Informal Land DeliveryProcesses in GreaterGaborone, Botswana

Constraints, opportunitiesand policy implications

Faustin T. Kalabamu

International Development DepartmentSchool of Public PolicyThe University of BirminghamEngland

Department of Architecture and PlanningUniversity of BotswanaGaboroneBotswana

Info

rmal

Lan

d D

eliv

ery

Pro

cess

es i

n A

fric

an C

itie

s

Pol

icy

Bri

ef -

2

Preface 1Introduction 4- Policy implications and recommendationsLand tenure 6- Pre-colonial land delivery and tenure systems- Colonial land tenure systems- Post-colonial land tenure systems- State land- Tribal landUrban growth 12- Economic growth and increased inequalities- Rapid urbanisation- Policies- LegislationGaborone 18- The development of Greater Gaborone- Formal land delivery- Demand for state land- Informal land deliveryAssessment 31- State-led land delivery systems- Land for women- Land for youths- Land for the poor- Interpreting Botswana’s successPolicy 34- ‘Ownership’, sale and leasing of state land- Codifying tribal/customary land tenure- Right of avail and entitlement- Demolition of ‘illegal’ structures- Compensation for arable land in tribal areas- Other issuesConclusion 37Footnotes 38Publications 41

Contents

International Development DepartmentSchool of Public Policy, J G Smith Building,

The University of Birmingham,, Edgbaston, Birmingham B15 2TT, UK

Website: www.idd.bham.ac.uk

PrefaceWhy research informal land deliveryprocesses?

The colonial powers in Africa introduced urban landadministration systems that were modelled on thesystems of their home countries. The extent to whichindigenous tenure systems were understood,recognised and incorporated varied from colony tocolony, but it was generally believed that only aformal system based on a European model couldprovide a framework for urban development andprotect the rights of urban property owners (who atthat time were expatriates). These landadministration systems, which were inherited atindependence, are governed by formal rules set outin legislation and administrative procedures.However, the legislative provisions and theadministrative systems that were established toimplement them proved quite unable to cope withthe rapid urban growth that occurred afterindependence.

The state-led approaches to development favouredin the 1960s and 1970s were associated with large-scale public intervention in urban land deliverysystems. However, the cost of implementation andcompliance has been too high for low-incomecountries, cities and inhabitants. At their extreme,land and property markets were perceived asineffective or exploitative. These views weretranslated into attempts to de-marketise land bynationalisation and/or government control over landmarket transactions. Whether or not the conceptson which such land policies were based were sound,limited capacity at national and municipal levelsensured their failure. Administered land supply hasvery rarely met demand and attempts to regulateand register all transactions in land and property havebeen universally unsuccessful. As a result, most landfor urban development has been supplied throughalternative channels.

In the early years of rapid rural-urban migrationmany households, including poor households, were

able to get access to land to manage the constructionof their own houses for little or no payment, through‘squatting’ or similar arrangements. Followingresearch in the 1960s and 1970s, there was a feelingthat the processes of ‘squatting’ and the allocationof customary land by legitimate rights holders werefairly well understood. Upgrading projects of the1970s were designed and implemented on this basis.

Most countries have now reversed some of the mostextreme versions of state intervention, but othercomponents remain despite serious implementationfailures. There is considerable doubt about whetherrecent attempts to improve land management willbe any more successful than previous approaches.In part, pessimism about the prospects for efficientand equitable urban land management arises fromthe continued lack of resources and capacity ingovernment, but it also stems from doubts aboutthe appropriateness of the principles and conceptson which recent urban land policies have been based.

Much research on land and property in Africantowns and cities assumes that the state has both theduty and the capacity to take on a majorinterventionist role in land management. Itconcentrates on documenting and explaining thefailures (and more rarely successes) of stateinterventions. Despite their significant role inproviding land for urban development, there hasbeen relatively little recent in-depth research onprocesses of informal land delivery or the institutions(rules and norms of behaviour) that enable them tooperate and that govern the relationships betweenthe actors involved. To improve policy and practice,a better understanding is needed of how formal andinformal systems operate, interact and are evolving.

Aims of the research

The aim of the project was to improve understandingof informal land delivery processes in six Africancities and their relationships with formal landadministrative systems. It analysed the

Informal land delivery processes in African cities

1

characteristics of informal land markets and deliverysystems

to increase understanding of the institutions thatunderpin and regulate transactions and disputesin landto assess the strengths and weaknesses ofalternative land delivery mechanisms, especiallywith respect to the extent to which they enablethe poor and other vulnerable groups (especiallywomen) to access land with secure tenure, andto identify and explore implications for policy.

The comparative research project

Coordinated by Carole Rakodi of the University ofBirmingham and Clement Leduka of the NationalUniversity of Lesotho, studies were undertaken insix medium-sized cities in Anglophone Africa, in allof which informal land delivery systems areimportant, but which also typify different colonialand post-colonial policies, legal frameworks,governance arrangements and experiences. Thecities and the local researchers were:

The aims of the project and the methodologicalapproach were jointly developed by the researchers.Findings and policy issues were discussed atworkshops in each of the cities, to obtain feedbackfrom relevant stakeholders and make a contributionto current debates about land policy andadministration in each of the countries studied. Theresearch teams generally identified some of the policyimplications of their findings rather than makingdetailed recommendations, because the researchersall believe that policy formulation and legislativechange should be negotiated processes involving allthe stakeholders in land management.

The research was funded by the UK Departmentfor International Development. DFID supportspolicies, programmes and projects to promoteinternational development. It provided funds for thisstudy as part of that objective but the views andopinions expressed are those of the authors alone.

Eldoret, Kenya Rose Musyoka, Department of Physical Planning, Government of Kenya

Enugu, Nigeria Cosmas Uche Ikejiofor, Federal Ministry of Works and Housing, Gusau,Zamfara State, Nigeria

Gaborone, Botswana Faustin Kalabamu, Department of Architecture and Planning, and SiamsangMorolong, Department of Law, University of Botswana

Kampala, Uganda Emmanuel Nkurunziza, Department of Surveying, Makerere University

Lusaka, Zamba Leonard Chileshe Mulenga, Institute for Social and Economic Research,University of Zambia

Maseru, Lesotho Clement Leduka, Department of Geography, National University of Lesotho

2

g A B O R O N EAB

OR

ON

E

The project in Gaborone

Botswana was included in the study as an exampleof a southern African country, in particular one inwhich the British colonial administration had adopteda protectorate system of indirect rule. Both incolonial times and since, the powers of the traditionalauthorities, especially those related to land, havebeen reduced but safeguarded. In Gaborone, therehas been extensive public sector formal plotprovision. Nevertheless, informal settlement hasincreasingly occurred outside the urbanadministrative boundary, demonstrating bothproblems with the formal land delivery system andprocesses by which traditional land allocationsystems, with state sanction, are adapting to urbandemand.

The research was carried out by Faustin Kalabamuof the Department of Architecture and Planning andSiamsang Morolong of the Department of Law,University of Botswana. The author would like tothank all the individuals and organisations that helpedin the research. In particular, the contributions ofthe research assistants (L. Gitonga, N. Tema, R.Mselle, A. Mogonono, K. Kgarathi, M. Madziba,R. Masilo-Rakgoasi and J. Motsuminyane) areacknowledged, as well as that of Dr B.K. Aquahwho participated at the beginning of the project. The

respondents and focus group participants in OldNaledi and Mogoditshane are thanked for theirpatience in answering our questions. Finally, thecontribution of participants in a policy workshop heldin Gaborone on 9th February, 2004 isacknowledged.

Following the workshop, a full report of the studywas published: Kalabamu, F.T. and Morolong, S.(2004) Informal Land Delivery Processes andAccess to Land for the Poor in GreaterGaborone, Botswana, Birmingham: University ofBirmingham, School of Public Policy, Informal LandDelivery Processes in African Cities Working Paper3 (ISBN No. 0 7044 2246 8, see alsow w w . i d d . b h a m . a c . u k / r e s e a r c h /researchprojs.htm

For further information contact Faustin Kalabamu,Department of Architecture and Planning, Faulty ofEngineering and Techology, University of Botswana,Private Bag 0061, Gaborone, Botswana, E-mailkalabamu@mopipi.ub.bw

3

g A B O R O N E

gA

BO

RO

N

Although Botswana is a large country (58,200 km2)with a small population (1.7 million inhabitants in2001) it has land tenure and management problemsthat arise, in the first instance, from its geographicand environmental characteristics. More than twothirds of the country’s land mass is covered by theKalahari Desert or sandveld, which supports someshrubs and grasses but lacks surface water. Due tothe lack of large natural water bodies and surfacerivers, more than 80% of the country’s population,settlements and economic activities are concentratedalong its eastern border or hardveld (Figure 1). Thehardveld has better soils and surface water andreceives more reliable rainfall than the sandveld.

In light of the above geographic and environmentalcharacteristics, the people of Botswana developedcustomary land tenure and management systems thatsought to ensure equitable access to and use of land-based resources. However, the advent of colonialismand Western influences appear to have constrainedthe customary land tenure systems, forcing them toadapt. The changed circumstances provided newopportunities and challenges for utilising land as asource of livelihoods. These called for new rules andprocedures for accessing, owning, transferring, anddisposing of land rights. Closely related to this hasbeen the question of who administers, controls andallocates land rights. This question is importantbecause, as Berry notes, those who control landaccumulate power and wealth and those who havepower control the distribution of land and wealth1.

The objectives of this summary are

a) to highlight the main features of customary landdelivery processes and how these have beeninfluenced by the legal, economic and politicalstructures introduced during the colonial and postcolonial eras;

b) to identify and analyse the characteristics ofinformal land markets and delivery processes inGaborone and its surrounding settlements;

c) to assess the strengths and weaknesses of formaland informal land delivery mechanisms in andaround Gaborone; and

d) to identify and explore the implications for policy.

The report is divided into eight sections. After thisintroductory section, land tenure systems in the pre-colonial, colonial and post-colonial periods arediscussed in turn. Next, urban development, policiesand legislation are analysed. These sections providea context for exploring formal and informal landdelivery processes in Greater Gaborone. The lastbut one section is an evaluation of land deliveryprocesses in Greater Gaborone, while the lastsection provides recommendations and variousoptions for improving land delivery processes in thestudy area (see also p. 5).

Introduction

4

Figure 1: The Republic of Botswana

Source: GOB, 1997

Policy implications and recommendations

i. It should be made clear to recipients of state land that they are leaseholders rather than outrightowners.

ii. Cost recovery for the capital costs of infrastructure provided to plots allocated by governmentshould be spread out over a period of time, to avoid the anti-poor bias caused by insisting on up-front payment.

iii. Leaseholders should be free to sell their outstanding rights and interests in land, in the interest ofpromoting more efficient urban land markets.

iv. Codification of customary land tenure is desirable and should include clarification of the role ofland boards.

v. Land boards should be regarded as trustees rather than owners of land. They should be guided byand accountable to the relevant local community.

vi. Each citizen should be entitled to a single lifetime grant of a residential plot on tribal land. Additionalplots should be purchased through the market.

vii. Structures erected without permission should become state property rather than being demolishedand alternative penalties devised for those who do not comply with regulatory requirements

viii. The basis and arrangements for paying compensation to indigenous land rights holders whose landis taken by a tribal land board for re-allocation should be reconsidered, to reduce resistance toand evasion of the official process for delivering land held under customary tenure for urban uses

5

g A B O R O N E

gA

BO

RO

N

Typical homestead

Land tenure

6

Pre-colonial land delivery andtenure systems

Besides the geographic and environmentalconditions highlighted above, pre-colonial landtenure and delivery processes in Botswana werelargely shaped by subsistence farming systems, aswell as the patriarchal structures and institutions thatcharacterised the various tribal groupings in theregion. Then, land was the most important sourceof livelihoods, as it supported survival through theextraction of wild vegetables, honey, hunting andfirewood, as well as the grazing of goats, sheep andcattle and limited production of cereal crops –notably sorghum. At that time, cattle ownership wasthe sign of wealth, social status and power.

Social and administrative structures

Each tribe was headed by a hereditary Chief, whowas by far the most important member of the tribe.He had outstanding privileges and authority2. Eachchief was head of a civil-cum-military structureentrusted with judicial, legislative, religious and otherpowers for maintaining law and order. He wasassisted by ward headmen who exercised authorityand privileges similar to the chief’s. Although chiefsand headmen were powerful leaders, they wereneither absolute rulers nor autocratic despots.Instead they ruled by consensus3.

In terms of social structures, Tswana tribes wereboth patrilineal and patrilocal in that they were basedon patriarchal structures and institutionscharacterised by male political leadership anddominance and corresponding female subservience4.Every girl who came of age was expected to bemarried to a man and to adopt the home of herhusband as her domicile throughout her life – unlessshe divorced the husband. Upon divorce, the womanwould return and live with her parents (if they werestill alive), her brother or any other male relative inher father’s lineage. Divorce and begetting childrenoutside wedlock were strongly discouraged although

men could marry more than one wife. Infidelity bymen was condoned and tolerated more than infidelityby women5.

Within each household, women, girls and young boyswere responsible for building and maintaining housesas well as food production and processing6. Womenwere socially and legally excluded from participatingin political, judicial, military, religious and otheractivities located in the public sphere of men. Men,as heads of households, represented their wives,unmarried women and male children in all tribal andcommunity meetings, courts and gatherings.However, the main duties of men were cattlemanagement and hunting.

Settlement forms and patterns

Due to ecological factors and the need to protecteach other, members of each tribe resided in anucleated village. Some large tribes, for example,the Bamangwato, lived in one huge village – the tribalcapital – and several satellite villages. Inhabitants ofsatellite villages were considered to be part andparcel of the tribal capital. Although villages movedfairly frequently, their general layout remained thesame. Land in each village was subdivided into fiveuse zones: residential, cultivation (masimo), grazing,hunting and forestry. The residential zone formedthe nucleus of the village and was often sited on ahill where the land was less fertile but the site easierto defend. Surrounding the residential zone werecultivation areas, which were, in turn, surroundedby pastureland, forestry and hunting areas7.

