 In South Africa, the state housing subsidy programme has impacted negatively on 
savings. Baumann (2003: 99) argues that the ?free? housing comes with changed 
lifestyle, increased expenditure and new financial obligations around ownership that 
reduces the capacity of the poor to save. The problem is further captured in the 
quotation from Mthweku and Tomlinson (1999: 1, 2):
 ?Trevor Manuel, the Minister for Finance, has stated that the government 
policies are being hamstrung by one of the lowest savings rates in the 
world?Despite this?community-based savings schemes-known as 
108
 ?imigalelo?, ?moholiswana?, ?stokvels?, and so on ? have played a prominent 
role in townships and informal settlements??
 While the housing subsidy in South Africa (from 2002) required beneficiaries to make
 a down payment of R 2 475.00 (with certain exceptions), there is no evidence that this 
has encouraged a culture of saving. Issues like incremental savings and the mobilising 
power of saving schemes for development are not fully explored; hence the potential 
of savings to strengthen the state?s housing programme remains unexploited. Savings 
can be an effective way of mobilising local communities, as demonstrated by such 
groups as the Homeless People?s Federation, in South Africa, which I discuss in 
subsequent sections.
 In Kenya the most common mechanism for savings as discussed later is the 
cooperative model. This is one of the main ways in which the middle-income groups 
access housing. However the impact of cooperatives, in housing finance, against a 
backdrop of a national housing deficit estimated at 750 000, is negligible. The 
cooperative model of saving is relatively weak in South Africa. 
Mobilisation of finance through grassroots is better developed in South Africa than 
Kenya. The Kenyan initiatives are relatively recent. In the case of Muuungano wa 
Wanavijiji their approach is influenced by the uTshani model in South Africa. While 
grassroots organisations are seeing the need to learn from each other, with positive 
results, the governments have not been as proactive. Support for savings in both 
countries comes mainly from the NGO sector.
 Savings are important in low-income housing finance for several reasons. While on 
the one hand they bring in the much-needed additional finances for home 
improvement, on the other hand consistent savings demonstrate affordability, not only 
of housing finance, but also of home ownership (Mthweku and Tomlinson, 1999:14). 
Savings, as reflected both in the Kenyan and South African cases, can be used to 
mobilise communities towards development. They provide alternative affordable 
finance and could be used to bargain with formal financial institutions, as 
demonstrated by uTshani and Letsema in South Africa. Commitment to consistent 
savings show perception of value of one?s housing, which could help avoid sale of 
109
 state subsidised houses way below market values as was observed by Boaden and 
Karam (2000: 20) and by Baumann (2003) in South Africa. In Kenya, savings could 
be another basis for developing an implementable housing policy. In the next sub-
 sections I outline various examples of savings schemes for housing in Kenya and 
South Africa.
 Pamoja Trust, Kenya6
 Pamoja Trust is an NGO founded in Kenya in 2000 to support the federation of the 
urban poor, Muungano wa Wanavijiji. The latter is composed of local savings groups. 
Pamoja Trust is a member of Slum/Shack Dwellers International (Weru, 2004: 33). 
They have been influenced in their approaches by the organisations of slum dwellers 
from India and South Africa, e.g. the uTshani Fund, which I discuss latter.
 Muungano wa Wanavijiji, with the assistance of Pamoja Trust has established a 
support infrastructure for local saving groups (Weru, 2004: 55). By 2002, there were 
54 savings schemes with 43 bank accounts (Weru, 2004: 41). They have developed 
100 savings groups in 60 settlements with a membership of 10 000 people in at least 8 
Kenyan towns (Weru, 2004: 56). These groups have their own internal governing 
structures and banking accounts. Their management is layered: with the local groups 
forming neighbourhood units; several neighbourhood units form the regional units 
and the regional ones, in turn, form the national federation. Daily savings schemes are 
encouraged. People tasked with management of daily savings walk from door-to-door 
collecting loan repayments and the savings (Weru, 2004: 40). They collect and 
disseminate other information at the same time. The trust supports savings campaigns, 
and local and international exchange visits, which have been done with India, South 
Africa, Zimbabwe, Namibia and Uganda.
 Pamoja Trust and Muungano wa Wanavijiji have recently set up a Trust Fund to lend 
to local savings schemes. The Fund is called Akiba Mashinani (grassroots savings). It 
has received contributions from Ruben and Elizabeth Rausing Trust (U.K.). The fund 
is to be capitalised through contributions from the community savings groups and 
  
6 Based partly on communications with Jane Weru, Director Pamoja Trust, in July 2004 and August 
2005.
