· D. Contracts and legality E. Landlord and tenant organizations IV. Public policy A. Rent control o 1. The introduction of rent controls o 2.

Page 1

Strategies for low-income shelter and services development:The rental-housing option

Table of Contents

Introduction

A. Contribution of rental accommodation to housing the poor

o 1. Now many tenants?

o 2. Variations between countries and between cities

o 3. Is the proportion of tenants increasing or decreasing through time?

B. Position of rental housing In national housing policy

I. The supply of rental housing

A. Public-sector housing for rent

o 1. Access to public-sector housing

o 2. Managing public-sector housing

B. The private rental sector

o 1. Who are the landlords?

o 2. How profitable is rental housing?

C. Low-income rental housing conditions

o 1. The public sector

o 2. The private sector

o 3. Housing conditions of tenants compared with those of owner-occupiers

II. The occupants of rental housing

A. Are tenants poorer than owners?

B. Other socio-economic characteristics of tenants

C. Methods of explaining tenure choice

o 1. Choice versus constraint

o 2. Is there a general preference for ownership?

o 3. Why is there a widespread desire for house ownership?

III. Landlord-tenant relations

A. The selection of tenants

B. Mobility of tenants, security of tenure and eviction

C. Conflicts between tenants and landlords

Page 2

D. Contracts and legality

E. Landlord and tenant organizations

IV. Public policy

A. Rent control

o 1. The introduction of rent controls

o 2. Which rents have been controlled?

o 3. Effects of rent controls

o 4. Removing rent controls

o 5. What is the best method of controlling rents?

B. Increasing the size of the rental housing stock

o 1. Incentives to the formal sector

o 2. Incentives for the informal sector

C. Environmental standards, slum demolition and urban renewal

D. The role of public rental housing

E. The role of rental housing in national settlement strategies

F. The roles of co-operative housing and non-governmental organizations

G. The role of the law and the courts

V. Recommendations for public policy

Rent controls

Public-sector housing supply

Increasing the supply of private rental housing

Housing conditions

The limits to State policy

The role of the law and the courts

VI. Priorities for research

Tenure patterns

Housing supply

The desire for house-ownership

The characteristics of tenants and non-owners

Landlord-tenant relations

Page 3

Rent controls

Maintenance and repair of the housing stock

The role of the law and the courts

A list of priorities

Bibliography

Page 4


List of Tables

Table 1. Housing tenure in urban areas by country

Table 2. Housing tenure in some cities of developing countries

Table 3. Level of services by tenure and housing sector

Table 4. Rent-to-income ratios for selected cities

Page 5

Table 1. Housing tenure in urban areas by country

Households (thousands)

Percentage O wnersa YearTenants O wners

Africa: Cameroon 122 230 58.1 1976 Libyan Arab Jamahiriya 69 107 55.0 1973 T unisia 129 327 64.3 1975Asia: Bangladesh 359 574 53.9 1973 India 10,112 9,010 47.1 1971 Pakistan 777 2,405 67.7 1980 Republic of Korea 2,562 1,981 42.9 1980 Singapore 202 280 55.0 1980 Sri Lanka 184 292 57.3 1981 T hailand 410 480 47.5 1976 T urkeyb 668 1,181 63.7 1975Latin America: Bolivia 131 198 47.0 1976 Colombia 1,131 2,355 64.8 1985 Ecuador 239 119 40.9 1978 El Salvador 168 156 47.8 1978 Venezuela 461 1,599 72.9 1981Sources: Compendium of Human Settlement Statistics 1982-1984 (United Nations publication) and International Symposium onHousing Organizing Committee (ISHOC), Country Profiles: Housing and Human Settlement Conditions (Yokohama, ISHOC, 1997)supplemented by the housing censuses of Colombia and Venezuela..Notes:a. Owners as a percentage of the total of owners, tenants and others.b. T he cities of Adana, Ankara, Istanbul and Izmir only.