Within residential zones, members settled accordingto wards. A ‘ward’ had both a social and spatialmeaning: it was a group of families related to oneanother through blood or marriage and who livedtogether in a small village or a section of a largevillage8. The chief’s ward, together with the tribalkgotla (village assembly or assembly point) andkraal, occupied the centre of the village and wassurrounded by other royal wards, while wards ofthe least royal units were sited on the village

7

g A B O R O N E

gA

BO

RO

N

periphery9. No person was permitted to settlepermanently outside the village residential zone underany circumstances. The continual shifting of villages,according to Hardie, “allowed for the settlementpattern to be dynamic, accommodating changesin the internal organisation as they becamenecessary”10.

Land tenure and rights

The most outstanding feature of the pre-colonialcustomary land tenure system was the Right of Avail– a right that was uniformly applied to all andautomatically shared by all people belonging to eachtribe11. It was from this right of avail that other rights– individual or communal – were deduced. All malesiblings had equal rights to be allocated land fromtheir fathers’ holdings or from the tribal reserve, orto inherit land from their fathers, without having tomake any form of payment or tribute to the chief orheadman. Allegiance to the chief was the only pricethat men paid for acquiring and maintaining landrights. Inheritance followed the paternal lineage.Women accessed land only through their husbands,fathers, brothers or other male relatives in theirfathers’ lineage. No land was allocated to women –whether single, divorced, widowed or in any othercircumstances – except among the Bakgatla, wherewomen could be granted arable land rights.

All allocated or inherited pieces of land remainedpermanently the exclusive property of the concernedfamily. Land rights could only be extinguished if theholder ceased to belong to the tribe. However, intimes of need (e.g. congestion or shifting of thevillage), land rights could be reallocated as part ofthe ward or village re-organisation process. Contraryto common European misconceptions at the time ofcolonisation, chiefs and headmen did not own triballand – they only administered its use, allocation andtransfer. Ownership of tribal land was vested in theentire tribe while chiefs acted as trustees orcustodians12. Except for land portions reserved forthe use of the chief and his family, none of the landwas said to belong to the chief in his individual andpersonal capacity.

Every individual and family belonging to the tribehad a right to access, graze cattle and harvest naturalresources (water, honey, firewood, etc.) fromunallocated tribal land or land that was not in activeuse at a particular time or season. Members of atribe “were free to travel, hunt and collect orharvest natural resources anywhere within [thetribal] territory provided they did not causedamage to improvements (e.g. crops) on land”13.The extended land rights were considered necessaryin view of the frequent and abrupt differences in landsuitability.

Figure 2: Colonial land tenure

Source: GOB, 1970:

8

g A B O R O N EAB

OR

ON

E

Colonial land tenure systems

Within four years of what is now Botswanabecoming a British Protectorate in 1985, thecountry’s land mass was divided into threecategories: (i) native reserve lands or tribal territories;(ii) crown lands; and (iii) freehold lands.

Freehold and crown lands

As shown in Figure 2, freehold land covered someof the most fertile land along the country’s easternborder and included farms (e.g. Tati Concession)acquired by European settlers through concessionsmade by chiefs14. The Ghanzi Block was createdand allocated to a column of Boers to act as a bufferagainst German expansion from South West Africa(now Namibia)15. However, the British colonialadministration discouraged widespread settlementand investment in Botswana by European farmers,miners, industrialists and traders and, instead,promoted labour migration from Botswana to SouthAfrica16.

Most of the land covered by the sandveld andutilised by the hunter-gatherer San and Hottentotstribes was considered unutilised and, consequently,declared crown land and placed under the authorityof the British crown. The High Commissioner wasgiven the power to make grants or leases on crownlands on any terms and conditions he deemed fit.

Changes to tribal land tenure

Five tribal reserves were defined in 1899: theBamangwato, Bamangwaketse, Bakwena,Batawana and Bakgatla17. Four additional territoriesfor the Bamalete, Batlokwa, Barolong and Bakalakawere later carved out of freehold farms. No reserveor territory was created for hunter-gatherer tribes.The administration of land within the tribal reservesremained with the respective chiefs and headmen.The British administration expected customary rulesand procedures for access, use and transfer to beapplicable to all land within each tribal reserve18.

However, contrary to these expectations, thetraditional land tenure system did experience somemodifications. The pressure for change emanatedfrom several interrelated factors: the introduction oftaxes, exposure to non-indigenous consumer goods,labour migration (mainly to South Africa) andcommodification of cattle.

To offset some of the costs of administering theProtectorate, the colonial administration introducedseveral taxes - all payable in cash by every Africanmale adult of the apparent age of 18 years or above.Men who did not own cattle were forced to seekpaid employment (mainly in South Africa) while thosewho owned cattle were forced to sell some of theirstock19. This resulted in economic inequalitiesbetween the majority who were taxpayers and thetax collectors (chiefs and headmen). The latteraccumulated considerable numbers of cattle aspeople sold or surrendered their beasts in lieu ofcash payments. Through labour migration some menbecame cash-earners and were able to acquire morecattle, while some women became beer brewers andsellers to returning migrants20. At the same time, dueto the prolonged absence of men while in paidemployment abroad, women increasingly assumedroles that had hitherto been undertaken by men.According to Larsson, by the 1930s unmarriedmothers had set up independent households, whilemany married women had become de factohousehold heads.

As rich people accumulated more and more cattle,they began to benefit from services (e.g. diseasecontrol and water from boreholes) that were initiallyoffered to European farmers. “[P]romoted bymarket forces, boreholes began to be recognisedas personal property in the 1930s, and movestowards the private ownership of adjacentgrazing lands, a direct and logical extension,soon followed”21. Owners of boreholes did notpermit cattle belonging to other people to graze closeto their boreholes lest the cattle were watered fromthese private facilities.

9

g A B O R O N E

gA

BO

RO

N

The above phenomena resulted in three changes tocustomary or tribal land tenure. First, individuals orsyndicates who owned boreholes created zones ofexclusive grazing rights around their boreholes, whileothers fenced their arable fields thereby retainingexclusive use of the land year round and denyingother tribe members their traditional rights of grazingon fields after crop harvesting22. Second, chiefsstarted allocating residential and arable land rightsto women who had become de facto or de jureheads of households. Third, money paid to chiefsby Europeans and concession hunters had started aprocess of land commercialisation. Thus both landand cattle became tradable commodities.

Post-colonial land tenure systems

The post-independence government adopted andmaintained the three tenure systems it inherited fromthe colonial administration. It has, however,progressively changed the rules and procedures onaccess, use, transfer and entitlements on state land

(formerly crown land) and tribal land, as well asplaced a moratorium on creation of new freeholdland from state land. The State Land Act 1966 (Cap.32:1) vested the trusteeship of state land in thePresident. Some state land has been converted totribal land and some freehold land purchased andconverted to state land. As a result, the proportionof tribal land increased from 49 percent in 1966 to71 percent by 1998 as shown in Figures 2, 3 and 4.Between 1966 and 1979, the proportion of land instate ownership decreased from 47 percent to 25percent and has remained the same thereafter. Theproportion of freehold land increased slightly fromfour percent in 1966 to six percent by 1979 andthen declined to four percent by 1988. Today,Botswana has one of the lowest rates of freeholdland in the region. In comparison to about fourpercent of freehold in Botswana, freehold/privatelyowned land accounts for about 72 percent in SouthAfrica, 44 percent in Namibia, 41 percent inZimbabwe and 38 percent in Swaziland23.

0

10

20

30

40

50

60

70

80

Tribal land State land Freehold land

1966

1979

1998

Source: Adams, M, Kalabamu, F.T. and White, R. (2003) “Land tenure and practice in Botswana: governance lessons forsouthern Africa”, Austrian Journal of Development Studies, XIX(1), p. 56

Figure 3: Land tenure categories in Botswana, 1966-1998

10

Figure 4: Contemporary land tenure Deeds registers FPSG leases as well as freeholds.A leaseholder may sell or transfer thetitle to any persons other than the stateonly after satisfying all developmentcovenants contained in the lease grant.Furthermore, the holder may only sellor transfer what remains of his or hergrant period, which leads to a declinein land values as the termination dateapproaches, to the extent that a“position is reached where it is notin the owner’s interest to furtherdevelop the land or even maintainthe property”24.

Tribal land

Within two years of independence in1968, the government passed theTribal Land Act (Cap. 32:02) which,among other things,

(i) Transferred the administration andpower to allocate land rights fromchiefs and headmen to land boardsestablished under the same act;(ii) Vested the right and title to land in

each tribal territory to a local land board, themembers of which served as trustees;

(iii)Made provision for the establishment of commonlaw leases on communal grazing land.

To the Botswana government, the passing of theTribal Land Act was necessitated by two factors:the need to improve livestock production 25 and theneed to provide people with land rights that couldbe used as economic assets, that is, rights which areboth saleable and bankable26. To Mathuba, the Actwas necessary in order to provide a written law thatcould easily be referred to27. However, accordingto Ng’ong’ola, the passing of the Act sought tostrengthen the hand of the Government by providinggreater political control over land administration28.At that time, chiefs and headmen were being

State land

The State Land Act (Cap. 32:1) enables thePresident or any person authorised by the Presidentto dispose of state land. In practice, centralgovernment departments (e.g. the Department ofLands and the Department of Wildlife and NationalParks) and urban councils administer state land.Since the 1976 moratorium on freehold alienation,state land is leased out to individuals and privatecompanies under Fixed Period State Grants (FPSG)and Certificates of Rights (COR). The FPSG is a50-99 years capitalised lease: a lease with the totalrental paid at its commencement rather thanperiodically. At the end of the FPSG lease period,the land together with all improvements on it revertsto the state without compensation. The Registrar of

g A B O R O N EAB

OR

ON

E

Source: Botswana National Atlas, Department of Surveys and Mapping

11

increasingly accused of favouritism and misuse oftheir powers to administer and allocate land. TheTribal Land Act was later – in 1975 – augmentedby government’s adoption of the Tribal Grazing LandPolicy (TGLP) that formalised the exclusive grazingland rights enjoyed by owners of boreholes andlarge herds of cattle.

The good intentions of the Act notwithstanding, anddespite several subsequent amendments to it, triballand remains a contentious issue because of anumber of factors, including:-

a) Failure to codify the customary land tenuresystem, thereby exposing it to various andcontradictory interpretations.

b) Failure to take cognisance of all forms of landrights – notably hunting and gathering.

c) Failure to appreciate the need for and the generaltendency towards commercialisation of all typesof land rights. Until 1985, common law leasescould only be granted to holders of grazing landrights. To date, arable (masimo) land rights arenot readily bankable or saleable.

d) Exclusion of certain types of land rights frombeing vested in land boards, notably grazing andborehole rights.

e) Land boards’ lack of grass roots representationand their dependence on chiefs and headmen toverify whether certain pieces of land are availablefor allocation.

Most contentious claims and counter claims havecentred on the interpretation of Sections 10(1) and(2). Section 10(1) in its original formulation vestedownership of tribal land in land boards on behalf ofthe people belonging to a particular tribe. Section10(2) removed pieces of land acquired before theact was passed from the jurisdiction of the landboards. Some individuals and the High Courtinterpreted section 10(2) to mean that all landallocated prior to the establishment of land boardsis held by individuals and their heirs in their privateand individual capacity29. The two sections were

amended in 1993 when land was vested in landboards on behalf of the citizens of Botswana andSection 10(2) deleted. Despite the deletions andamendments, ownership of tribal land and the extentof customary land rights remain contentious issues,particularly in peri-urban and urbanising villages.

In order to address shortfalls in land board structures,the Government has continuously changed theircomposition, the qualifications of land boardmembers and the procedures for electing themembers. At present all land board members areeither appointed by the Minister or vetted by acentral government ministry. Chiefs and activepoliticians have been excluded from standing for landboard membership to avoid their politicisation,although the Minister’s power to appoint land boardmembers may itself be regarded as being political.

Tribal land rights may be held under Certificates ofCustomary Land Grant or common law leases.According to Section 20(2) of the Tribal Land Act,customary land rights may not be granted for non-customary land uses such as trading, manufacturing,business or commerce, nor may they be granted toa non-citizen. Customary land rights may beconverted into 50 or 99 year common law leases atthe grantee’s initiative and costs, in addition to theproduction of a diagram or plan approved by theDirector of Surveys and Mapping. Common lawleases are renewable, saleable and bankable andmay be registered by the Registrar of Deeds.

Upon expiry of a common law lease, landautomatically reverts to customary land tenure forcitizens and to the land board for non-citizens,without compensation for any improvements effectedby the lessee (Section 25) unless the land boardelects to pay compensation. Leaseholders are,however, permitted to remove any improvementsthat can be removed without causing irreparabledamage to the land within six months of thetermination of the lease.

g A B O R O N E

gA

BO

RO

N

12

Urban growthUrbanisation in Botswana is basically a post-colonialphenomenon. As of 1960, only four settlements -Lobatse, Kasane, Ghanzi and Francistown - wererecognised as towns by the 1955 TownshipProclamation. Although the Proclamation providedthat only settlements located on crown land couldbe called ‘towns’, Francistown, which was locatedon freehold land, was included in the list of townspresumably because the majority of residents wereEuropeans. At the time of independence in 1966,the country did not have clearly stated long-termurban development policies and legislation partlybecause over 95 percent of the population lived inrural areas( Figure 5).

However, since the discovery and exploitation ofdiamonds, copper and nickel in the late 1960s,Botswana has experienced rapid economic growthand far-reaching social and demographictransformations. These have forced the governmentto devise and adopt a series of administrativearrangements, laws and policies to meet thechallenge.