110
 local development agencies. More funding has been agreed with the Kenya 
Community Development Foundation. Akiba Mashinani will lend to savings scheme 
for onward lending to members. The members are expected to contribute 10%, their 
local savings schemes 10%, and the fund 80% loan, towards house construction 
(Weru, 2004: 44). However, this organisation is relatively new and its housing output 
is still limited, with most being at planning stage. Some 39 units have been built by a 
group in Huruma, Nairobi (Weru, 2004: 60). Its? successes or failures in housing 
finance can only be assessed fairly in the future. However, it shows a great potential 
in helping the urban poor access housing finance.
 SACCOs in Kenya
 SACCOs are the main savings groups in Kenya. Syagga, et al. (2001: 109) report that 
in 1989 there were more than 200 SACCOs in Nairobi. They loaned members Kshs 
7.5 billion. SACCOs used to lend at subsidised rates of 4%, which was initially 
pegged by law through the Cooperatives Act. This was very low as banks charged 
over 25% on their mortgages at the time (Syagga, et al., 2001: 109). The fixed interest 
rate has been withdrawn. Members now decide how much interest they want to pay 
on their loans. 
One example of the cooperatives is the National Housing Cooperative Union 
(NACHU), which was registered in 1970 through the Cooperative Societies Act, CAP 
490 of the Laws of Kenya (Mwaura, 2002:3). It is unique since most of its members 
are predominantly low-income, with 45% of its members earning below Kshs 1 
500.00 per month (Syagga, et al., 2001: 110). It lends to several local cooperatives 
namely: Huruma (928 members); Marura (9 members); Soweto-Kayole (39 members) 
and Embakasi (18 members). NACHU operates more as a micro-finance, lending 
short-term housing loans, i.e. repayable in 48 months, at an interest of 19% (Syagga, 
Mitullah and Karirah-Gitau 2001: 111). It lends wholesale to the cooperatives for 
purchase of land and the cooperative facilitates repayment and takes liability for 
default by individual members. 
111
 uTshani Fund, Village Banks, SACCOs, Kuyasa, Ithala, Letsema in South Africa
 The uTshani Fund is an initiative of the South African Homeless People's Federation. 
The Federation represents some 100 000 households, including 1 200 savings 
schemes and 1 000 homeless communities. The Federation was formed in 1991 out of 
a network called the People?s Dialogue on Land and Shelter (Huchzermeyer, 2004a: 
120). Its membership is 85% female, and the average household income of Federation 
members is under R715.00 per month. The Federation?s primary goal is ?to develop 
its members? capacity to conceive, control and implement their own poverty 
alleviation strategies via the development of their own communities? (SAHPF, 2005).
 uTshani Fund was set up in 1994 as a Section 21 company, owned by the South 
African Homeless People?s Federation (Bolnick, 1996). ?uTshani Fund has been 
established for the sole purpose of obtaining, consolidating and delivering finance, on 
a collective basis, to savings schemes affiliated to the South African Homeless 
People?s Federation?  (SAHPF, 2005). It was conceived as an instrument to leverage 
funds from other actors in the private and public sector. It has an asset base of R77 
million. It was awarded R 10 million by the first post-apartheid housing Minister, Joe 
Slovo in 1994 (received in 1995). The current Minister of Housing, Lindiwe Sisulu, 
further gave the fund an additional donation of R 10 million on the 29th of May 2005 
(SDI Bulletin 13, May 2005). uTshani started giving loans in 1995, with money from 
a German funding agency Misereor. ?By late 2003, using loan funds, members? 
savings, subsidies, and other resources, the Federation had built more than 14 000 
houses? and its savings groups had accumulated R11 million in savings (SAHPF, 
2005).
 There are three types of savings through the uTshani initiative. They are: daily 
savings (nsuku zonke), the ?granary? (inqolobane) and housing savings. The first 
category is for consumption and crisis loans. Inqolobane is for investment in 
production and small enterprises (SAPHF, 2005). The last one is dedicated housing. 
uTshani runs nine regional accounts, which are the main means of access to credit by 
members (Mthweku and Tomlinson, 1999: 8; Rust, 2002: 16). The fund?s loans are 
offered at subsidised interest rates fixed at 12% for 15 to 24 years. Fixing the interests 
is a way of hedging the impacts of inflation on housing finance (Mthweku and 
112
 Tomlinson, 1999:8). However the question remains why this initiative is not 
influencing policy.
 The other model in South Africa, are the Village Banks, developed by the 
International Fund for Agriculture and Development (IFAD) and African Rural 
Agricultural Credit Association (AFRACA). In this model members save and are 
given loans on the basis of their savings (Mthweku and Tomlinson, 1999: 9). 