Page 6

Table 2. Housing tenure in some cities of developing countries (percentage) O wners Tenants O thers Year

Africa: Brazzaville 56 34 9 1979 Cairo 31 --- 69 --- 1981 Beni Suef 74 --- 26 --- 1981 Kumasi 10 62 25 1986 Rabat 33 52 15 1981 Sale 47 42 11 1981Asia: Dhaka 22 55 23 1973 Chittagong 33 65 2 1973 Bombay 38 --- 62 --- 1981 Calcutta 24 --- 76 --- 1981 Delhi 53 --- 47 --- 1981 Madras 32 --- 68 --- 1981 Colombo 56 29 15 1981 Bangkok 55 31 14 1980 Hong Kong 26 17 58 1981 Ankara 64 --- 36 --- 1975 Istanbul 56 --- 44 --- 1975 Jakarta 64 23 13 1985 Seoul 41 59 0 1987Latin America: Bogotá 57 40 3 1985 Medellin 65 30 5 1985 Cali 68 27 4 1985 Barranquilla 74 22 4 1985 Caracas 63 31 6 1981 Maracaibo 80 14 7 1981 Valencia 79 16 5 1981 Lima 49 28 23 1980 Mexico City 53 --- 47 --- 1980 Guadalajara 52 39 8 1980 Monterrey 68 25 7 1980 Puebla 48 46 7 1980 Santiago 64 20 16 1982 La Paz 43 35 22 1976 Santa Cruz 52 28 19 1976Sources: Compendium of Human Settlement Statistics 1982-1984 (United Nations publication); ISHOC County Profiles: housing andhuman settlement conditions (Yokohama, ISHOC, 1987) supplemented by the housing censuses of Colombia, Mexico, Peru andVenezuela and by J. Soto, "El acceso a la vivienda de los sectores mas pobres de la región metropolitana" (s.l., Universidad Catolica,1987), (mimeo.); A.G. T ipple and K.G. Willis "Who owns, who rents: tenure choice in a West African city' (University ofNewcastle-upon-T yne, 1989) (mimeo.); R. Keles and U. Kano "Housing and the urban poor in the Middle East - T urkey, Egypt,Morocco and Jordan", MES Studies No.20 (T okyo, Institute of Developing Economics, 1987), Abt Associates; Inc. "Informal housingin Egypt (si., 1982) (mimeo.); and K. Wadhva, "Rental Housing in India: Compulsion or Choice?” Papers presented to the ExpertGroup Meeting on Rental Housing in Developing Countries, held at IHS, Rotterdam, October 9-13, 1989 (Rotterdam, IHS, 1989).Note: Some census authorities Classify data into only owners and tenants, and exclude a category for other kinds of non-owners.

Page 7

Table 3. Level of services by tenure and housing sector in Kumasi

SectorTenure

O wner Family house Tenant Sub-tenant AllT enement 7.1 5.3 5.5 5.2 5.6Indigenous 4.7 4.3 4.5 2.4 4.5Government 7.8 6.8 7.6 9.6 7.8High cost 10.4 8.9 10.9 12.0 10.7T otal 6.8 5.1 5.5 5.2 5.6Source: A.G. T ipple and K.G. Willis, "who owns, who rents: tenure choice in a West African city" (University of Newcastle-upon-T yne,1989) (mimeo.)Calculated on the basis of possession of kitchen, bathroom, piped water and toilet No service = 0, shared service =1 and exclusive use =2 Double weighting was given to possession of water and toilet. Hence the maximum scores are 12 for exclusive services and 6 forshared services.