Urban development, land policies and legislation

Figure 5: Population growth in Botswana

Sources: Government of Bechuanaland Protectorate (1965) Report on the Census of Bechuanaland Protectorate, MardonPrinters, Bulawayo; GOB (Government of Botswana) (1971) Report on the Population Census 1971, Government Printer,Gaborone; GOB (1983) 1981 Population and Housing Census: Guide to Villages and Towns of Botswana, Central StatisticalOffice; GOB (1992) Population of Towns, Villages and Associated Localities, Government Printer; GOB (2002) Populationof Towns, Villages and Associated Localities, Government Printer.

0

200000

400000

600000

800000

1000000

1200000

1400000

1600000

1800000

1964 1971 1981 1991 2001

Botsw ana

Total urban population

Tow ns

13

g A B O R O N E

gA

BO

RO

N

Economic growth and increasedinequalities

Although Botswana received minimal investmentfrom the British and remained a poor country untilthe early 1970s, when exports of diamonds andcopper-nickel began, it has since experiencedtremendous economic growth. The country hasrecorded steady economic growth averaging 6percent annum during the entire post-independenceperiod. Consequently, the per capita GrossDomestic Product (GDP) rose, at constant prices,from Pula 1,683 in 1966 to Pula30 7,863 in 1995,and grew in real terms by 8 percent in 1999/2000and 9 percent in 2000/131. While the contributionof the agricultural sector to GDP fell from 43 percentin 1966 to 4 percent in 1995, the contribution ofmining rose from zero at independence to about 17percent in 1975/76, reached a peak of about 49percent in 1985/86 and then declined to about 34percent in 1994/95. The Government has utilisedthe mineral revenue to finance infrastructuredevelopments, provide services and diversify theeconomy. Wildlife-based tourism, which wasinsignificant during the colonial era, is currently thesecond most important source of foreign currencyearnings. Formal sector employment increased from69,500 in 1978 to 234,500 in 199532.

Colclough and McCarthy attribute Botswana’sexceptional economic performance to the“efficiency, planning capacity, ability to spend,and negotiation skills” of the Botswana leadership,rather than to political stability, regional groupings,or economic factors such as the discovery of mineralresources33. The leadership is composed of “menof landed wealth, with a considerable personalstake in the economy”34 who deserted thechieftaincy “located in commercial agriculture,which also possessed long experience ofgovernment and social control” 35. Unlike othercountries on the continent “where government isusually in the hands of petty bourgeois or

bureaucratic elites, and policy often has an urbanand consumptionist bias, when not downrightpredatory and corrupt”, the leadership inBotswana has been keen to ensure that neither theeconomy nor the administration founder36. It has,therefore, strived to spend within its means,accumulate foreign reserves and diversify theeconomy. This strategy has enabled the country toattract more foreign aid and investment than othercountries in the region.

Despite the country’s economic boom and thegovernment’s generous expenditure on developmentprojects and public welfare programmes, poverty(defined as “an inability to meet basic needs”) iswidespread, especially in rural areas, whereapproximately 60 percent of all poor householdsand nearly 70 percent of very poor households inBotswana are found37. The predominance of povertyin rural areas has been blamed on a high andincreasing proportion of people without cattle, poorcrop production and unemployment. The proportionof rural households without cattle is estimated tohave increased from 16 percent in some areas (e.g.Katleng District) during the 1930s, to about 33percent in the 1960s, 58 percent in the early 1980sand about 74 percent in the 1990s38.

Due to high unemployment rates among youths andwomen, poverty is more widespread among womenand women-headed households. In 1993/94 about41 percent of all women -headed households werecategorized as poor compared to 34 percent of men-headed households. The relatively high incidence ofpoverty among women-headed households is alsodue to high dependency ratios and low incomes, asthe majority of women who earn cash incomes areengaged in poorly paid jobs39.

Rapid urbanisation

As noted above, at the time of independence mostpeople in Botswana resided in rural areas. Accordingto the 1964 census, the three towns of Lobatse,

14

Francistown and Gaborone had a combinedpopulation of 21,000 inhabitants, accounting forabout 4 percent of the country’s total population.However, by 1971 the total urban population hadmore than doubled, to 59,000 and accounted for10 percent of the country’s total population (Figure5)40. The number of towns had increased to five(with the establishment of the mining towns of Selibe-Phikwe and Orapa). The rapid growth of the urbanpopulation between 1964 and 1971 is generallyattributed to rural-urban migration due to ruralpoverty. The latter arose from low agriculturalproductivity and investment of public funds in urbancentres. While rural expenditure concentrated onimproving veterinary services and construction ofadministrative infrastructure, investments in urbanareas went into mining, construction, manufacturingand other sectors with high job and income multipliereffects. Most of the rural-urban migrants weresingle, male and young (aged 15-24)41.

By 1981, the number of urban settlements hadincreased to eight – with the addition of Jwanengand the villages of Tlokweng and Palapye – and theurban population to 163,000, constituting 17 percentof the national population. The employment patternsin Palapye were determined by the adjacentMorupule coal mining activities, while those inTlokweng were influenced by its proximity toGaborone. Ten years later, in 1991, the number ofurban centres had increased to 24 and the urbanpopulation to 619,000 or 47 percent of the country’stotal population. As of 2001, the urban populationstood at about 877,000 people and accounted for52 percent of the entire population in the country. Itis worth noting that, while in 1981 about 50 percentof the country’s population lived within a radius of200 km from Gaborone, by 1991 the radius hadreduced to 100 km42, thereby greatly increasing thedemand for land, housing and other services inGaborone and its surrounding settlements.

In brief, post-independence urbanisation inBotswana has largely been due to the establishment

g A B O R O N EAB

OR

ON

E

of mining towns (Orapa, Selibe-Phikwe, Jwanengand Sowa) and in-situ urbanisation of traditionalsettlements – especially district and sub-districtadministrative centres, (e.g. Serowe, Kanye,Mmadinare and Moshupa), peri-urban settlements(e.g. Tlokweng and Mogoditshane) and touristvillages (Kasane and Maun). All the former chiefs’capitals are now administrative, service, commercialand industrial centres, while peri-urban settlementsare dormitory towns for people working in the cities.

Rapid urbanisation has had numerous effects on theenvironment, employment, and social and economicwell-being of various groups in Botswana. Theseeffects have been both positive and negative: theyinclude increased demand for land for urbandevelopment and an increased volume of informaltransactions in land and house construction.

Policies

Urban planning and land management policiesadopted by the government since the 1970s haveattempted to deal with the problems andcontradictions caused by rapid urbanisation. Untilthe mid-1970s, the government concentrated on theacquisition and servicing of land for the establishmentand construction of new mining towns and on buildinghouses for civil servants. It introduced self-helphousing schemes in the mid-1970s when it realisedthat the Botswana Housing Corporation (BHC) -established in 1970 - was unable to meet the housingneeds of all urban residents, notably the lowestincome group, without substantial subsidy, which thegovernment could not afford43.

Between 1973 and 1982, under what weretechnically squatter upgrading and site-and-servicesschemes, Self-help Housing Agency (SHHA)departments in various towns serviced some 7500new plots and upgraded an equal number44. All theself-help housing schemes were undertaken on stateland, tribal land converted to state land or freeholdland bought by the state and then converted to state

15

g A B O R O N E

gA

BO

RO

N

land. By the end of 1989 SHHA departments hadserviced and allocated 26,709 plots, whichaccounted for over 53 percent of all residential plotsin Gaborone, Lobatse, Francistown, Selibe-Phikwe,Jwaneng and Kasane45.

Other major policies adopted in the 1970s and1980s included the 1978 Urban Development andLand Policy, the 1982 National Policy on Housingand the 1985 National Policy on Land Tenure, whichcomplemented each other and reflected theprovisions of the Tribal Land Act and the TribalGrazing Land Policy. The main thrust of thesepolicies was to provide serviced land for owner-occupied houses to all urban households and therebypre-empt squatting or the development of informalsettlements.

The launching of the Accelerated Land ServicingProgramme (ALSP) and the introduction of newurban development standards proposed byconsultants and published in 1992 represented animportant departure from previous policies. TheALSP and the new standards sought “to maximizeresidential and marketable land while providingadequate space to meet the physical, social andaesthetic needs of the citizens” and to rapidly placea substantial quantity of serviced land in the marketto meet immediate and future land requirements inthe major urban areas46. Under the ALSP, pit latrinesand public water standpipes, which had until thenhelped to minimise public infrastructure costs for low-income plots, were replaced with private waterconnections and waterborne sewerage systems.

The ALSP produced a total 22,228 plots of which21,840 were for residential uses, 249 industrial and139 for commercial uses47. About 47 percent of theplots were located in Gaborone followed byFrancistown (15 percent). Although all the plots havebeen allocated, a substantial proportion remainsundeveloped or partly developed. By 2001, overhalf the low-income plots serviced and allocatedunder the ASLP were undeveloped (5,185 of

9,325), while 7,799 plots had not been paid for orhad outstanding balances. Some plot holders aresaid to be selling the plots, either because they haveno funds to develop them or to avoid repossessionby the government. According to Datta and Jonesthe ALSP has failed where SHHA programmessucceeded because SHHA plots were granted forfree. In addition, water and sanitation services, aswell as building material loans, were heavilysubsidised48. ALSP plot holders have to pay for theirplots and borrow from financial institutions withoutany form of subsidy. Furthermore, the ASLP appearsto have been rushed, too big for the country’sdomestic economy and to have fuelled ‘demand’for state land plots, as evidenced by the dramaticincrease in plot applications.

According to the 1999 National Policy on Housing,the government decided to change its role from thatof a financier, producer and landlord to that of afacilitator, in partnership with the private sector.Under the state-private sector partnership, thegovernment provides primary services (e.g. majorroads, water and sewer lines) and allocates blocksof land (up to 10 hectares) to private developerswho then subdivide the land and provide secondaryand tertiary infrastructure services. Developers maysell undeveloped plots in the market or develop andsell the products. They are required to designate aproportion of the serviced land for low and middle-income groups. To make plots affordable to low-income households, the government can adopt oneof two strategies. It may buy plots designated forlow-income households from the developer and thenreallocate them to beneficiaries at affordable prices.Alternatively, it may pay a subsidy to the developerupfront. The latter is then expected to sell the plotsto low-income beneficiaries at affordable prices49.In contrast to the SHHA schemes, in which subsidieswent directly to the poor, under this partnership thegovernment is subsidising the private sector in thehope that the subsidies will be passed on to the poor,which is highly unlikely.

16

g A B O R O N EAB

OR

ON

E

The responsibility for supplying houses to thedestitute has been assigned to the Department ofSocial and Community Development, which alsoprovides them with food, clothing and shelter.However, the department “provides free housingwhere the destitute individual has no recourseto any other alternative assistance”50 To assistvery poor and middle poor households who do notqualify for either destitute housing or low-incomeplots under the private-sector partnership, theGovernment introduced Integrated PovertyAlleviation and Housing schemes. The schemes,which are “based on the assumption that povertyis the underlying cause of homelessness”51, seekto integrate skills acquisition, employment creation,income generation and shelter provision by engagingyouths and other unemployed individuals in theproduction of building materials (e.g. bricks) andthe construction of houses52.

Another scheme, the Turnkey Project for SHHA, isin the offing. Under this scheme, the Government, ina joint venture with the Botswana HousingCorporation (BHC), will build houses for the richpoor and middle-income households that cannotafford houses currently offered by BHC and theprivate market. Beneficiaries, to be drawn fromcurrent SHHA waiting lists, will be able to buy theproperties under tenant purchase arrangements.

With the recent demolition of houses inMogoditshane, Gaborone, Selebi-Phikwe and otherurban settlements, the government appears to havereturned to the squatter clearance policy that itabandoned in the mid-1970s. The reneweddemolitions of unauthorised housing developmentshave been rekindled by the increased demand forland and apparent ‘lawlessness’ that characterisedinformal land markets in Mogoditshane between1980 and 2000. The Ministry of Lands and Housinghas been particularly concerned that while “the[Kweneng] Land Board was trying without anysuccess to allocate [plots] in accordance withapproved plans, some field owners started

changing the use of their fields without theauthority of the Land Board, subdividing andselling the subdivided portions as residentialplots” (emphasis added)53. Much as demolitionof ‘illegal’ houses serves to remind people of thepolice power reserved by the state, this policy needsto be revisited and appropriate alternativesinvestigated, because it is flawed in several ways.As will be argued later, the policy impoverishespeople, destroys national wealth, wastes national andland board manpower resources, and createsantagonism between residents, land boards and thegovernment.

Legislation

The Townships Act (Cap 40:02) that replaced the1955 Townships Proclamation deals with thecreation of townships and cities, as well as theestablishment of structures for administering thetownships. Urban development, settlement planningand land use planning and management are governedby the Town and Country Planning Act (TCPA) of1977. This act provides for (a) orderly andprogressive development of both urban and ruralsettlements; (b) the preservation and improvementof amenities; (c) granting of permission to developland; and (d) monitoring the use of land.

The TCPA empowers the responsible Minister todeclare any part of the country or a settlement as a“planning area”. Until 1995, only 7 townships(Francistown, Gaborone, Ghanzi, Jwaneng, Kasane– Kazungula, Lobatse, Selebi-Phikwe and Sowa)had been declared planning areas. With the exceptionof Kazungula and Francistown, all the abovetownships are sited on state land. Kazungula islocated on tribal land while a large part ofFrancistown lies on freehold land. In 1995, the peri-urban settlements of Tlokweng, Mogoditshane, TatiSiding, Matshelagabedi and parts of the North Eastdistrict around Francistown were declared planningareas as were the district and sub-districtheadquarters of Serowe, Maun, Kanye, Molepolole,

17

g A B O R O N E

gA

BO

RO

N

Mochudi and Ramotswa, Mahalapye and Palapye.Section 5 of the TCPA provides for theestablishment of the Town and Country PlanningBoard (TCPB), the functions of which are todetermine applications for permission to developland and to advise the Minister on the preparationor revision of a development plan for any planningarea. The Board consists of a representative of theministry responsible for town and country planning- in practice, the Director of the Department of Townand Regional Planning (DTRP) - who also servesas the Chairman of the Board; five members drawnfrom ministries responsible for public works, land,agriculture and/or any other ministry with an interestin town and country planning; and three peopleappointed by the Minister. In 1980, the Ministertransferred development control functions to District,Town and City councils.