Another local savings group, initiated by the NGO Development Action Group 
(DAG) operating in Western Cape, is Kuyasa Fund. It offers households loans 
equivalent to three times a member?s savings. One must save for at least six months to 
qualify for Kuyasa loans (Rust, 2002: 16). 
There are also savings and credit cooperatives (SACCOs) in South Africa, operating 
under the Savings and Credit Cooperative League. There were about 3000 people 
belonging to these in 1999. They then had an asset base of R4 million. They had given 
70% of their loans for home improvement. They charge low interest rates between 6 
and 10%, and give loans as high as R 25 000.00 for up to 5 years. Default rates on 
loans are low, as members are shareholders and one does not borrow more than they 
have in shares (Mthweku and Tomlinson, 1999: 10). 
The other model is the Ithala Savings Scheme. It is a deposit and loan scheme in 
Kwazulu Natal. It had 45 branches and was competing with the banks by offering 2% 
higher interest on savings. The disappointment though was that it was only lending 
2% of its accumulated R640 million to the poor (Mthweku and Tomlinson, 1999: 11). 
There is the Letsema National Women?s Group, which acts an umbrella body for 
various provincial women?s organisations. Started in 1998, its aim was to pool 
together various members? savings with a view to negotiate better borrowing rates 
from the formal financial sector. Members contribute at least R 50.00 monthly, which 
a representative collects and deposits in Letsema?s accounts and receives a deposit 
slip. They pay their members 2% interest. Between 1998 and 1999, their savings had 
risen from R 30 000.00 to R 500 000.00 (Mthweku and Tomlinson, 1999: 12).
113
 Another group, which I study in more details in Chapter Six, Section 6.3.2, is 
Masisizane Women?s Cooperative, Johannesburg. It was started as a Rotating Savings 
and Credit Association (stockvel) in Ivory Park, Johannesburg, in 1991. Members, 
who were mainly women from Ivory Park, contributed R 20 every Sunday and built 
their houses through sweat equity. At it?s peak, in 2002 Masisizane had over 3 000 
savers and had constructed over 300 houses. However, it started experiencing 
problems when it was transformed into a cooperative to support the state?s People?s 
Housing Process7. 
3.4.  Labour, materials and building standards
 In this section I discuss issues relating to access to labour, materials and technology, 
and building standards. I argue that there is diminishing relevance of auto-
 construction in self-help housing, with an increase of focus on self-management. I 
also discuss why uptake of alternative materials is very low, both in Nairobi and 
Johannesburg. Lastly, I explore the subject of building standards. While alternative 
and appropriate building standards for the poor are recognised in both cities, the 
approaches for engagement with them help retain the status quo, and where there is 
limited success, these end up enforcing the already well-entrenched patterns of 
development, where the poor get only inferior services. I suggest need to revisit 
incremental standards for all, as the way forward.
 3.4.1. Local labour and small-scale contractor networks in Nairobi and 
Johannesburg
 The first issue that I discuss is that of self-build. In both cities, there was widespread 
evidence that self-managed housing processes have replaced self-build. In both cities, 
this is attributable to availability of cheap labour. Where sweat equity persisted, it was 
because it was enforced by the supporting organisations for various reasons. 
Examples from Nairobi include the Welfare Advisory Council (WAC) in Dandora 
  
7 This information is based on in-depth interviews with leaders of Masisizane, Mobre and Jotham 
Moyo, on the 12th and the 15th of August 2004, and also with discussions with Planact staff in 
September 2004. Earlier discussions with Masisizane, including a talk with the late leader Anna 
Mofokeng was carried out together with Wits Housing Students in May 2003.
114
 and Kabiro Human Development Programme (KHDP) in Kawangware (see Chapter 
Five for more detailed discussion). In South Africa, the People?s Housing Process 
requires beneficiaries to contribute sweat equity in lieu of the R 2475.00 (US$ 
393.00) contribution towards subsidised housing. Given that most beneficiaries have 
no construction skills, coupled by the fact that there is cheap labour available, this has 
limited the beneficiary contribution to self-build. In the cases I investigated in Ivory 
Park and Diepsloot, semi-skilled local labourers drove the house construction process. 
Self-build is unnecessary, even undesirable in current self-help housing processes.
 Most construction in Nairobi is through a network of small-scale contractors, working 
both in the formal and informal sectors. In many instances they provide labour for 
bigger construction companies, through subcontracting arrangements. There is a much 
stronger participation of small-scale firms and groups of individuals in delivery of 
housing in Nairobi than in Johannesburg (see for example Wells and Wall, 2001). The 
development of the labour intensive cottage industry in Kenya is associated with the 
increase in informal access to housing, particularly linked to the collapse of public 
housing programmes. In Nairobi, informal contractors do not merely undertake 
building projects for the low-income, but also for some high-income developments. 