Page 8

Table 4. Rent-to-income ratios for selected cities

O wners Renters Ratio Percentageowners

Bogotá 0.20 0.18 1.13 57Cali 0.17 0.19 0.89 68Cairo 0.10 0.07 1.47 31Bangalore 0.25 0.10 2.50 26Busan 0.41 0.16 2.60 47Seoul 0.40 0.22 1.79 41Davao 0.04 0.08 0.51 73Manila 0.27 0.09 3.03 33Sources: S.J. Malpezzi and S.K. Mayo, “User cost and housing tenure in developing countries”, Journal of Development Economics, vol.25 (1987), pp. 210-211; ISHOC, Country Profiles: Housing and Human Settlement Conditions (Yokohama, 1987); Economic andSocial Commission for Asia and the Pacific, Physical Profile of Cities in the ESCAP Region (Bangkok, 1982); Compendium of HumanSettlement Statistics 1982-1984 (United Nations publication) ; and Colombian census.

Page 9


List of Figures

Figure 1. Relative likelihood of access to piped water for renters compared with owners for selected cities

Figure 2. Likelihood of house-ownership vis-à-vis household income by deciles for selected developing-countrycities

Page 10

Figure 1. Relative likelihood of access to piped water for renters compared with ownersfor selected cities

Source: A.C. Lemer, The Role of Rental Housing in Developing Countries: a Need for Balance (Washington, D.C.,World Bank, 1987), Water Supply and Urban Development Discussion Paper No. UDD-104.Note: Income groups represent increasing household incomes in equal percentage groups. Likelihood ratio is ratio ofpercentage of renters with piped water to percentage of owners with piped water.

Page 11

Figure 2. Likelihood of house-ownership vis-à-vis household Income, by deciles, forselected developing country cities

Source: A. C. Lemer, The Role of Rental Housing in Developing Countries: a Need for Balance (Washington, D.C.,Works Bank, 1987), Water Supply and Urban Development Discussion Paper, No. UDD-100, p.15.Note: Data from surveys, various years.

Page 12


Bibliography

Aaron, H., "Rent controls and urban development: a case study of Mexico City", Social and Economic Studies, vol. 15(1966), pp. 314-328.

Abt Associates Inc., "Informal housing in Egypt" (1982) (mimeo.).

Amis, P., "Commercialized rental housing in Nairobi, Kenya", in Patten, C.V., (ed.), Spontaneous Shelter: InternationalPerspectives and Prospects (Temple University Press, 1988), pp. 235-257.

Amis, P., "Migration, urban poverty, and the housing market: the Nairobi case", in Eades, J., (ed.), Migrants, Workers,and the Social Order (London, Tavistock Publications, 1987), pp. 249-268.

Amis, P., "Squatters or tenants? The commercialization of unauthorized housing in Nairobi", World Development, vol.12 (1984), pp. 87-96.

Amis, P.. "A shanty town of tenants: the commercialization of unauthorized housing in Nairobi 1960-80" D. Phil.dissertation (University of Kent at Canterbury, 1983).

Angel, S., and Pornchokchai, S., "Bangkok slumlands", Cities, vol. 6 (1989), pp. 136-146.

Aradeon, D., "Regional assessment of human settlements policies in Nigeria", Paper presented at the Symposium onNational Human Settlement Policies and Theory held at University of Sussex, February 1978.

Armus, D., and Hardoy, J.E., "Vivienda popular y crecimientro urbano en el Rosario del novecientos", RevistaLatinoamericana de Estudios Urbano-Regionales, vol. 31 (1984).

Azuela, A., "De inquilinos a propietarios. Derecho y politico en of Programa de Renovacion Habitacional Popular",Estudios Demográficos y Urbanos, vol. 2 (1987), pp. 53-74.

Badcock. B., "Land and housing policy in Chinese urban development, 1976-86", Planning Perspectives, vol. 1 (1986),pp. 147-170.

Bahr, J., "The impact of socio-political structures on intraurban migration of low-income groups and the peripheralgrowth of Latin American metropolitan areas (including examples from Lima and Santiago de Chile)", inKleinpenning, J.M.G., (ed.), Competition for Rural and Urban Space in Latin America, NederlandseGeografische Studies, vol. 25 (1986), pp. 140-163.