Section 17 of the Tribal Land Act, as amended in1993, appears to have delegated land use planningfunctions in tribal areas (including peri-urban areas)to land boards. However, physical planners havestrongly objected to this, arguing that “DTRP shouldbe the only institution mandated to prepare landuse plans and that the work of the land boardsshould be confined to land allocation”54. Thisargument is mistaken, because only the Minister ismandated, under the TCPA, to prepare land useplans. The Minister may delegate these functions toany institution, including the land boards.

Decisions of the TCPB only become effective after14 days and only then if not disallowed by theMinister. The decision of the Minister is final andmay not be challenged in any court (Section 12 (3)of the Act). The TCPA also empowers the Ministerto consider any application made to the Board andeither grant or reject the application, therebybypassing the TCPB. Again, decisions of theMinister in such cases are final and may not bechallenged in any court (Section 14 (3) of the Act).Furthermore, any developer aggrieved by the

decision of the TCPB has leave to appeal to theMinister who “may allow or dismiss the appeal ormay reverse or vary any part of the decision of theBoard whether or not the appeal relates to that part,and deal with the application as if it had been madeto the Minister in the first instance” (Section 15 (3)of the Act). Once again, the decision of the Ministeron any appeal is final and may not be challenged inany court (Section 15 (3) of the Act). Although theMinister’s decisions are said to be final and notchallengeable in any court, the High Court hasjurisdiction to entertain any appeal if it is shown thatthe Minister acted in bad faith.

The same Minister administers several other statutesrelated to land – namely, the State Land Act (Cap.32.01), Land Control Act (Cap. 32: 11), Tribal LandAct (Cap. 32:02) and the Townships Act (Cap.40:02). These Acts, together with TCPA, highlightthe extensive powers that the Minister and therefore,the state, enjoy in the control, administration, useand management of land and developments thereon.Suspicions abound that Ministers may have at timesabused their powers, as recently alleged by theLesetedi Commission55. The Commission was setup by the President to find out, among other things,whether the Minister responsible for land mattershas powers to allocate State land and whethervarious successive Ministers had followed correctprocedures in certain land allocations and changesof land use. It found that the Minister has no powersto allocate State land and that neither administrativenor legal procedures were consistently followed inmaking land allocations and changing land uses.

18

GaboroneLand delivery in Greater Gaborone

The development of GreaterGaborone

At the time of its conception as the capital ofindependent Botswana in the early 1960s, Gaboronewas a small trading centre with a population of about3,900 Europeans. Then “all that existed of presentGaborone was a railway station with a few landgrants, a hotel, a store … and approximatelythree and half kms to the east [of the railwaystation] was the ‘Government Camp’ housingthe Protectorate government officials, a store,

church, various government offices etc.”56. Thefew non-Europeans who worked at the camp andrailway station resided in Tlokweng - a traditionalvillage east of Gaborone. By 1966 a considerablenumber of government offices had been built andhouses for middle and senior government officialserected between the railway station and theGovernment Camp as shown in Figure 6. A few plots(about 511) were set aside in the Bontleng area forself-help housing because it was then assumed that

labourers employed in the construction of the newcapital would return to their home villages later57.

Gaborone has since grown rapidly in terms of status,population and area covered. As the national capital,Gaborone is the seat of the national government,the Parliament and the House of Chiefs. It housesmost central government ministries and departmentsand the Southern Africa Development Community(SADC) headquarters and secretariat. By 1981,Gaborone, with about 60,000 inhabitants, hadalmost trebled the ultimate population envisaged bythe 1963 Master Plan. In 1986, Gaborone wasraised from Municipality to City status. In spatialterms, it had grown northwards and almost doubledits physical size (Figure 7). Ten years later, in 1991,Gaborone had doubled its 1981 population to nearly134,000 residents and extended westwards acrossthe railway line (Figure 8). According to the 2001census, Gaborone’s population was recorded atabout 186,000 inhabitants.

Gaborone’s population and spatial growth has beenparalleled by the growth of surrounding villages –notably Mogoditshane. As shown in the table, duringthe period 1964-1971, Gaborone recorded anannual population growth rate of almost 24 percent,while the population in surrounding villages declined,which implies that people were migrating from theminto Gaborone. However, since 1971 most of thesesettlements surrounding Gaborone have beengrowing rapidly. Between 1971 and 1981,Metsemotlhaba’s annual population growth rate wasalmost twice that of either Gaborone orMogoditshane. However, during the 1981-1991 and1991-2001 periods, the growth of Mogoditshanesurpassed all the settlements surrounding Gaboroneand the city itself. The slowest growing village hasbeen Tlokweng, while Mogoditshane andMetsemotlhaba have been the fastest. The rapidpopulation growth experienced in these settlementsappears to reveal the inability of formal institutionsto satisfy the demand for land, housing and otherservices.

Source: Department of Survey and Mapping, Gaborone

Mogoditshane

Station

Gaborone

Camp

Figure 6: Gaborone in 1963

19

g A B O R O N E

gA

BO

RO

N

Formal land deliveryGaborone was built on state land surrounded byfreehold farms except on its eastern side where itshared a border with the Batlokwa tribal territory.

During the 1960s, the government supplied about2,800 plots, most of which were developed by thethen Public Works Department for seniorgovernment officers. Only 35 percent of the 2,786plots were designated for low-income houses, ofwhich 515 were built by the Public WorksDepartment and the rest constructed by individualson a self-help basis.

In the meantime, most labourers, who wereexpected to ultimately return to their home villages,built houses for themselves on a piece of state landthat later became known as Old Naledi. Thedevelopment of Old Naledi forced the GaboroneTown Council to initiate its first self-help housingproject at Bontleng. In 1966 the council surveyedand serviced plots about 380 of which wereallocated with very little control over either theselection of beneficiaries or the building process58.The second self-help housing project, Extension 14,was launched in 1972 with financial assistance fromthe UK Overseas Development Administration(ODA). The project, with about 300 plots, waspartly designed to accommodate ‘squatters’ whowere being relocated from state land earmarked forindustrial development.

Source: Department of Survey and Mapping, Gaborone

Figure 7: Gaborone in 1979/1980

Sources: GOB (Government of Botswana) (1972) Report on the Population Census 1971; GOB (1983) 1981 Populationand Housing Census: Guide to Villages and Towns of Botswana; GOB (1992) Population of Towns, Villages and AssociatedLocalities; and GOB (2002) 2001 Population and Housing Census: Population of Towns, Villages and Associated Localities,Gaborone: Government PrinterN.B. Figures in brackets and italics indicate annual population growth rates from the previous census year.

Location 1964 1971 1981 1991 2001

Gaborone 3855 18799 (23.6%)

59657 (12.2%)

133468 (8.4%)

185891 (3.4%)

Tlokweng 3711 3906 (0.5%)

6657 (5.3%)

12501 (6.3%)

21133 (5.3%)

Mogoditshane 2548 1075 (-0.9%)

3125 (10.6%)

14246 (15.2%)

32843 (8.4%)

Gabane 5402 1936 (-0.1%)

2688 (3.3%)

5975 (8.0%)

10399 (5.5%)

Mmopane - 539 584 (0.8%)

1249 (8.4%)

3512 (10.3%)

Metsemotlhaba - 50 395 (20.7%)

1586 (13.9%)

4997 (11.5%)

All the above settlements

15516

26305 (7.4%)

73106 (10.2%)

169025 (8.4%)

258775 (4.3%)

Botswana 514876 596944 (1.5%)

941027 (4.6%)

1326796 (3.4%)

1678891 (2.4%)

Population growth in Gaborone and the surrounding villages

20

g A B O R O N EAB

OR

ON

E

In the mid-1970s, the government purchasedBroadhurst and Tsholofelo freehold farms, which atthat time lay north of the urban boundary, convertedthem into state land and expanded the townshipnorthwards59. The new developments in Broadhurstand Tsholofelo provided about 6,000 plots for low,medium and high-income housing, industries andshopping facilities60. Self-help housing plotsaccounted for 70 percent of all plots in the extension.To decongest Old Naledi, some tenants in OldNaledi were allocated SHHA plots in Broadhurstand Tsholofelo, which according to Dickson,somewhat slowed the growth of Old Naledi61.

The development of Broadhurst and Tsholofelo,“constituted practically a doubling of thephysical size of the town”62. The project wasundertaken in three phases –two of which werepartly funded by the USAID, CIDA and ODA. Thethird phase (Tsholofelo) was wholly funded by theGovernment of Botswana. While low-income plotsin Broadhurst and Tsholofelo were developed on aself-help housing basis, medium and high-incomeplots were developed by the Botswana HousingCorporation, the Botswana DevelopmentCorporation and other private sector developers.

During the 1980s, the government acquired severalfreehold farms - Bonnington Farm, Glen Valley andpart of Content Farm – and converted them to stateland for urban uses. Most of the Glen Valley farmwas put to military use, while Content Farm becamean agricultural college and research centre.Bonnington Farm, which measured about 3,000hectares, was subdivided into residentialneighbourhoods referred to as Gaborone WestPhases I – IV. Gaborone West Phase I consistedof self-help housing plots, medium and high-incomeplots and sites for blocks of flats. The BotswanaHousing Corporation (BHC) developed most of theplots in the latter three categories. In Gaborone WestPhases II and IV, the Botswana HousingCorporation (BHC) developed all the plotsdesignated for ‘low-income’ and medium-income

housing, flats and town houses. High-income plotsin these three phases (commonly known as ‘surveysand lands’ plots) were allocated to parastatals (e.g.the Bank of Botswana, Water Affairs and theBotswana Development Corporation) and individualdevelopers. It is worth noting that, while SHHAschemes were designed to serve the middle poorand rich poor, BHC low-cost houses are meant forthe rich poor and lower middle-income households63.BHC does not build for destitute, very poor or middlepoor people. Gaborone West Phases I, II and IVgenerated close to 15,000 residential plots for amixture of income groups.

Despite the above efforts on the part of theGovernment to service and develop sufficient landfor urban development, supply appears to havefailed to satisfy demand. In 1987 the BHC had atotal waiting list of over 26,000 applicants from all

Source: Department of Surveys and Mapping, Gaborone

Figure 8: Gaborone in 1999

21

g A B O R O N E

gA

BO

RO

N

urban centres in the country. About 20,000 (or 77percent) were applying for houses in Gaboronealone. Three years later, in 1990, the waiting list hadincreased to about 29,000, of which 22,000applications were for Gaborone alone. The demandfor residential plots in Botswana’s seven major urbancentres for the period 1990-2000 was estimated at69,768, of which 44,100 were required in Gaboronealone64. On the basis of these estimations, theGovernment introduced the Accelerated LandServicing Programme (ALSP) discussed earlier.

Under the ALSP initiative, the Government was ableto service a total of 10,210 residential plots inGaborone. About 40 percent of these have beendeveloped by the Botswana Housing Corporationfor rich poor, middle and high-income groups.Parallel to the ALSP scheme, the private sectorserviced 683 residential plots in Kgale Hill (part ofForest Hill farm) and 850 in Phakalane Estate -both of which are freehold land65. However, privatedevelopments in Phakalane and Kgale Hill accountfor only 4.6% of all serviced plots within GaboroneCity boundary.

As a result of the above Government and privatesector efforts, by 1997 the City of Gaborone had atotal of 33,34366 serviced residential plots, of which23,294 (almost 70 percent) were fully developedor being developed. A total of 10,049 state landplots were yet to be allocated or had been allocatedto people who were still paying the ‘purchaseprice’67.

The state has not serviced any additional residentialplots since the ALSP initiative. It has instead, underthe new state-private sector partnership, allocatedblocks of state land to private sector developerswho are required to service the land and either sellplots or build houses for sale. Between 1998 and2001 about twenty companies were allocated a totalof about 75 hectares of state land to service andbuild blocks of flats. The largest block (almost 23hectares) was allocated to Universal Builders, which

has since serviced, developed and sold most of theplots and houses it built. The government intends tocomplement private sector efforts by servicing anddelivering approximately 7,873 plots between 2001and 200568.

Demand for state land

From the foregoing we note that the governmenthas assumed responsibility for supplying residentialplots and housing since the establishment ofGaborone Township. This may have been promptedby several factors, including Tswana culture in whichthe chief allocated land to his subjects, and a coloniallegacy whereby the Government provided housingfor its senior employees. After 1980 generousmineral revenues could have cementedgovernment’s commitment to supply enough land tomeet everyone’s needs.

The above resolve, intentions and effortsnotwithstanding, the demand for land in Gaboroneis far from being satisfied. As noted above, in 1990the Botswana Housing Corporation had a waitinglist of 22,000 applications for houses in Gaboronealone69. As of January 2001, the waiting list forresidential plots in Gaborone stood at 23,50670.Given Gaborone’s population, these figures grosslyexaggerate the demand for land. About 48 percentof the plots serviced and allocated under theAccelerated Land-Servicing Programme remainundeveloped and/or not paid for71. This shows thatmany people have made several applications forland, behaviour to which the government has tendedto be sympathetic.

Speculation in land is fuelled by the relatively cheapprice charged for the residential plots offered by thestate. While the private sector offers plots at a priceof not less than Pula 120 per m2, the highest cost forstate land residential plots is Pula 32 per m2. Theexcessive demand for plots delivered by governmenthas prevented those in need or purchasers withsufficient resources from obtaining land through

22

g A B O R O N EAB

OR

ON

E

formal delivery channels and had forced them toresort to informal land markets.