There are a series of linkages enabled through small scale sub-contracting. There are 
also exploitations and benefits, which accrue from these. However, their role in larger 
scale housing delivery is largely unrecognised in policy. 
In South Africa, the developer driven approach dominates low-income, mass-
 produced, housing as discussed in Chapter Two. Formally, there are contractual 
processes that allow backward linkages to small contractors. However, this is not the 
case in practice. The definition and facilitation of emerging contractors leave no room 
for informal players (see Fitchett, 2001). There are several regulations relating to 
empowerment in the construction industry, including the Preferential Procurement 
Policy Framework Act No. 5 of 2000. The Act uses a point system to assess bids for 
state programmes. The points are linked with the extent the bidder uses/intends to use 
previously disadvantaged groups through sub-contracting arrangement. This had very 
limited success, as many major contractors have chosen to focus on the private 
commercial sector rather than meet affirmative action requirements in the public 
115
 sector. The other issue is that this requirement is not a Department of Housing 
Requirement per se, the onus of enforcement lies with the Department of Labour.
 3.4.2. Low uptake of alternative materials and technologies for low-income 
housing
 The Kenyan government created the Housing Research Development Unit (HRDU) at 
the University of Nairobi in 1965. It was later renamed Housing and Building 
Research Institute (HABRI). Its principal focus was on low-cost housing 
technologies. It is interesting that in the 1980s South Africa similarly created the 
National Housing Research Foundation (NHRF) to resolve the Bantu housing 
problems with focus on mass production technologies (Wilkinson, 1998). In Kenya, 
there were no real savings from alternative technologies, as the scale was never big 
enough to recover costs of production8. Market distortions also undermined any 
benefits that could accrue from alternative technologies. In South Africa, debate has 
been on the quality of the living environments in the mass-produced settlements, as 
they ignored broader socio-economic and ecological aspects of housing. In Nairobi, 
such groups as the Intermediate Technology Development Group (ITDG), Shelter 
Forum and HABRI, remain some of the key players in housing research. My 
fieldwork indicated that although there is a disconnection between the originators of 
alternative technologies and the targeted users, the local labourers and contractors are
 informed about them to some extent. But they do not use them none-the-less. Why? I 
argue in Chapter Seven, that low-uptake of alternative materials and technologies is a 
reflection of consumption patterns favoured by the market and often indirectly 
promoted by the state, through regulation or entrenched practices.
 3.4.3. Contradictions in search for appropriate building standards
 Closely related to the issue of appropriate technologies is the issue of building 
materials and standards. The generic debate focuses on two central issues: that the 
building standards inherited from European colonial powers tended to be unrealistic 
and unattainable by the low-income in African cities; on the other hand, poor 
  
8 Discussion with Prof. Paul Syagga, former head of Housing and Building Research Institute 
(HABRI), University of Nairobi.
116
 construction standards are responsible for various problems experienced in the post-
 colonial African cities. The debate has erroneously assumed that the poor need 
inferior standards, while higher standards should be preserved for the rich. I take 
Watermeyer?s (2003) position that high standards are good for all, the only difference 
should be the way the poor and the rich access these. The former could access then 
incrementally, while the later should pay to access them instantly.
 The appropriate building standards debate is old in Kenya (see Tutts, 1996). The need 
for appropriate building standards was recognised, at least on paper by 1979, when 
the National Development Plan stated that:
 ?All municipalities were to review their housing standards in order to make 
them appropriate for the settings to which they were to be applied and to 
reduce them to a minimum, consistent with(in) the provision of low cost 
housing needs at reasonable cost. The standards applicable would be the ones 
capable of being raised/upgraded as development and incomes become higher? 
(Syagga, et al. 2001: 98).
 However there is a general feeling both in literature and in the policy that building 
regulations are still too restrictive (Republic of Kenya, 2003:30). The Building Code 
prescribes the building standards, materials, foundations, openings, etc. Nairobi 
allows for relatively low building standards, on paper, based on Code 95, from the 
1993 ?Building By-Laws and Planning Regulations Review Final Report? (Wasike, 
2000). The Nairobi City Council is aware of the need for enabling building standards, 
but is either unwilling or unable to implement them9. Otherwise it is not uncommon to 
allow selective application of alternative standards. Large-scale examples where this 
has been done include Ayany Estate, in Nairobi, where the whole estate was built on 
stabilised earth blocks, and Komarok Estate (also in Nairobi), where sisal fibre re-
 enforced tiles were used. Current efforts in Kenya are led by ITDG ? Kenya, who 
have implemented alternative construction and planning standards in smaller towns 
like Nakuru and recently in Mavoko Municipality, in the neighbourhood of Nairobi.