Bahr, J., and Mertins, G., "Desarrollo poblacional en el Gran Santiago entre 1970 y 1982: analisis de los resultadoscensales en base a distritos", Revista de Geografía del Norte Grande, vol. 12 (1985), pp. 11-26.

Barnes, S.T., Patrons and Power. Creating a Political Community in Metropolitan Lagos (Manchester UniversityPress, 1987).

Barnes, S.T., "Public and private housing in urban West Africa: the social implications", in Morrison. M.K.C., andGutkind, P.C.W., (eds.), Housing the Urban Poor in Africa (Syracuse, NY, Maxwell School, SyracuseUniversity, 1982), pp. 5-29.

Baross, P., "The articulation of land supply for popular settlements in Third World cities", in Angel, S., and others,(ads.), Land for Housing the Poor (Singapore, Select Books, 1983), pp. 180-209.

Batley, R., Power through Bureaucracy: Urban Political Analysis in Brazil (Gower, 1983).

Beijaard, F., "On conventillos; rental housing in the centre of La Paz, Bolivia", Urban Research Working Paper No. 5(Free University of Amsterdam, 1986).

Benton, L.A.. "Reshaping the urban core: the politics of housing in authoritarian Uruguay", Latin American ResearchReview, vol. 22 (1987), pp. 33-52.

Page 13

Bogus, L.M.M., "Villa do encontro: a oidade chega a periferia" (PRODEUR Cademos de Estudos e Pequisas)(Universidade de Sao Paulo, Faculdade de Arquitectura e Urbanizmo, 1981), vol. 4.

Booth, A., "Raising the roof on housing myths", Shelter (1989).

Bradbury, K.L., and Downs, A., (eds.), Do Housing Allowances Work? (The Brookings Institute, 1981).

Byong Key, S., "Rental housing policy for the low income households - experience and prospects of the Republic ofKorea", Paper presented to the Commission on Human Settlements at its tenth session, Nairobi, 6-16 April 1987.

Carlson, E., "Evaluation of housing projects and programmes: a case report from Venezuela", Town Planning Review,vol. 31 (1961), pp. 187-209.

Castillo Palma, J., "El movimiento urbano popular en Pueblo", in Garcia, M., and Castillo-Palma, J., (eds.), LosMovimientos Sociales en Pueblo (Universidad Autonoma de Puebla, 1986), Tomo 2, pp. 201-360.

Centro de Estudios Urbanos (CEU), "Primeros avances de la investigation "El inquilinato y la vivienda compartida deciudades latinoamericanans", (Caracas, 1989) (mimeo.).

Centro de la Vivienda y Estudios Urbanos (CENVI). "Instituciones publicas y organizaciones sociales frente almejoramiento de cuatro asentamientos en la Ciudad de Mexico" (Medellin, Colombia, Centro de Estudios delHabitat Popular (CEHAP), 1986), Documentos 6.

Centro de la Vivienda y Estudios Urbanos (CENVI), "Vivienda y propiedad en cinoo colonias populares de la Ciudadde Mexico" (Mexico City, 1989) (mimeo.).

Chant, S., and Ward, P.M., "Family structure and low-income housing policy", Third World Planning Review, vol. 9(1987), pp. 5-19.

Cohen, M.A., Urban Policy and Political Conflict in Africa (The University of Chicago Press, 1974).

Collier, D., Squatters and Oligarchs, (Baltimore, Johns Hopkins University Press, 1970).

Connolly, P., "La politica habilacional despues de los sismos", Estudios Dernograficos y Urbanos, vol. 2 (1987), pp.101-120.

Conway, D., and Brown, J., "Intraurban relocation and structure: low-income migrants in Latin America and theCaribbean", Latin American Research Review, vol. 15 (1980), pp. 95-125.

Cooperative Housing Foundation (CHF) "The cooperative development system", CHF Concepts and Methods Series,No. 7 (1985).