In brief we note that, despite government’s concertedefforts to supply land, the demand for land inGaborone has not been satisfied due to severalinterrelated factors – namely, the policy ofsubsidising land supply for the rich and poor alike;the inability of poor households to afford the plotssupplied since 1990; and the evolution of a ‘cultureof entitlement’ which drives people to ‘demand’ plotswhen they have neither the means nor the immediatedesire to develop them.

Informal land delivery

The inability of the Government and the formal systemto deliver serviced plots to those who have eitherthe need or the means and an immediate intention todevelop them, has resulted in potential developersacquiring land through informal means (e.g. self-allocation and purchase of undeveloped tribal andstate land) and the emergence of unplannedsettlements such as Old Naledi and Mogoditshane.

Old Naledi

The first and the only substantial unplannedsettlement within the city of Gaborone is Old Naledi,which is as old as the city itself and has its genesisfrom the construction of the national capital. Thesite that is now Old Naledi was part of a block ofstate land that was zoned for industrial uses underthe 1963 Gaborone Master Plan. The settlementstarted as a temporary labour camp for peopleemployed by the construction companies that werebuilding government offices, but later became thebasis for a bigger settlement. Aerial photographstaken in 1963 (Figure 6) reveal no houses on thesite that later became Old Naledi. However, some388 people claim to have resided at Old Naledibefore 196572. Aerial photographs taken on the eveof independence in 1966 reveal a substantial numberof houses on the site (Figure 9). At that time, Old

Naledi covered an area of approximately 24 ha. By1971 the settlement had increased in density andspatial extent (Figure 10), covered about 42 ha andhad a population of 4,075. Old Naledi served as“the first point of call by migrants from thehinterland who had no formal employment andrelied heavily on their kinsmen or tribesmen foraccommodation”73.

The government initially tolerated the existence ofOld Naledi on the assumption that the labourers whooccupied the area would later return to their homevillages. However, later it attempted to remove them.First, it attempted to resettle Old Naledi residentsin the self-help housing project at Bontleng74.Second, the Government mandated the BotswanaHousing Corporation (BHC) to erect low-costhouses in an area called New Naledi – just acrossthe road from Old Naledi – in the expectation thaton completion of the small, mass-produced houses,residents of Old Naledi would be relocated and theiroriginal settlement demolished. However, the OldNaledi residents who were relocated to those houses

Source: Department of Surveys and Mapping, Gaborone

Figure 9 Old Naledi in 1966

Old Naledi

23

g A B O R O N E

gA

BO

RO

N

their plots through self-allocation, 10 percent throughpurchase and 2 percent through inheritance77.

Using building materials criteria, 27 percent of thehouses were classified as low quality (mud, thatch,cardboard or plastic); 48 percent as medium quality(mud, brick, cement block, tin, thatch/asbestos); and25 percent as high quality (brick or cement blocks,tin, asbestos or tiles).

The survey further revealed that about half of theresidents had originated from rural districts close toGaborone- namely Kweneng and Southern Districts.About 48 percent of the 3,303 adults in Old Nalediwere in wage employment, while 39 percent wereself-employed and 13 percent were unemployed.88 percent of the economically active adults weremen - of whom 94 percent had wage employment.However, only 66 percent of the economically activewoman received wages. The dependency ratio wasestimated at 1:2 that is, each working personsupported two other adults. Furthermore, the surveyrevealed that Old Naledi residents - especially men- had strong rural connections. However, most ofthe self-employed female plot ‘owners’ had weakor no rural connections, indicating that they fell inthe category of the very poor.

Upgrading of Old NalediDuring the upgrading process, the number of plotsin Old Naledi was reduced from about 2000informal holdings to 1,610 officially recognised plotsand the population from 10,000 to about 8,600residents78. About 150 tenants and 350 plot holdersdisplaced by the plot rationalisation werecompensated, allocated free plots in either OldNaledi resettlement area or Broadhurst or Tsholofeloself-help housing projects. The plot holders amongstthose displaced were permitted to salvage theirbuilding materials for reuse in their new locations.The upgrading involved rationalisation andregularisation of plot boundaries; construction ofearth roads, storm water drains, communal waterstandpipes, two schools, a health clinic, a community

were completely dissatisfied with their newaccommodation and quickly returned to their originalplots in Old Naledi. It is also possible that rents inNew Naledi may have been unaffordable75.Consequently, by 1971, over 6,000 people or aquarter of Gaborone’s population lived in OldNaledi, which was the first informal settlement inBotswana to be upgraded76.

According to a base line survey undertaken in 1977/78 just before the settlement was upgraded, OldNaledi had a population of 10,019 residents wholived in 2,344 households on 2003 plots. Plot sizesranged from 66m2 to 1200m2 but averaged 320m2.At that time, only 15 percent of the plotsaccommodated more than one household, while thenumber of persons per plot ranged between 4 and8. Of all the plot ‘owner’ households, 83 percentwere headed by men and 27 percent by women.About 88 percent of plot ‘owners’ had obtained

Figure 10 Old Naledi in 1971

Source: Department of Surveys and Mapping, Gaborone

Old Naledi

24

g A B O R O N EAB

OR

ON

E

centre and a police station; and the provision of otherbasic services. The Gaborone City Councilencouraged plot holders to improve the quality oftheir houses by giving them Certificates of Rightsand financial assistance in the form of building materialloans.

In 1988 the City Council tarred all the major roadsin the area, lined all the storm water drains andconnected all the public institutions to the city’scentral sewerage system79. Furthermore, in 1995all the roads and major pedestrian ways wereprovided with street lighting. In 2002, the GaboroneCity Council commissioned Haas Consult, anengineering firm, to advise it on the feasibility ofupgrading Old Naledi to meet the current urbandevelopment and planning standards. Haas Consultenvisages that the exercise will involve demolitionof houses on almost 800 plots to enable expansionof roads, schools, clinics, playgrounds etc. Itestimates that a total of P9,102,050 will be requiredto compensate plot holders whose properties willbe affected and that the whole upgrading exercise islikely to cost over P75 million80.

Assessment of the upgrading and consolidationprocessesOld Naledi appears to have reached saturation point,as evidenced by its slow growth between 1991 and2001. While the population of the area increasedby 200 percent between 1981 and 1991, it onlyincreased by 5 percent between 1991 and 2001.Haas Consult puts the total number of plots in OldNaledi at about 1,700, an increase of about 90 plotssince 1981. Of the plot owners surveyed as part ofthe present study, only 1 percent had subdividedtheir plots.

Instead of subdividing their plots, most owners havebuilt additional structures. Although the maximumnumber of structures per plot permitted under theoperative Development Control Code is 2 –consisting of a main house plus a servant quarter oroutbuilding - there are, according to Gwebu,approximately 6,307 structures on 1,704 residentialplots giving an average of 3.7 structures/plot81. Theaverage number of people per plot doubled from6.1 in 1981 to 12.7 in 2001. Overall, the populationdensity had increased from 20 to 80 persons perhectare between 1978 and 2001. Considering thatOld Naledi is characterised by single storeystructures, these are high densities.

Old Naledi today

25

g A B O R O N E

gA

BO

RO

N

Of the initial sample of plots in our survey in 2002/3, perhaps a third had absentee owners. Of thosefor which questionnaires were completed (which hadresident owners), 71 percent accommodated morethan one household compared to 15 percent in 1978.In 2002/03 one plot accommodated eleven separatehouseholds while, on average, 3.25 households livedon each plot (Figure 11). The average number ofhabitable rooms on each plot was seven, with aminimum of two and a maximum of 19.

Data from our household survey show that 96percent of owner households have access to theirown pit latrine (the rest share a pit latrine with theirneighbours) while 26 percent have their own watertap and 74 percent share a tap with their neighbours.Of all the households interviewed, 12 percent hadno working adult, 30 percent had only one workingadult, 39 percent had two working adults and theremainder (19 percent) more than two. The averagenumber of adults working per household was two.Furthermore, 45 percent of the householdsinterviewed had cattle (an average of 30 head perhousehold). Using the frequency of cooked mealstaken by the family each day, 15 percent of the house

owners in Old Naledi may be categorised as verypoor, 54 percent as middle poor (two meals a day)and 31 percent as rich poor (three meals a day).The high incidence of non-resident landlords indicatesthat many richer plot owners have moved out ofOld Naledi.

Despite improved tenure security, increased incomes(rent plus wages/incomes from employment),building material loans and improved infrastructure,plot holders have not improved the quality of theirhouses as envisaged. They have found it moreprofitable to invest in additional rental rooms than inimproving the quality of the structures. This apparentcontradiction between government expectations andplot holders’ interests is one reason for government’sinsistence on high development standards for lowincome plots under the ALSP projects and theproposed turnkey SHHA project. Completerelocation of Old Naledi was one of the initial optionsmooted in the recent upgrading feasibility study,which has instead recommended demolition of allthe houses on almost half of the plots for roadwidening.

0

5

10

15

20

25

30

35

1 2 3 4 5 6 7 8 9 10 11

%

Figure 11: Number of households per plot in Old Naledi, 2002/03

Source: Field surveys, 2002/03

26

g A B O R O N EAB

OR

ON

E

Mogoditshane

Mogoditshane pre-dates Gaborone Township. Itwas established by the Kwena chief and hisheadmen, initially as a cattle post and later as a borderpost to guard against the encroachment of freeholdland alienation into the Bakwena tribal territory.Although, as shown in the table on p.19,Mogoditshane’s population decreased by more thanhalf between 1964 and 1971, it has since grownquite rapidly – especially during the 1981-1991inter-census period when its growth rate surpassedthat of Gaborone.

With the upgrading of Old Naledi and strictmonitoring of ‘squatter’ housing within GaboroneTownship, effective demand for land spilled overinto peri-urban areas – notably Mogoditshane.According to a study undertaken by KwenengDistrict Administration in 1981, people found iteasier, cheaper and quicker to get a residential plotin Mogoditshane than in Gaborone82. During thisperiod the responsibilities for allocating tribal land

had been transferred (under the Tribal Land Act)from the chief and ward headmen to the ThamagaSub-land Board with its offices in Thamaga (about25 km south of west of Mogoditshane). The latteroperated under delegated authority from theKweneng Land Board, which was situated inMolepolole, some 45 km west of Mogoditshane.In practice, the land board relied on local headmenfor the identification and allocation of vacant plots.The land board would simply confirm allocations toapplicants who presented ‘no objection’ formssigned by the headman83.

Due to increased demand for residential plots asMogoditshane’s population grew in tandem withGaborone’s growth, land allocations started to extendbeyond residential zones to arable areas. Most ofthe people applying for plots in Mogoditshane hadbeen rejected /neglected by land and housinginstitutions in Gaborone84. Concerned about therapid sprawl Mogoditshane was experiencing,Kweneng Land Board assumed direct control over

Source: Sithole, T. R. (1995) An Evaluation of Urban Planning in ‘Per-Urban Settlements of Botswana, unpublishedMasters of Urban and Regional Planning Thesis, Heriot-Watt University of Edinburgh College of Arts, Edinburgh

Figure 12: The division of Mogoditshane into traditional and modern areas

27

g A B O R O N E

gA

BO

RO

N

the tribal land in the village towards the late 1980s.Headmen and land applicants were directed not toidentify plots beyond the ring road (Figure 12)introduced by physical planners as part of theGaborone West structure plan. The area beyondthe ring road was made subject to ‘modern’ townplanning procedures, whereby plots were to bedemarcated before they were allocated.

The first attempt by the land board to demarcateplots beyond the ring road was stopped by variousfield owners, who denied ever being consulted withregard to the land board’s intention to repossesstheir fields. Some also queried the inadequateamounts of compensation offered by the landboard85. Consequently, dissatisfied field owners(including ward headmen) subdivided their fields into40x40m plots, which they sold to people wantingto build houses. Although they subdivided withoutconsulting the land board, field owners expected thatthe land board would later formalise the allocations,as was the case with plots allocated by wardheadmen86. These allocations became the subjectof the 1991 Presidential Commission of Inquiry intoland problems in Mogoditshane and other peri-urbansettlements.

The Commission’s findings confirmed, inter alia,that arable land had indeed been subdivided andconverted to residential use without the authority ofthe Land Board. Residential plots were being ‘sold’for at least P8,000 per 40 x 40m plot. The paymentswere called madi-a-selepe, meaning ‘thank you fee’.Some sellers issued fictitious ‘certificates ofcustomary land grants’ stamped with fictitious namessuch as ‘Mogoditshane Land Board’. Buyers wereasked to develop the plots as soon as possible.Cabinet Ministers and high-ranking governmentofficials were among the people who ‘illegally’acquired land in Mogoditshane.

The Commission cited the failure of state institutionsto deliver a sufficient number of serviced plots; easyaccess to tribal land; the high price of serviced landin Gaborone; and the belief that land boards had no

authority over land allocated before 1968 when theTribal Land Act was enacted as the main factorsthat contributed to the development of informal landmarkets in Mogoditshane and other peri-urbanvillages87.

Following the Commission’s report, the Governmentdecided that all those citizens who had “illegally”acquired and developed land in Mogoditshane andother peri-urban areas should pay a penalty fee ofP5,000 per 40 x 40m plot and a proportionateamount for larger plots in order to have their holdingsregularised by the land board88. One of the Ministersimplicated in the report was forced to resign. Inaddition, a subordinate land board for Mogoditshanewas established.

According to the report of a Technical Team set upby the responsible Ministry eight years later inJanuary 2000, about half of 1,558 people who had‘illegally’ acquired and developed land inMogoditshane failed to pay the P5,000 penalty. Inthe meantime, more arable land had been ‘illegally’subdivided for housing purposes. Furthermore, anumber of people who had not previously beenidentified as so-called ‘squatters’ paid the P5,000penalty fee in order to have their rights regularised.Some had paid between P25,000 and P43,000indicating attempts to regularise large tracts of land89.

Besides its proximity to Gaborone, four other majorfactors appear to have fuelled informal land deliveryin Mogoditshane.