 According to Pottie (2003: 128), in South Africa, appropriate construction standards 
were meant to have been ensured by the National Home Builders Registration Council 
  
9 Personal communications with Elijah Agevi, the Regional Director, ITDG, in July 2004. See also 
Agevi (1987)
117
 (NHBRC) established in June 1995. It was to set minimum standards, register 
contractors and provide warranty for the houses for five years. Pottie (2003: 128) 
suggests that they were ineffective in the low-income sector as they only covered 
housing valued at R 25 000.00 upwards, thus not benefiting most of the low-income 
households. In the National review of the People?s Housing Process (PHP)10, it was 
felt that NHBRC was more of an obstacle in realisation of housing through self-help, 
which assumes that the owner ? not an NHBRC registered contractor ? would build 
the house (see also Napier, 2003: 344). They were seen to collect money from small 
contractors without providing any meaningful protection to the homeowners. The fact 
that PHP houses are exempted from NHBRC requirements further helps entrench the 
view that they are of inferior standards. For example they cannot be accepted as 
collateral by financial institutions. These issues are critical for incremental housing 
and standards. Watermeyer (2003) interprets accessible building standards as starter 
standards acceptable to all, but which are to be improved upon gradually. Initial low 
construction costs could be reconciled with long term needs for the highest standards 
by all through this approach. The other angle of accessible standards relates to 
infrastructure and service provision, which I will discuss in the next section.
 3.4.4. Collapse of developmental control in Nairobi
 Another issue, relating to labour, materials and technology, mainly in Kenya, is weak 
enforcement of development control. According to the Commission of Inquiry into 
Local Authorities in Kenya (Republic of Kenya, 1995, in Syagga, Mitullah and 
Karirah-Gitau, 2001: 85) ?there have been numerous cases of blatant disregard of 
planning regulations, resulting in mushrooming of unplanned settlements, misuse of 
road reserves and hence serious strain on services and facilities provision.? Syagga, et 
al. (2001:102) lists the following as some of the factors that contribute to the Nairobi 
City Council?s ineffectiveness to enforce development control as per Land Planning 
Act, CAP 303: 
??there are minimal funds to enforce development control by the NCC and 
those who enjoy political protection and/or those who can compromise the 
  
10 I was one of the resource persons in the National People?s Housing Process Review, from 1st to 2nd
 December, 2003, in Pretoria, South Africa, where I presented a paper titled: ?Potential Professional, 
Technical, Regulatory, Institutional and Training Conflicts and Opportunities Opened Through the 
People?s Housing Process (PHP) Policy Shift in South Africa?.
118
 system, develop in direct contravention of the existing laws and regulation. In 
the uncommon event that such people land in court, they are able to easily pay 
there [sic] way out due to the ridiculously low fines, if at all, and continue 
with the developments.?
 I will now look at some of the reasons why development control has failed in Nairobi, 
through scrutiny of one of the units of the NCC that is meant to be in charge of 
development control in the low income area where one of my case study settlements, 
Dandora, is situated.
 The Housing and Community Development Department (HDD)11 is a government 
department, established in 1983, as a branch of the Nairobi City Council, to deal with 
planning, survey, community development and finance. The technical department of 
HDD is in charge of architectural, technical and planning issues in Dandora, Umoja, 
Mathare North and Kayole settlements in Nairobi. It is staffed by technicians, who 
draw and approve building plans in these areas. It has surveyors to deal with land 
surveys.  The Finance Section manages loan repayments and land transfers. They also 
receive plot rents and land rates on behalf of the NCC. The Community Development 
Department was meant to have been in charge of local self-help groups. However, it is 
the development control function of HDD that I am interested in.
   
11 Interviewed Mr. Ndirangu, Assistant Director, Community Development Section, Housing and 
Community Development Department, Nairobi City Council. This was done on the 1st of April 2004, 
at the NCC/HDD offices in Dandora Phase One, Nairobi. A follow-up interview was done together 
with Research Assistants on the 8th of April, 2004. During this time we also interviewed the Acting 
Director, Mr. Karanja. Additional information was accessed through an interview with Mr. Omunga, 
Planning Division, the Nairobi City Council.