Corbett, P., "Council housing for low-income Indian families in Durban: objectives, strategies and effects", in Smith,D.M., (ed.), Living under Apartheid (London, George Allen and Unwin, 1982), pp. 183-197.

Cornelius, W., Politics and the Migrant Poor in Mexico City (Stanford University Press, 11975).

Coulomb, R., "La legislación an materia de vivienda en arrendamiento para el Distrito Federal: Situacion actual (1985) ypropuestat reglamentarias", Cuademos dal CENVI (Mexico City, 1985).

Coulomb, R., "La producción y consurno de la vivienda on alquiler con especial referencia a la Area Metropolitana dela Ciudad de Mexico" (1985) (mimeo.).

Coulomb, R., "La vivienda de alquiler on Las areas de reciente urbanización", Revista de Ciencias Sociales yHumanidades, tomo VI (1985), pp. 43-70.

Coulomb, R., "Rental housing and the dynamics of urban growth in Mexico City", in Gilbert, A.G., led.), Housing andLand in Urban Mexico (San Diego, Centre for U.3. Mexican Studies, University of California, 1989), pp. 39-50.

Crooke, P., "Popular housing supports and the urban housing market", in Skinner, R.J., and Rodell, M.J., (ads.),People, Poverty and Shelter: Problems of Self-help Housing in the Third World (London, Methuen, 1983), pp.173-191.

Page 14

Cuenya, B., "El submercado de alquiler de piezas en Buenos Aires", Boletin de Medio Ambiente y Urbanizacidn, tomo17 (1986), Suplemento especial, pp. 3-8.

Dasso, J., "An evaluation of rent supplements", Land Economics, vol. 44 (1968), pp. 441-449.

Daunton, M.J., A Property Owning Democracy? Housing in Britain (London, Faber and Faber, 1987).

Daunton, M.J., Harare and Home in the Victorian City. Working Class Housing 1850-1918 (Edward Arnold, 1983).

Desai, A.R., and Pillal, S.D., (eds.), Slums and Urbanization (Bombay, Popular Prakashan, 1970).

Dietz, H.A., Poverty and Problem-solving under Military Rule: The Urban Poor in Lima, Peru (Austin, University ofTexas Press, 1981).

Downs, A., Rental Housing in the 1980s (The Brookings Institute, 1983).

Drakakis-Smith, D.W., High Society: Housing Provision in Metropolitan Hong Kong 1954 to 1979 (Hong Kong,Centre of Asian Studies, University of Hong Kong, 1979).

Drakakis-Smith, D.W., Urbanization, Housing and the Development Process (London, Croom Helm, 1981).

Dreier, P., "The tenants' movement in the United States", International Journal for Urban and Regional Research,vol. 8 (1984), pp. 255-279.

Duncan, S.S., "The housing question and the structure of the housing market", Journal of Social Policy, vol. 6 (1977),pp. 385-412.

Durbad-Lasserve, A., "L'exclusion des pauvres dans les villas du tiers-monde" (Paris, L'Harmattan, 1983).

Dwyer, D.J., People and Housing in Third Word Cities (London, Longman, 1975).

Dwyer, D., "Urban housing and planning in China", Transactions of the Institute of British Geographers, vol. 11(1987), pp. 479- 489.

Eckstien, S., 'The political economy of lower class areas in Mexico City: some constraints on local businessprospects", Latin American Urban Research, vol. 3 (1975), pp. 125-146.

Economic and Social Commission for Asia and the Pacific (ESCAP), Physical Profile of Cities in the ESCAP Region(Bangkok, 1982).

Edwards, M., "Cities of tenants: renting as a housing alternative among the Colombian urban poor", unpublisheddoctoral dissertation, University College London (1981).

Edwards, M., in Gilbert, A.G., Hardoy, I.E., and Ramirez, R., (ads.), Urbanization in Contemporary Latin America(Chichester, John Wiley, 1982), pp. 129-158.