First, until the establishment of the MogoditshaneSub-Land Board in the early 1990s, there was apower vacuum created by the extinction of chiefs’supervisory role in land matters and the laxity ofland boards located tens of kilometres away.Consequently, unsupervised headmen -motivated by the ‘thank you fees’ - assumedpowers to themselves, rezoned arable land andallocated residential plots, allocations which theland board confirmed.

28

g A B O R O N EAB

OR

ON

E

Second, the land board undermined its ownauthority when it condoned allocations made byheadmen during the 1970s and early 1980s. Theland board made yet another procedural errorwhen it directed headmen and land applicants toidentify vacant plots within Old Mogoditshane.Headmen and land seekers may have beenconfused by these rules and procedures ortempted to utilise the loophole in the landallocation process.

Third, due to poor record keeping and failure toissue certificates, there were confusions as to whohad or had not been legally allocated land or hadtheir plots regularised by the land board90.

Fourth, as noted earlier, holders of arable landrights are generally dissatisfied with the amountof compensation paid for their loss of use rights.As a result, rights holders thought it wise tosubdivide their holdings and sell the plots orotherwise transfer their rights before land boardofficials arrived with their meagre compensationpayments.

Informal land acquisitions in MogoditshaneAccording to the 1991 census, three quarters of allowners in Mogoditshane reported that they had beenallocated their plots by the chief or land board, 13percent had inherited their plots, 4 percent hadpurchased land and only 7 percent had allocatedland to themselves. While the 1991 PresidentialCommission identified 841 plots that had beeninformally acquired, a follow-up Task Forceidentified a total of 1,558 ‘illegally’ acquired plots inthe same area – 64 percent of all plots inMogoditshane. The Commission’s report revealedthat formal land delivery systems had completelybroken down in the area. Those who allocated andsold land wanted cash payments, while the buyerswished to build houses. Security of tenure was‘guaranteed’ by written and witnessed sale/transferagreements and through immediate fencing and/orconstruction of houses, as well as the prospects ofregularisation by the land board on presentation of

‘no objection’ papers signed by headmen. Buyerswere urged to build as quickly as possible – even atnight91.

The proportion of owners in a sample survey carriedout by Malibala in 1998 who reported that they hadbeen allocated their plots by the chiefs or land boardwas 32 percent. 16 percent had inherited and 27percent purchased their plots, while 26 percent hadself-allocated their plots or been given them forfree92. Although Malibala included those ‘allocatedfor free’ under self-allocation, the proportion of 26percent is high compared to the 1991 census results.

Contrary to Malibala’s findings, our householdsurveys and focus group discussions did not produceevidence to illustrate widespread informal landsupply systems in Greater Gaborone. Only fivepercent, four percent and eight percent of householdswe interviewed, respectively, in Old Mogoditshane,New Mogoditshane and Old Naledi said they hadallocated plots to themselves. The low rate of self-allocation in Old Naledi is surprising since we knowfor a fact that the people who initially settled in OldNaledi were squatters. We also expected NewMogoditshane to have a higher proportion of self-allocatees than Old Mogoditshane. A number ofreasons may account for these discrepancies.

First, it appears that once self-allocations havebeen regularised, land rights holders considerthemselves to have been legally and duly allocatedland by the responsible authority - SHHA or theland board. As a result, respondents who mayhave initially bought undeveloped land, self-allocated land or used other ‘illegal’ methods toacquire land have since ‘regularised’ orformalised their acquisitions with the land boardand, therefore, consider the land board to haveallocated them the land in question.Second, due to a reluctance to admit wrongdoingin a country where the culture detestsdisobedience to authority or a fear of retribution,some ‘squatters’ may have hidden their ‘true’method of land acquisition. Our findings are,however, consistent with the 1991 census data.

29

g A B O R O N E

gA

BO

RO

N

Third, a thousand or so properties belonging to‘illegal’ land occupiers in New Mogoditshanewere demolished between 2000 and 2002, whichmeans that the majority of remaining houses hadbeen erected on legally acquired land.

Thus our field surveys show that, as expected,allocation by traditional authorities was moredominant in Old than New Mogoditshane while‘allocations’ by the land board were more commonin New than Old Mogoditshane (see table below).Acquisition of property through purchase was similar

Old Mogoditshane

New Mogoditshane

Method of property acquisition by owner households

Male head

(n=56) %

Female head

(n=42) %

All

(n=98) %

Male head

(n=58) %

Female head

(n=34) %

All (n=92)

%

Purchased 5.4 16.7 10.2 17.2 0.0 10.9 Allocated by traditional chief / headman

35.7 28.6 32.7 12.1 11.8 12.0

Allocated by party functionary 0.0 0.0 0.0 8.6 5.9 7.6 Allocated by public sector body (land board)

26.8 16.7 22.4 39.7 64.7

48.9

Self-allocated 3.6 7.1 5.1 3.4 5.9 4.3 Inherited 25.0 23.8 24.5 15.5 11.8 14.1 Gift 3.6 7.1 5.1 3.4 0.0 2.2 Total 100.0 100.0 100.0 100.0 100.0 100.0

at about ten percent of plots in both parts of thearea but substantially lower than in Malibala’s study.The proportion of self-allocations was also lower inour study (4-5 percent) than in Malibala’s study (26percent). The differences may be due to thedemolitions that recently took place in Mogoditshaneor may reflect fear on the part of plot holders toadmit purchase or self-allocation. As expected,acquisition of land through inheritance is very high inthe older part of the area (25 percent) and moremodest (14 percent) in the more recently developedpart.

About 41 percent of all households interviewed inMogoditshane – 43 percent and 37 percent in Oldand New Mogoditshane respectively - were headedby women. In the older area, a quarter of thehouseholds had inherited their plots, compared to14 percent in the newer area. In both areas almostas many women household heads had inherited plotsas men. It seems that inheritance is generally definedto include the transfer (excluding sale) of land torelatives before one dies. Inheritance is also used tojustify transfer of land without land boardauthorisation. For example, in the matter betweenKgatleng Land Board vs Bontsi Chelane and 13others, 14 people claimed to have inherited landfrom one woman who had subdivided her fields. It

is doubtful whether the 14 individuals were relatedto the said woman by blood, adoption or marriage.Just under two thirds of all the heads of householdsinterviewed in both areas belong to the indigenoustribe – Bakwena. As expected, a larger proportion(about 50 percent) of the heads of households inOld Mogoditshane were born in the village comparedto 44 percent in New Mogoditshane.

Excluding inheritance and the less common methodsof acquisition, it is clear that the majority of plots inOld Mogoditshane had been allocated by the chiefor headmen and an even larger majority in NewMogoditshane by the land board. However, evenin Old Mogoditshane, the land board had played an

Source: Field surveys, 2002/03

Methods of land acquisition in Mogoditshane, 2002/03

30

important role and in New Mogoditshane someowners had been allocated plots by the chief andheadmen. The traditional authorities appear toslightly favour men, but not markedly so, an apparentdeparture from custom. More male plot owners hadbeen allocated their plots by the land board in OldMogoditshane but a much higher proportion ofwomen in New Mogoditshane, demonstratingoverall practice that slightly favours women.

It was not possible to collect detailed informationon the income and expenditure of surveyedhouseholds, but in order to assess whether low-income households can obtain access to land in thearea, information on work status, householdpossessions and consumption of food, all quite goodproxies for estimating the wealth status ofhouseholds, was collected. Over 70 percent of theheads of households who had acquired plots in eacharea had done so while they were working93. Ofthese 80 percent or more were in full-time wageemployment, mainly with public sector bodies orlarge-scale private firms. However, fewer female thanmale heads were working (67 percent) and a higherproportion of male heads in New than in OldMogoditshane (86 percent compared to 73percent). Moreover, fewer women were inprofessional/technical occupations.

Today, about 39 percent and 45 percent of therespondents in Old and New Mogoditshanerespectively have at least one car, pick-up or van.Three quarters have gas stoves and only a quarterused coal braziers/woodstoves for cooking. Abouta third of the respondents interviewed eat meat,chicken or fish everyday; and only 2 percent in NewMogoditshane and 7 percent in Old Mogoditshanecan only afford one meal a day. Although thosewho were allocated land through the customarysystem may have included relatively low wageemployees and low-income households, most ofthose who have acquired plots in Mogoditshanemore recently have not been poor. Today, nonewho own urban land and a house can be categorised

as poor, while the vehicle ownership rates and foodconsumption patterns indicate that around a third ofthe owners are relatively wealthy.

At the time that they occupied their plots, only 58percent of the households in Old Mogoditshane hadaccess to proper toilets while 42 percent used thebush for excreta disposal. This was very unhygienicconsidering that 29 percent of the householdsobtained water from wells, streams and ponds. InNew Mogoditshane, only 22 percent used the bushat the time they first occupied the plots, 59 percenthad their own toilets and 20 percent shared toiletfacilities with neighbours. Over the years, pipedwater has gradually been extended, enabling somehouseholds to upgrade their water supply andmethod of sewage disposal, while many others haveinvested in on-site facilities.

At present, most households have access to propersanitation facilities and drinking water. In OldMogoditshane, 84 percent have their own pit latrines,14 percent have flush toilets and the remaining onepercent share toilet facilities with their neighbours.The situation is similar to that in New Mogoditshane,where 75 percent have their own pit latrines, 18percent have flush toilets and 6 percent sharefacilities with their neighbours. Unlike when they firstoccupied the plots, no household uses the bush forexcreta disposal. Between 55 and 60 percent ofthe respondents in Old and New Mogoditshane havetheir own taps while about a third share a tap withneighbouring plots. However, about ten percent ofthe respondents still obtain drinking water from wells,streams and ponds in each part of the area.

g A B O R O N EAB

OR

ON

E

31

AssessmentStrengths and weaknesses of the land delivery systems in Greater Gaborone

With the exception of Old Naledi, there has beenno widespread informal housing development withinGaborone Township. Within the urban boundaries,the government has quickly demolished the few illegalhouses that have sprung up. While the urban areanow extends beyond the city boundaries, the landdelivery system in the areas concerned has bothformal and informal characteristics. In effect, theGovernment of Botswana has adopted a ‘carrot andstick’ approach in the management andadministration of land in Greater Gaborone andother settlements.

State-led land delivery systems

Land delivery systems in Greater Gaborone (andBotswana in general) are strongly dominated, intheory and practice, by state structures andinstitutions. The government owns and controls about95 percent of the land within Gaborone Townshipand, through land boards, indirectly administers andcontrols all the tribal land outside the townshipboundaries. It has achieved this control throughcontinual purchase of freehold land on the one handand the Tribal Land Act on the other.

Having assumed the role of urban land supplier, thestate further undertook to supply subsidised, cheapor free-of-cost land to all citizens who reside or wantto reside in towns such as Gaborone. This strategy,however, has led to ever increasing ‘demand’ forland. This has, to some extent, been for speculativepurposes, whereby individuals allocated land at nocost or cheaply have not developed it but have soldit at market prices, sometimes years later. As a result,state and tribal land allocated to some individualsremains vacant or undeveloped for many years.

Worse still, most land in Greater Gaborone – as inother townships - is neither made available to thosewho have immediate need for it nor to those whohave the means to pay for it and/or develop itexpeditiously but rather to those who are eligible.To be eligible an applicant must belong to a category

for which plots are available (e.g. first-time-applicant,low-income earner, and resident for a number ofyears). The eligibility criteria have somehow beenconstrued to mean ‘entitlement’ criteria. Thus someindividuals may be applying under various categoriesbecause they meet the requisite criteria, rather thanbecause they need the land or can develop it. The‘entitlement’ mentality is so entrenched that someparticipants in the Focus Group Discussionsconducted during this research expressed a desireto inherit their relatives’ positions on waiting lists.

Land for women

As observed earlier, the data collected in this studyshow that state agencies in Greater Gaborone donot have open policies or practices that discriminateagainst the allocation of land to women. All theparticipants in focus group discussions concurredon this, as summarised by one woman who said that,“There is no gender bias in land allocation. Thebiasness is found between the rich and the poor.The rich get plots because they manage to paybribery”.

“We do not know of any specific issue” regardingdiscrimination against women, said the youths in OldNaledi. However, there

“is gender biasness in land ownership, tothe extent that our mothers here haveregistered their own plots in the names oftheir partners because they [malepartners] were earning income. If youinquire from plot holders here on how theygot their plot you find that most of the plotshere in reality belong to women butbecause of the respect of men they areregistered under men’s names”.

Plots are often registered in men’s names because itis assumed by officials that they are the householdheads, whether they are cohabiting or married.

32

Land for young people

Young people feel disadvantaged by the presentwaiting list procedures for accessing land. First,youths feel “discouraged to apply for landbecause those who applied many years ago arestill waiting”. “… if I apply now for a plot I willwait for about 5 or 12 years without any replyand if you ask you are told to go and wait asthey will write you a letter”, said one respondentfrom New Mogoditshane.

Second, youths said that they are further discouragedto apply for land because “we are not employedso we think of how we will develop those plotsin case we get them”, in the words of a youngwoman in Old Naledi.

Third, youths in Mogoditshane feel disadvantagedby the present system whereby the land boardrepossesses agricultural land, converts it toresidential uses and allocates demarcated plots toapplicants from all over the country. To them, theprocess denies them a right to their ancestral land.

Fourth, young people are against the practice ofselling and buying land. They decry parents who sellland because when their children grow up they willhave no land to inherit.

Furthermore, because they themselves are notemployed, they are unable to buy land. Instead, theywant to inherit their parents’ positions on the waitinglists and to be given preferential treatment. “... weapply for land as youth and we are made to waitin a queue with people who already have plots.It could be very helpful if the applications of theyouth could be given a special treatment” arguedone young man.

Land for the poorThe formal land delivery system in Gaborone hasgenerally ignored the land requirements of the poor– especially during the 1960s and early 1970s.During this period informal delivery systems enabledthe poor to access land at Old Naledi, which wasfree and within walking distance of the city centre,industries and other workplaces. The self-helphousing programmes undertaken between 1975 and1990 benefited the middle and rich poor, since theyexcluded people without verifiable regular incomes.With effect from 1990, when allocation of free self-help housing plots came to an end, only the rich poormay have benefited from formal state land deliverymechanisms. The poor are likely to be furtherexcluded by the policies and programmes adoptedin the 1999 National Housing Policy (see p.15-16).