119
 Photo 3.1. Housing and Community Development Department Offices, Dandora Phase 1
 There are several factors why HDD has failed in controlling development in the 
settlements under its mandate. Lack of funding, lack of staff, corruption, unofficial 
withdrawal from its mandate of development control, lack of knowledge of council?s 
regulations, unrealistic requirements by the NCC and operations of powerful 
individuals within and outside of the state were mentioned as some of these factors. 
The previous officers of HDD were said to have been corrupt. HDD?s relatively large 
initial workforce, of 589 persons, also developed many informal linkages, which 
further undermined the department?s efficacy in development control. Further, local 
people were largely unaware of the council?s requirements for development relating 
to plot sizes, plot ratios, etc. There are people in Dandora, involved in informal sale of 
plots as small as 50m2, with no access to infrastructure and services (see photo 3.3). 
Unrealistic expectations by the HDD also undermined development control. In 
Dandora, Umoja, Kayole and Mathare North, only one storey development ?is 
allowed?. Developments of up to six storeys are very common (see photo 3.2). The 
prototype plans developed by HDD for the area still reflect the single storey 
requirements. Powerful individuals continue to build without consulting HDD.  They 
use their individual ties within the government for protection. On the other hand the 
poor develop informally, backed by local chiefs and politicians.
120
 Photo 3.2. Six-storey development in Dandora Phase 5
 Photo 3.3. Informal developments on un-serviced plots in Dandora
 At a broader level, Syagga (2002: 121,122) gives a list of alternative development 
control mechanisms that have evolved in Nairobi to attempt to deal with the problems
 described above. These include: controlled displacement of informal settlements; 
blanket approvals to land buying companies without requiring that they provide 
121
 infrastructure and services before issuance of individual titles; permissive plan 
approval through the former president?s political interventions; permissive building 
standards, where organisations like the National Housing Cooperative (NHC) were 
allowed to use their own alternative standards; private sector development control, the 
best example being the Kahawa Sukari settlement in the outskirts of the city, where 
residents developed their own approval mechanisms. The major problem with all of 
these alternatives is that they are stopgap measures and as such are not widespread.
 3.5. Infrastructure and services
 The two central issues I discuss in this section are the factors inhibiting provision of 
basic infrastructure and services to the residents of both cities, and inaccessibility to 
infrastructure and services standards. It is clear that both cities have not engaged with 
these issues adequately. How could the problems be resolved? Is privatisation of 
infrastructure and service delivery the answer? My case studies in Johannesburg 
clearly suggest that it is not. Is local control by residents the alternative? My studies 
from Nairobi tend to suggest that while this works for higher income areas it is met by 
many obstacles in low-income areas. My position therefore is that the government is 
still the best placed to provide these, although some aspects of infrastructure and 
service provision and maintenance could involve the private sector and the 
communities, in partnership with the state.
 3.5.1. Access to basic services
 In Nairobi, the City Council (NCC) provides infrastructure and services. There is
 consensus however that the NCC has not been able to deliver these adequately (see 
Syagga, Mitullah & Karirah-Gitau, 2001: 52). Informal areas totally lack basic 
services. The other problem is maintenance (see also Syagga, et al., 2001: 52). The 
Nairobi City Council has failed to maintain infrastructure and services. There are 
plans for privatisation of services, starting with garbage collection and water supply. 
The Nairobi Water Company, established through the Water Act 2002, is already 
dealing with water billing. Sewerage and waste management is yet to be privatised. 
This was after failure of an initial agreement with Jacorossi Impresse SPA, an Italian 
122
 firm that had been single sourced to deliver the services (Daily Nation, 25th May 
2004). The inefficient local authority has chosen privatisation as a panacea to all its 
problems. There might be useful lessons to learn from the City of Johannesburg?s 
experiences on this issue.
 Johannesburg delivers most of its services through ?corporatised? structures: ?utilities? 
and ?agencies?. Each utility and agency is run as a business entity, with management 
hired on performance contracts and tasked to cut subsidy levels (over five years) by R 
100 million (US$ 15.9 million). ?Utilities? are those services, which individuals 
consume and pay for directly e.g. power, water, etc. ?Agencies? are those that are 
consumed collectively, e.g. parks, zoos, etc. The question of reduction of state 
subsidies for infrastructure and services was dealt with through the 1995 Urban 
Infrastructure Investment Framework (RSA, 1995 in Bond, 2003: 50, 51). It says in 
part:
 ?Services and infrastructure will be introduced in line with the affordability 
levels of the communities affected. The principle that people should pay for 
the services to which they have access is central. This means that the level of 
services in each area should relate to what consumers there can afford and are 
willing to pay. Where government support is needed to ensure basic service 
delivery, it will be provided transparently. Deliberate steps will be taken to 
remove any disguised subsidies. Limited cross-subsidies to enhance household 
affordability and secure ?lifeline? consumption will be necessary?.