Ekpenyong, S., "Housing, the state and the poor in Port Harcourt", Cities, vol. 6 (1989). pp. 39-49.

El Kadi, G., "Market mechanisms and spontaneous urbanization in Egypt: the Cairo case", International Journal ofUrban and Regional Research, vol. 12 (1988), pp. 22-37.

Englander, D., Landlord and Tenant in Urban Britain, 1838-1918 (Oxford, Clarendon Press, 1983).

Evers, H-D., "Urban landownership, ethnicity and class in southeast Asian cities", International Journal of UrbanRegional Research, vol. 8 (1984), pp. 481-496.

Fass, S., "Housing the ultra-poor: theory and practice in Haiti", Journal of the American Planning Association, vol. 53(1987), pp. 193-205.

Fass, S., "Port-au-Prince: awakening to the urban crisis", in Latin American Urban Research, vol. 6 (1978), pp. 155-180.

Fedelonjas, La Vivienda Arrendada corno Alternative al Deficit Habitacional (Bogotá, 1982).

Page 15

Field, B., "Public housing in Singapore", Land Use Policy, vol. 4 (1987), pp. 147-156.

Findlay, A.M., and Paddison, R., "Planning the Arab city: the cases of Tunis and Rabat", Progress in Planning, vol.26 part one (1986), pp. 1-82.

Foster, J., "Review of Wohl, A. How imperial London preserved its slums", International Journal of Urban andRegional Research, vol. 3 (1979), pp. 93-114.

Garcia Peralta, B., "Situación de la vivienda on las delegaciones afectadas por los sismos", Revista Mexicana deSociologia, tomo 48 (1986), pp. 283-292.

Garza, G., and Schteingart, M., La Accion Habitacional del Estado on Mexico (El Colegio de Mexico, 1978).

Ghai, Y.P., and McAuslan, J.P.W.B., Public Law and Political Change in Kenya (Nairobi, Oxford University Press,1970).

Gilbert, A.G., "Home enterprises in poor urban settlements", Regional Development Dialogue, vol. 9 (1988), pp. 21-39.

Gilbert, A.G., "Housing during recession: illustrations from Latin America", Housing Studies. vol. 4 (1989), pp. 155-166.

Gilbert, A.G., "Pirates and invaders: land acquisition in urban Colombia and Venezuela", World Development, vol. 9(1981), pp. 657-678.

Gilbert, A.G., "Latin America's urban poor: shanty dwellers or renters of rooms?", Cities, vol. 4 (1987), pp. 43-51.

Gilbert, A.G., "The tenants of self-help housing: choice and constraint in the housing markets of less developedcountries", Development and Change, vol. 14 (1983), pp. 449-477.

Gilbert, A.G., and Gugler, J., Cities, Development and Poverty: Urbanization in the Third World (London, OxfordUniversity Press, 1982).

Gilbert, A.G., and Varley, A., "Housing tenure and the urban poor in Third World cities", Final Report to OverseasDevelopment Administration (1988) (mimeo.).

Gilbert, A.G., and Ward, P.M., Housing, the State and the Poor. Policy and Practice in Three Latin American Cities(Cambridge, Cambridge University Press, 1985).

Gilbert, A.G., and Ward, P.M., "Residential movement among the poor: choice or constraint?", Transactions of theInstitute of British Geographers, vol. 7 (1982), pp. 129-49.

Green, G., "Finding a home in a frontier city: the dynamics of housing tenure in Santa Cruz, Bolivia", Doctoraldissertation, University College London (1988).

Green, G., "The quest for tranquilidad: paths to home ownership in Santa Cruz, Bolivia", Bulletin for Latin AmericanResearch, vol. 7 (1988), pp. 1-16.

Grimes, O. F., Housing for Low Income Urban Families (Baltimore, Johns Hopkins University Press, 1976).