Formal land delivery processes in tribal areas mayhave benefited the poor until the mid-1980s wheninformal land delivery systems started to mushroom.According to focus group informants, it has becomedifficult to obtain land board allocated plots inMogoditshane, Tlokweng and other peri-urbanareas due to long waiting periods and bribery. Claimsof bribery, nepotism and corruption among landboard officials were aired in all the six focus groupdiscussions

According to focus group participants, “You haveto know someone at the land board to get land.At times when people go to apply, they will betold to come with /at five (tla ka 5) and thosewho know will come with P5000.00 for bribery.While those who do not know will come at 50’clock and find offices closed,” they claimed.According to youths, land is no longer as easilyaccessible as it was in the past. “Land used to beallocated by chiefs in their own way, and whenland board took over, the old ways of allocatingland were abandoned and the land boardregulations are not working well. The land boardregulations should have been built on the onesthat were used by chiefs”, they said.

g A B O R O N EAB

OR

ON

E

33

Interpreting Botswana’s successes

Despite the imperfections noted above, the state-led land delivery systems have been able to createand maintain relatively well-planned and servicedresidential suburbs in Gaborone, as well as relativelyorderly land delivery systems. This study hasconfirmed earlier findings that, unlike other cities insub-Saharan Africa, perhaps three quarters of allowners of residential plots have legitimate and/orlegal land rights over their plots. Although the studyareas of Old Naledi and Mogoditshane wereoriginally unplanned settlements, the government hasprovided room for land rights holders to formalisetheir rights and punished those who were reluctantto do so.

Maipose attributes Botswana’s successful ‘state-leddevelopment’, and remarkable and sustainedeconomic growth to ‘good luck’, ‘good governance’and ‘good management’ –which are inter-related butindependently constituted. To Maipose, the countryis lucky to be endowed with rich mineral depositsand to have discovered these deposits after, ratherthan before independence. Otherwise the mineralrevenues would have benefited the colonisinggovernment. Botswana is further lucky to haveavoided widespread European settlement and landdispossessions94.

Furthermore, the country is lucky to be composedof a “small and largely homogenous populationat least in the sense that ‘tribalism’ and‘sectionalism’ are not sharply used to explainelectoral behaviour and leadership competitionin Botswana”95. To Good, Botswana’s trackrecord of good governance owes its origin tocontinuity in the political leadership throughout thethree political episodes: from pre-colonial times,during the colonial era and to the present. Unlikemany countries in the region, Botswana’spostcolonial leadership was experienced, respectedand trusted as an extension of the familiar traditionalchiefs’ administration96.

Botswana’s government has managed the country’smineral resources and economy exceptionally well,which in turn has enabled the country to attract aconsistent flow of foreign aid and direct investment.According to Good, the leadership is committed togood management of state/public resources becauseit has a considerable stake in the economy. Goodmanagement of public sector resources enables theleadership to remain in power and to attract foreigninvestment into the country, which results in windfallgains/profits to the private sector.

Wealth and good governance has been translatedinto considerable government administrative capacityand a consistent policy of redistribution from growththat has benefited most of Botswana’s citizens. Thishas been demonstrated in the case of land deliveryby a considerable capacity to purchase, subdivideand service land with significant levels of subsidy, asdescribed above. Despite a variety of problematicaspects, this approach has been quite consistent andhas pre-empted the widespread emergence ofinformal settlements within the city boundary.

g A B O R O N E

gA

BO

RO

N

34

PolicyPolicy implications and recommendations

While the government has done remarkably well withdelivery of state land for urban development, thissummary has highlighted areas that requireimprovements. The major issues of concern includethe conceptions of land ownership, rights of availand entitlement, and sale of land or land rights.Although some of these relate to state land, mostrelate to tribal land.

Ownership, sale and leasing ofstate landContemporary rules and definitions of ‘ownership’of state land are both confusing and ambiguous. Inthe first instance, the concept of “selling” or“purchasing” of state land is a misnomer becausewhile “selling” implies transfer of ownership, sale ofstate land in Botswana does not lead to transfer ofland ownership from the state to buyers. State landis leased, rather than sold, to individuals. WhileCertificate of Rights (COR) beneficiaries are grantedusufruct rights, Fixed Period State Grant (FPSG)beneficiaries are holders of capitalised leases. Webelieve beneficiaries have a moral right to know thatthey are leasing rather than buying land.

It should be made clear to recipients of stateland that they are leaseholders rather thanoutright owners

Furthermore, the so-called ‘prices’ for state landare only costs for servicing land rather than ‘lumpsum rental payments’ as is widely claimed97. As notedabove, the cost of land has never been included whencomputing the ‘prices’ for state land plots. Costsincurred by the government when buying freeholdland have never been passed on to beneficiaries. Inpractical terms, state land is granted ‘free of costs’to all beneficiaries. We believe the time is ripe forthe state to consider regular collection of rentalcharges (perhaps every two, three or five years)

instead of ‘lump sum collection’, which makes landunaffordable to the poor. Acknowledging that stateland is being leased will facilitate repossession ofplots that remain undeveloped for long periods anddiscourage people who think that they are buyingland to own in perpetuity.

Cost recovery for the capital costs ofinfrastructure provided to plots allocated bygovernment should be spread out over aperiod of time, to avoid the anti-poor biascaused by insisting on up-front payment

Understanding and acknowledging the fact that CORand FPSG title owners are leaseholders helps toclarify the fact that only the government can sell stateland. As argued elsewhere, COR and FPSGleaseholders may only sell, transfer, cede etc theirrights and interests in land for the remainder of theirlease periods98. It is, therefore, morally, politicallyand economically justifiable for leaseholders to sellor transfer their land rights and interests as and whenthey wish. This appreciation will go a long inpromoting formal land markets in Greater Gaboroneand other urban areas – which the government hasbeen trying to promote since 1982 (p.15-17).

Leaseholders should be free to sell theiroutstanding rights and interests in land, in theinterest of promoting more efficient urban landmarkets

Codifying tribal/customary landtenure

As noted above, the Tribal Land Act (in its presentform) deals with administrative aspects of customaryland tenure, leaving the rest to personal interpretationand misinterpretation. The term ‘tribal’ is also

misleading since such land belongs to all citizens ofBotswana. In light of these omissions and differinginterpretations, it is recommended that studies andconsultations should be undertaken to identify issues,elements and aspects of customary land tenure thatneed to be codified. The codified land tenure facilityshould state whether land boards are owners,trustees or administrators of tribal land, andthereafter, define their roles and responsibilitiesaccordingly. It should also state whether the‘trusteeship’ ends with allocation or is forever.

Codification of customary land tenure isdesirable and should include clarification ofthe role of land boards

As trustees, the powers of land boards (and theirpredecessors) should be inferior to those of the tribeor jural community to which the land belongs, butsuperior to those of individuals within the tribe.Although in the past the chief could overrule decisionsof individual families or wards, he could not overrulethe wishes of the tribe. At present land boardsconsider themselves de facto ‘owners’ of tribal landunder their jurisdiction. They claim that they ‘ownthe land’ and only allocate usufruct rights to landoccupiers. As trustees, land boards should be guidedby policies and wishes made by the respectivecommunities - each land board should be answerableto local residents through kgotlas, district councilsor other forums.

Land boards should be regarded as trusteesrather than owners of land. They should beguided by and accountable to the relevantlocal community.

The codified land tenure system should be flexible,providing for local inputs and variations. It shouldalso define the extent to which land occupation andutilisation indicate private and/or exclusive ownershipof rights over land or provide for extended rights(e.g. collection of wild fruits, honey etc on TribalGrazing Land Policy ranches), where suchprovisions would not compromise superior landrights.

Rights of avail and entitlement

The government should reconsider its commitmentto ensure every citizen gets a free or subsidisedresidential plot in several townships. Thecommitment raises expectations that the governmenthas increasingly found hard to fulfil – the harder ittries the more applications it receives. To curtail the‘culture of entitlement’ among middle and high-income groups, we support the Land Policy Reviewproposal whereby each citizen will be entitled to asingle lifetime grant of a residential plot on tribal land.If people wish to acquire additional plots, they shouldpurchase them through the market. The success ofsuch a policy will depend on a complete and up-to-date records system.

Each citizen should be entitled to a singlelifetime grant of a residential plot on tribal land.Additional plots should be purchased throughthe market.

Demolition of ‘illegal’ structures

In view of the high quality of residential andcommercial structures ‘illegally’ erected in someareas, the negative social and economic effects ofwholesale demolitions, and the high costs ofdemolishing permanent buildings, we recommendthat such buildings become state property, whichthe state may convert to appropriate uses or auction

g A B O R O N E

gA

BO

RO

N

35

off to the highest bidders. Demolition of buildingsshould only be carried out in the public interest ratherthan as retribution for infringements of regulations.There are many better ways through which the statecan exercise its authority over law-breakers.

Structures erected without permission shouldbecome state property rather than beingdemolished and alternative penalties devisedfor those who do not comply with regulatoryrequirements

Compensation payable for arableland in tribal areas

The compensation payable to indigenous land rightsholders when their land is repossessed by tribal landboards for subdivision and allocation should be seenas a fair recompense for the loss of use rights, orlandholders will continue to resist and evade officialprocedures for the conversion of rural tribal landfor urban use. Two possible solutions should beinvestigated

a) Dispossessed masimo owners should beallocated two to three residential or commercialplots in lieu of the loss of their user rights.

b) Compensation should be calculated in terms of‘lost lifetime opportunities for growing crops,grazing livestock, etc on the said piece of land’.‘Lifetime’ could be set at 99 years or 999 yearsif perpetuity is the main contention. A farm withan annual crop yield valued at P100 would havea present lifetime value of P9,900.

These proposals take into account the common viewof land as an asset that should be bequeathed fromone generation to another in a family.

The basis and arrangements for payingcompensation to indigenous land rightsholders whose land is taken by a tribal landboard for re-allocation should bereconsidered, to reduce resistance to andevasion of the official process for deliveringland held under customary tenure for urbanuses

Other issues

Other pertinent issues that require re-examinationinclude

the powers of the Minister in terms of the Townand Country Planning Actensuring a supply of smaller plots (about 200 m2

each) for the very poor, andcreation of land delivery partnerships betweenland boards, local communities, the private sector(including the Botswana Housing Corporation)and District Councils for the servicing and deliveryof plots for residential and other uses.

These issues were recently covered by the landpolicy review report and are still under considerationby the government, so are not discussed in detailhere99.

g A B O R O N EAB

OR

ON

E

36

ConclusionBoth in the colonial and pre-colonial eras, thegovernment in Botswana has attempted to meetgrowing demand for urban residential land throughthe administrative allocation of publicly owned land.Because Botswana has, since independence,experienced rapid economic growth and achievedpolitical stability, the government has had substantialresources and considerable administrative capacityto subdivide and service relatively large areas ofland, purchasing freehold farms on the urbanperiphery when necessary to expand its stock ofland. Its ability to make large numbers of servicedplots available for middle and lower-income urbanhouseholds has enabled it to prevent informalsettlement within the urban administrativeboundaries, with the exception of one settlement.Tolerated in the early days of Gaborone’s growthbecause it accommodated what was then seen as atemporary construction labour force, Old Naledi hasgradually been regularised and upgraded. Today itcontinues to perform an important function in theurban housing market, providing a large number oflow cost rental dwellings in a central location, althoughits continued role may be threatened by currentupgrading proposals.

The three main contemporary problems with thegovernment’s strategy of meeting the urban demandfor middle and low-cost housing through takingresponsibility for land delivery are as follows:

i. Demand, as expressed in waiting lists for plots,is constantly outstripping both supply andpopulation growth, fuelled by subsidies to the costof land and eligibility rules. As a result, manywho have not already received a plot have towait for a long period to obtain one.

ii. Government’s ambition to successively improvethe planning, infrastructure and constructionstandards it adopts and the new financingarrangements it is encouraging have, in recentyears, placed publicly provided residential plotsbeyond the reach of lower income households.

iii. Many individuals have been able to obtain twoor more plots through this system. Many haveneither the need nor the resources to develop(or even complete payments for) these plots,resulting in large numbers of undeveloped plots.As a result, not only is government capital tiedup unnecessarily but also those in need of a plotfor their own occupation have to wait for anunduly long time before they are allocated one.

For these and other reasons, land has increasinglybeen supplied for urban residential use in the ‘tribal’land area just outside the city boundary, where it isformally under the control of a land board. This wasset up by government to take over the responsibilityof traditional authorities for allocating land. Theresearch revealed that, although much of theagricultural land subdivided for residential use isformally repossessed and re-allocated by the board,there are a number of problems with this process.Rights holders belonging to indigenous groups havedifficulty reconciling their understanding of the rightsand responsibilities of actors in the customary tenuresystem with the rights and responsibilities of the landboard. In addition, they are also led by the perceivedunfairness and inefficiency of official procedures tocircumvent them when they can. It is difficult toassess the exact extent of informality, first, becauseindigenous rights holders recognise that some of whatthey do does not comply with official requirementsand disguise their actions. In addition, governmentattempts at enforcement (including demolitions)encourage actors to conceal informal practices.Moreover, periodically the tenure of existing rightsholders has been regularised, thereby rendering thepreviously informal formal. Basically, the Botswanagovernment has put in place a workable system fordelivering large volumes of rural land held undercustomary tenure for planned urban residential use.Nevertheless, there are a number of legal andpractical problems with the system, some of themquite significant. This research has suggested anumber of ways in which these could be tackled.