 The question remains on what corporate management portends for service delivery to 
the urban poor. The challenge from a theoretical perspective to this mode of delivery 
is: to what extent can a city consider its citizens as its clients in the strict sense of 
market operations and profitability? In any event, evidence from my case studies 
show that these corporatised units are not necessarily more efficient than the public 
sector agencies, e.g. more than 2000 households in Diepsloot do not receive any 
billing for their water consumption12. What does this portend for equity in 
infrastructure and service provision? The question Bond (2003: 50,51) asks is whether 
this is tenable in a country where there is such a high incidence of poverty (and 
inequality) (see also Beal, Crankshaw & Parnell, 2000: 114-116). 
  
12 Information based on interviews with Diepsloot Councillor, Salphina Mulanzi in August 2004, and 
February, 2005.
123
 In both cities, the approach to service provision is impacting negatively in access to 
other resources for self-help housing. While I understood the general logic of the 
Nairobi City Council, that they do not provide services in unplanned areas, as services 
are linked with the planned development of the city, my concern was that those in 
informal areas needed these even more than those in planned areas; from an 
emergency point of view. Evidence from my Johannesburg case study areas showed 
clearly that where services had been provided people were investing more in their 
houses, because service provision to the informal areas was considered as a sign of 
legitimisation of informal development. There is a link between service provision, 
legitimacy of informal settlements and unlocking of housing resources that need to be 
appreciated in policy.
 3.5.2. Infrastructure and service standards
 I presently shift back to the debate on standards. Apart from the house construction 
standards discussed previously, other issues that affect low-income housing, 
particularly the self-help are prohibitive infrastructure and services standards that are 
required by local authorities. In Nairobi alternative standards are accepted in principle 
but not in practice, while in Johannesburg, high standards for all remain intact.
 In Kenya, infrastructure and service standards are stipulated in many pieces of 
legislation and regulation. The Public Health Act, for example, prescribes standards 
for lighting, ventilation, safety and temporary buildings. Planning Standards also 
stipulate infrastructure and services standards, in addition to prescribing urban land 
use. Road Planning Standards prescribe statutory road sizes and other details. The 
other tools that affect infrastructure and service standards directly are: Development 
Control Codes, and Land Use Zoning and Fire Regulations. In Kenya, standards for 
construction of roads, including local access roads, are considered to be the most 
difficult to attain (Syagga, et al. 2001: 99). However, accessible infrastructure and 
services standards remain illusive (see Republic of Kenya, 2003).
124
 In South Africa, tools like Less Formal Township establishment Act 113 of 1991 give 
a window of opportunity for alternative service and infrastructure standards, although 
it does not expressly say so. Instead it gives the ?Administrator? discretionary powers: 
?If he is of the opinion that such a servitude or condition is inconsistent with, 
or undesirable in relation to, the use, occupation, development or subdivision 
of that land as a township, or that the cancellation of the servitude or condition 
in accordance with formal procedures will unnecessarily delay the opening of 
a township register in respect of that land? ?.
 ?then such conditions may be cancelled as authorised in Section 12 of the Act, titled: 
?Exclusion of laws and suspension of servitudes and restrictive conditions?. 
According to Napier (2003: 344) in spite of the differentiation enabled by legislation 
with regard to standards, the authorities draw no lines between the different 
requirements for different areas. Infrastructure and service standards remain high for 
all in Johannesburg.
 3.6. Summary and conclusions
 In both cities, means of access to land by the poor are limited, in spite of extensive 
legislation to deal with this in South Africa. Invasion, informality and illegality 
dominate access to land, with Nairobi being the worse off of the two. Poor 
constitutional and legislative frameworks, together with corruption and interests of the 
political class in Kenya have worsened the problem. The case of South Africa shows 
that dealing with these, while keeping property rights sacred, is only likely to entrench 
the status quo. There is a need to look at the ways the poor access land in both cities, 
with a view to developing policy and legal instruments to support these.
 The two cities case studies show that while private sector and mortgage finance can 
be used to access housing by the middle-income, they are not likely to work for low-
 income groups. Savings and state subsidies are the better bedrock for funding low-
 income housing. However, there are no clear mechanisms in both cities to support 
savings by the poor towards housing. In Kenya, there is need to open up debate on 
targeted subsidies towards low income housing. Micro-finance also is inadequate and 
inappropriate in both contexts.
125
 Cheap labour is available in both countries due to high incidence of unemployment. 