Gupta, D.B., "Urban housing in India", World Bank Staff Working Papers, No. 730 (Washington, D.C., 1985).

Gutierrez, J., and others, "The recent worldwide economic crisis and the welfare of children: the case of Cuba", WorldDevelopment, vol. 12 (1984), pp. 247-260.

Ha, Seong-kyu, "Korea", in Ha, Seong-kyu, (ed.), Housing Policy and Practice in Asia (London, Croom Heim, 1987).pp. 80-116.

Hai, T.S., "Low cost housing in Malaysia: a review of public sector involvement", in Young, Y.M., (ed.), A Place toLive: More Effective Low-cost Housing in Asia (Ottawa, International Development Research Centre, 1983), pp.65-78.

Hamer, A.M., "Las subdivisiones no reglamentadas de Bogotá: los mitos y realidades de la constructión de is viviendasupplementaria", in Corporación Centro Regional de Población, La Ciudad Documento No. 24 (1981).

Page 16

Handelman, H., 'High rises and shantytowns: housing the poor in Bogotá and Caracas", American Universities FieldStaff Reports No. 9 (1979).

Hansen, E., and Williams, J., "Economic issues and the progressive housing development model", in Patter, C.V., (ed.),Spontaneous Shelter: International Perspective and Prospects (Temple University Press, 1988), pp. 303-325.

Hardoy, J.E., and Satterthwaite, D., Shelter: Need and Response - Housing, Land and Settlement Policies inSeventeen Third World Nations (Chichester, John Wiley, 1981).

Harloe, M., Private Rented Housing in the U.S. and Europe (London, Croom Helm, 1985).

Harms, H., "Production of housing in Cuba", Proceedings of the Bartlett Summer School 1981: The Production of theBuilt Environment (London, 1982).

Heskin, A.D., Tenants and the American Dream: Ideology and the Tenant Movement (New York, Praeger, 1983).

Hoenderdos, W., Lindert, P., and Verkoren, O., "Residential mobility, occupational changes and self-help housing inLatin American cities: first impressions from a current research programme", Tijdschrift voor Economische etSociale Geografie, vol. 74 (1983), pp. 376-386.

Hollnsteiner, M., "The case of 'the people versus Mr. Urbano Planner y Administrator"', in Abu Lughod, J., and Hay,R., (ads.), Third World Urbanization (Cambridge, Cambridge University Press, 1977).

Iliffe, J., The African Poor. A History (Cambridge, Cambridge University Press, 1967). International Symposium onHousing Organizing Committee (ISHOC), Country Profiles: Housing and Human Settlement Conditions(Yokohama, 1987).

Jackson. H.E., "Intro-urban migration of Mexico City's poor", Unpublished doctoral dissertation, University ofColorado (1973).

Jaramillo, S., "Alcances y limites del componente 'arrendaminetos' del IPC como indicator de la evolucion de losprecios de la vivienda", CEDE Analisis del Indice de Precios al Consumidor, Documento de Discussion No. 2(1987).

Jaramillo, S., "Entre el UPAC y la autoconstrucción: comentarios y sugerencias a la política de vivienda", Controversia(Bogotá, CINEP, 1985), pp. 123-124.

Jones, C., "Housing: the element of choice", Urban Studies, vol. 16 (1979), pp. 197-204.

Jurema, A., Sa Feria 13: Os ultimos digs do Goveno Joao Goulart (Rio de Janeiro, Editorial O Cruzeiro, 1964).

Kabagambe, D., and Moughtin, C., "Housing the poor: a case study in Nairobi", Third World Planning Review, vol. 5(1983), pp. 227- 248.

Karst, K., Schwartz, M., and Schwartz, A., The Evolution of the Law in the Barrios of Caracas (s.l., University ofCalifornia Press, 1973).

Keare, D.A., and Parris, S., "Evaluation of shelter programs for the urban poor: principal findings" World Bank StaffWorking Papers No. 547 (Washington, D.C., 1982).