37

Footnotes1 Berry, Sara S. (2001) Chiefs Know Their

Boundaries. Heinemann, James Currey andDavid Philip, Portsmouth, Oxford and CapeTown

2 Schapera, I. (1994) A Handbook of TswanaLaw and Custom. International African Institute,London and The Botswana Society, Gaborone,p. 30; Helle-Valle, Jo (2002) “Seen from below:conceptions of the state in a Botswana village”Africa, 72(2) pp. 179- 202.

3 Schapera, 1994, op.cit. p. 84.4 Dow, Unity and Kidd, Puseletso (1994) Women,

Marriage and Inheritance. Women and Law inSouthern Africa, Gaborone, p. 1

5 Dow and Kidd, 1994, ibid, p. 28.6 Kalabamu, F. T. (2001) “Housing delivery

systems in Botswana: The inadequacy of genderneutral policies” in Rwomire, A. T. (Ed.) AfricanWomen and Children: Crisis and Response.Praeger, Westport, pp 205 – 224; Larsson, Anita(1989) “Traditional versus modern housing inBotswana: An analysis from the user’sperspective” in Bourdier, Jean-Paul and Alsayyad,Nezar (eds) Dwellings, Settlement andTradition: Cultural Perspectives. UniversityPress of America, Lanham, pp. 503-525;Larsson, Anita (1990) Modern Houses forModern Life. University of Lund, Lund.

7 Kalabamu, F. T. (1993) “An appraisal of theevolution of Tswana settlement” PlanningHistory, 15(2) pp. 12-18; Schapera, 1994, op.cit.

8 Schapera, 1994, op. cit. p.19.9 Schapera, 1994, op. cit, p.23.10 Hardie, G.J. (1982) “The dynamics of the internal

organization of the traditional tribal capital,Mochudi” in Hitchcock, R. R. and Smith, M. R.(Eds.) Settlement in Botswana. BotswanaSociety and Heinemann, Gaborone, pp. 205-219.

11 Schapera, I. (1943) Native Land Tenure in theBechuanaland Protectorate. Lovedale, CapeTown; Jeppe, W.J.O. (1980) Bophuthatswana:Land Tenure and Development. Maskew Miller,Cape Town; Kalabamu, F. T. (2000) “Land tenure

and management reforms in East and SouthernAfrica – the case of Botswana” Land Use Policy,17, pp. 305-319

12 Schapera, 1994, op.cit. p. 195-19613 Kalabamu, 2000, op. cit. p.30614 Colclough, Christopher and McCarthy, Stephen

(1980) The Political Economy of Botswana:A Study of Growth and Distribution. OxfordUniversity Press, Oxford, p.14.

15 Ng’ong’ola, Clement (1997) “Land rights formarginalized ethnic groups in Botswana, withspecial reference to the Basarwa” Journal ofAfrican Law, 41, pp. 1-26

16 Colclough and McCarthy, 1980, op.cit.17 Colclough and McCarthy, 1980, op.cit, p. 1418 Dickson, W.L. (1990) Land Tenure and

Management in a Developing Country. NIR,University of Botswana, Working Paper No. 54.

19 Colclough and McCarthy, 1980, op.cit. p.22;Good, Kenneth (1992) “Interpreting theexceptionality of Botswana” The Journal ofModern African Studies. 30(1) pp.69-95.

20 Larsson, Anita (1999) “Gender system andgender contract: the case of Botswana” inGoodman, S. and Mulinari, D. (Eds.) FeministInterventions in Discourses on Gender andDevelopment. Lund University, Lund, p. 73-4.

21 Good, 1992, op.cit, p. 7222 Mathuba, (1982) Customary and Modern Land

Tenure Systems in Botswana. Unpublished MScDissertation, University of Wisconsin, Madison,USA

23 Adams, M., Kalabamu, F.T., and White, R.(2003) “Land tenure and practice in Botswana:governance lessons for Southern Africa” AustrianJournal Of Development Studies, XIX(1) pp.55-74.

24 Dickson, 1990, op.cit. p. 2625 Government of Botswana (1970) National

Development Plan: 1970-75. GovernmentPrinter, Gaborone, p. 31

26 Government of Botswana (1983a) Report of thePresidential Commission on Land Tenure.Government Printer, Gaborone, p. 2 and 5

38

27 Mathuba, B.M. (1989) Report on the Review ofthe Tribal Land Act, Land Policies and RelatedIssues. Government Printer, Gaborone, p.2-3.

28 Ng’ong’ola, 1997, op.cit. p.1429 Government of Botswana (1992a) Report of the

Presidential Commission of Inquiry into LandProblems in Mogoditshane and other Peri-Urban Villages. Government Printer, Gaborone,p. 91.

30 Pula is Botswana’s currency. In 1995, 1 US$ =4 Pula. At present 1US$ = 5 Pula.

31 Government of Botswana, (2002) 2001 Ministryof Finance and Development Planning, Gaborone,p. 17-20.

32 Government of Botswana (1997) NationalDevelopment Plan 8: 1997/98 –2002/03.Population and Housing Census: Populationof Towns, Villages and Associated Localities.Government Printer, Gaborone, p.9

33 Colclough and McCarthy, 1980, op.cit. p.105-107

34 Good, Kenneth (2002) “Rethinking non-accountability and corruption in Botswana”Africa Insight, 32(3) pp. 11-18.

35 Good, 1992, op.cit. p. 7336 Good, 1992, op.cit. p.7337 Jefferis, Keith (1997) “Poverty in Botswana” in

Salkin, J.S. et al., (Eds.) Aspects of theBotswana Economy. Lentswe la Lesedi,Gaborone, p. 484.

38 Colclough and McCarthy, 1980, op.cit. p. 23;and Good, Kenneth (1999) “The state andextreme poverty in Botswana: the San anddestitutes” The Journal of Modern AfricanStudies. 37(2) pp.185-205.

39 Jefferis, 1997, op.cit. p. 48640 For census purpose, an urban area is defined as

a settlement with a minimum population of 5,000inhabitants and where at least 75 percent of thelabour force are employed in non-agriculturalactivities.

41 Silitshena, R.M.K. (1984) “Urbanisation inBotswana” Nordic Geographical Journal, 38,pp. 109-128

42 NRS (Natural Resource Services) (2003) Reviewof Botswana National Land Policy. Departmentof Lands. (2 volumes), p.96. Not yet published.

43 Government of Botswana (1983b) AnEvaluation of the Self-help Housing Agencies.Ministry of Local Government and Lands,Gaborone, p.28.

44 GOB, 1983b, ibid, p. 2845 Ministry of Local Government, Lands and

Housing (1992a) Review of the Self-helpHousing Agency. Unpublished, p.23.

46 PADCO (Planning and DevelopmentCollaborative International) (1989) Impact ofDesign Standards on Cost of UrbanDevelopment in Botswana. Ministry of LocalGovernment and Lands, Gaborone, p.3.

47 Ministry of Lands and Housing (2001) Reviewof the Accelerated Land Serving Programme.Unpublished, p.10.

48 Datta, Kavita and Jones, Gareth A. (2001)“Housing and finance in developing countries:invisible issues on research and policy agendas”Habitat International, 25, pp. 333-35.

49 Government of Botswana (2000) NationalPolicy on Housing in Botswana. GovernmentWhite Paper No. 2 of 2000. Government Printer,Gaborone, p. viii.

50 Gwebu, Thando D (2003) “Environmentalproblems among low income urban residents: anempirical analysis of Old Naledi – Gaborone,Botswana” Habitat International, 27, pp. 407-427

51 Gwebu, 2003, ibid, p. 42552 GOB, 2000, op. cit.; and Government of

Botswana (2003) National Development Plan9: 2003/04 –2008/09. Ministry of Finance andDevelopment Planning, Gaborone, p.333.

53 MLH (Ministry of Lands and Housing) (2000)Mogoditshane Land Problems Report.Unpublished memo, p.4.

54 NRS, 2003, op.cit. p. 13255 Government of Botswana (2004) Report of the

Judicial Commission of Inquiry into State LandAllocations in Gaborone, Government Printer,Gaborone.

g A B O R O N E

gA

BO

RO

N

39

56 Dickson, 1990, op.cit. p. 1057 Dickson, 1990, op.cit. p. 1158 GOB, 1983a, op.cit. p.3659 Dickson, 1990, op.cit. p.1360 DTRP (Department of Town and Regional

Planning) (1998) Gaborone City DevelopmentPlan. Final Draft. Unpublished, p.12

61 Dickson, 1990, op.cit. p.1362 GOB, 1983a, op.cit. p.3663 MLGLH (Ministry of Local Government, Lands

and Housing) (1997) Review of the NationalPolicy on Housing of Botswana. Draft Report,p.98.

64 MLGL (Ministry of Local Government andLands) (1990) An Analysis of ProvidingServiced Land and Housing in Botswana’sSeven Centres. Not published, p.21-6.

65 DTRP, 1997, op.cit.66 This figure includes about 1,600 plots serviced

under a squatter/settlement upgrading exercise inOld Naledi.

67 DTRP, 1997, op.cit. p.92-107.68 MLH, 2001, op.cit. p. 4069 MLGL, 1990, op.cit. p. 2570 MLH, 2001, op.cit.71 MLH, 2001, op.cit. p. 15-1672 MLGL (Ministry of Local Government and

Lands) (1978) A Baseline Survey of the OldNaledi Upgrading Project Area, not published,p.13.

73 DTRP, 1998, op.cit. p. 1174 van Nostrand, J (1982) Old Naledi: The Village

Becomes a Town. James Lorrimer & Company,Toronto; GOB, 1983a

75 van Nostrand, 1982, ibid.76 MLGL (Ministry of Local Government and

Lands) (1983) An Evaluation of the Self-helpHousing Agencies. Unpublished report, p.33

77 MLGL, 1979, op.cit.; van Nostrand, 1982,op.cit.; and Larsson 1986, op.cit.

78 GOB, 1983a, op.cit. p. 35; and van Nostrand,1982, op.cit.

79 Gwebu, 2003, op.cit. p. 41480 Haas Consult (2003) Upgrading of

Infrastructures in Old Naledi: ExecutiveSummary – Draft Detail Report. Reportsubmitted to the Gaborone City Council, p.42-43

81 Gwebu, 2003, op.cit. p.41782 KDA (Kweneng District Administration) (1982)

Mogoditshane: Gaborone’s Refuge orTraditional Village. Unpublished memo, p.22.

83 Sithole, T.R. (1995) An Evaluation of UrbanPlanning in Peri-Urban Settlements ofBotswana: A case study of Mogoditshane.Unpublished Masters of Urban and Rural Planningdissertation, Heriot-Watt University andEdinburgh College of Art, p.56.

84 Sithole, 1995, ibid, p.6185 Sithole, 1995, ibid, p.6286 Sithole, 1995, ibid, p.62-6387 GOB, 1992a.88 GOB (Government of Botswana) (1992b) Land

Problems in Mogoditshane and other Peri-Urban Villages. Government Paper No.1 of1992. Government Printer, Gaborone, p.24.

89 Coordination Unit, Mogoditshane Land Problems(2002) Briefing notes on Mogoditshane landproblems – squatter removal exercise-11.10.200. Unpublished memo, p.3

90 GOB, 1992a, p. 1291 GOB, 1992a, p. 14-1592 Malibala, Nakale Marro (1999) The Evolution

and Effects of Informal Land Markets on Peri-urban Environments: the case ofMogoditshane. Unpublished M.Sc. Thesis,University of Botswana, p.43.

93 Refers to income earning work and excludessubsistence farming.

94 Maipose, Gervase S. (2003) “Economicdevelopment and the role of the state inBotswana” DPMN Bulletin, X(2) pp. 19-25.

95 Maipose, 2003, ibid, p.2096 Good, 1992, op.cit.97 E.g. Dickson, 1990, op.cit.; Kalabamu, 2000,

op.cit.98 Kalabamu, 2000, op.cit. p. 31699 NRS, 2003, op.cit.

g A B O R O N EAB

OR

ON

E

40

Informal Land Delivery Processes in African Cities

Working papers

1. Rakodi, C. and Leduka, C. (2003) Informal Land Delivery Processes and Access to Land for thePoor in Six African Cities: Towards a Conceptual Framework.

2. Ikejiofor, C.U. (2004) Informal Land Delivery Processes and Access to Land for the Poor in Enugu,Nigeria

3. Kalabamu, F.T. and Morolong, S. (2004) Informal Land Delivery Processes and Access to Land forthe Poor in Greater Gaborone, Botswana

4. Musyoka, R. (2004) Informal Land Delivery Processes and Access to Land for the Poor in Eldoret,Kenya

5. Leduka, R.C. (2004) Informal Land Delivery Processes and Access to Land for the Poor in Maseru,Lesotho

6. Nkurunziza, E. (2004) Informal Land Delivery Processes and Access to Land for the Poor in Kampala,Uganda

7. Mulenga, L.C. and Rakodi, C. (2004) Informal Land Delivery Processes and Access to Land for thePoor in Lusaka, Zambia

Policy briefs

1. Ikejiofor, C.U. (2004) Informal Land Delivery in Enugu, Nigeria: Summary of Findings and PolicyImplications

2. Kalabamu, F.T. (2004) Land Delivery Processes in Greater Gaborone, Botswana: Constraints,Opportunities and Policy Implications

3. Musyoka, R. (2004) Informal Land Delivery in Eldoret, Kenya: Summary of Findings and PolicyImplications

4. Leduka, R.C. (2004) Informal Land Delivery Processes and Access to Land for the Poor in Maseru,Lesotho

5. Nkurunziza, E. (2004) Informal Land Delivery Processes in Kampala, Uganda: Summary of Findingsand Policy Implications

6. Rakodi, C. and Leduka, R.C. (2004) Informal Land Delivery Processes and Access to Land for thePoor: A Comparative Study of Six African Cities

Publications can be obtained by either telephoning Carol Fowler on 44 (0) 121 414 4986or Email: c.a.fowler@bham.ac.uk and also downloaded as a PDF file from http://www.idd.bham.ac.uk/research/researchprojs.htm

Publications

41