This has resulted in decline in self-construction. What one saw, both in Nairobi and 
Johannesburg, is self-management of the construction process. Big contractors are 
leaving housing, in Nairobi and in Johannesburg. Nairobi shows that small-scale 
contractors can support housing development, but need state support, while 
Johannesburg shows that the government can institute formal linkages between big 
and small contractors, although this yields little when the big contractors have other 
choices on where else to work. Nairobi shows that weak development control can 
have long-term negative impacts on the quality of the built environment, although it is 
a haven for self-builders, while Johannesburg shows how strict development control 
inhibits self-help initiatives. This calls for a search for an appropriate level of control
 and support. There is also need to explore incremental standards and promote 
alternative and accessible use of materials and technologies in both cities.
 Privatisation of supply of infrastructure and services being pursued by both cities will 
result in an increased asymmetry in access to these. Poorer neighbourhoods are likely 
to be increasingly marginalised. Privatisation is not necessarily the answer to 
inefficient supply as seen in Nairobi. In addition, uniform requirement for cost 
recovery for these is not appropriate in countries where the majority of the population 
live below the poverty line. There are unexplored opportunities in both cities for 
public/private/communities partnerships to supply, manage and maintain local 
infrastructure and services. Lastly, alternative standards for infrastructure and services 
have not been adequately explored in the two countries. Tables 3.1 and 3.2 outline 
some of the similarities and differences in access to resources in Nairobi and 
Johannesburg.
126
 Table 3.1. Similarities and differences in access to land in Nairobi and 
Johannesburg
 Similarities in access to land
 Nairobi Johannesburg
 Invasions and illegality Invasions sometimes result in upgrading; 
increase in illegality
 Weak anti-eviction legislation laws. Ad-hoc 
evictions.
 Stronger anti-eviction legislation, weak 
implementation
 Differences in access to land
 Nairobi Johannesburg
 No subsidies Land subsidies
 Weak land registration, only 6% of the 
country regularised
 Strong but slow land regularisation, long 
wait for titles
 Private access to land only
 Predominantly rental tenure
 Access through state-subsidised housing 
programme
 Predominantly owner occupation
 Outdated constitutional order & inadequate 
land management framework
 Transformative constitutional order, but 
conservative application in urban areas
 Political abuse, corruption, land grabbing Detailed legislation
 Table 3.2. Similarities and differences in access to finance in Nairobi and 
Johannesburg
 Similarities in access to finance
 Nairobi Johannesburg
 Market has failed to avail affordable finance 
to poor because of high state domestic 
borrowing
 Market has failed to lend to the poor in spite 
of state?s incentives
 Exploitation through micro-finance, weak 
regulation; fluctuating interest
 Exploitation through micro-finance, 
fluctuating interest, in spite of regulation
 Mortgage finance unavailable to the low-
 income
 Mortgage finance unavailable & 
inappropriate for low-income
 Differences in access to finance
 Nairobi Johannesburg
 Private sector has failed to fund low income 
housing
 State subsidies for housing
 No attempts Stronger attempts to get the private sector to 
lend to the poor
 Micro-finance loosely regulated Micro-finance unavailable beyond R 10 000, 
impact of the Usury Act
127
 Table 3.3. Similarities and differences in access to labour, materials and technology 
in Nairobi and Johannesburg
 Similarities in access to labour, materials and technology
 Nairobi Johannesburg
 Big contractors left housing when direct 
state sponsorship stopped
 Big contractors in developer driven 
approach; overall capacity declining, 
because of interest in other sectors
 Low-uptake of alternative technologies Low uptake of alternative technologies
 Recognition of appropriate and incremental 
standards, but no implementation. Lower 
standards for the poor
 Few opportunities for incremental standards. 
Lower standards for the poor
 Differences in access to labour, materials and technology
 Nairobi Johannesburg
 Better developed network of small scale 
contractors, better linkages between formal 
and informal labour markets
 Little space for small contractors, in spite of 
legislation, none-for the less formal ones
 Weak development control, counter 
productive
 Quality control through NHBRC, counter 
productive
 Table 3.4. Similarities and differences in access to infrastructure and services in 
Nairobi and Johannesburg
 Similarities in access to Infrastructure and services
 Nairobi Johannesburg
 Privatisation perpetuates asymmetry in 
access to basic resources along income lines.
 Privatisation perpetuates asymmetry in 
access to basic resources along income lines.
 Differences in access to Infrastructure and services
 Nairobi Johannesburg
 Inefficient provision of infrastructure and 
services, based on cost recovery
 Cost recovery for basic services & subsidy
 Beginning of privatisation. Privatisation through corporatised units 
showing some weaknesses