Keles, R., and Kano, H., Housing and the Urban Poor in the Middle East - Turkey, Egypt, Morocco and Jordan(Tokyo, Institute of Developing Economies, 1987), M.E.S. Series No. 20.

Kemp, P., "Housing landlordism in late nineteenth century Britain", Environment and Planning A, vol. 14 (1982), pp.1437-1447.

Kemp, P., "Some aspects of housing consumption in late nineteenth century England and Wales", Housing Studies,vol. 2 (1987), pp. 3-16.

Kirkby, R.J. R., Urbanization in China: Town and Country in a Developing Economy 1949-2000 A.D. (London,Croom Helm, 1985).

Page 17

Klak, T., "From socially progressive policy to socially regressive program: the case of Jamaica's National HousingTrust". Paper presented at Association of American Geographers annual meeting, Baltimore (1989).

Kojima, R., Urbanization and Urban Problems in China (Tokyo, Institute of Developing Economies, 1987).

Kowarick, L., and Ant, C., "Cem anos de promiscuidade: o cortiço na Cdade de Sao Paulo", in Kowarick, L., (ed.), AsLutas Socials e a Cidade: Sao Paulo: Passado e Presente, Paz e Terra (1988), pp. 49-74.

Kowarick, L., and Bonduki, N., "Espaco urbano e espaco politico: do populismo a redemocratizacao", in Kowarick, L.,(ed.), As Lutas Socials e a Cidade: Sao Paulo: Passado e Present Paz e Terra (1988), pp. 133-168.

Laquian, A.A., Slums are for People (Manila, D.M. Press, 1969).

Laun, J. I., "El estado y la vivienda en Colombia: analysis de urbanizaciones del Instituto de Credito Territorial enBogotá", in Castillo, C., (ed.), Vida Urbana y Urbanismo (Bogotá, Instituto Colombiano de Cultura, BibliotecaBasica de Colombia, 1977) vol. 30.

Lee, D.S., "Summary report: public housing programs in Korea", paper presented to Yokohama InternationalSymposium on Housing (1987) (mimeo.).

Lemer, A.C., The Role of Rental Housing in Developing Countries: A Need for Balance (Washington, D.C., WorldBank, Water Supply and Urban Development Department, 1987), discussion paper, Report No. UDD-104.

Lendeck, A., The Political Economy of the New Left (New York, Harpor and Row, 1972).

Leong, L.K., and others, "Management of Singapore's new towns", in Yeung, Y.M., (ed.), A Place to Live: MoreEffective Low-cost Housing in Asia (Ottawa, International Development Research Centre, 1983), pp. 19-64.

Lewin, A.C., Housing Co-operatives in Developing Countries (Chichester, John Wiley, 1981).

Lozano, E., "Housing the urban poor in Chile: contrasting experiences under 'Christian Democracy' and 'UnidadPopular'", Latin American Urban Research, vol. 4 (1975), pp. 135-155.

MacLennan, D., "Maintenance and modernization of urban housing", OECD Urban Affairs Programme (1986) (mimeo.).

Macoloo, P., "Housing the urban poor: a case study of Kisumu town, Kenya", Third World Planning Review, vol. 10(1988), pp. 159-174.

Malpezzi, S.J., "Rent control and housing market equilibrium: theory and evidence from Cairo, Egypt", Doctoraldissertation, George Washington University (1986).

Malpezzi, S.J., "Rent controls: an international comparison", Paper presented to American Real Estate and UrbanEconomics Associations (1984).

Malpezzi, S., and Mayo, S.K., "The demand for housing in developing countries: empirical estimates from householddata", Economic Development and Cultural Change, vol. 35 (1987). pp. 687-721.

Malpezzi, S., and Mayo, S., "User cost and housing tenure in developing countries", Journal of DevelopmentEconomics, vol. 25 (1987), pp. 197-220

Marris, P., "The meaning of slums and patterns of change", International Journal of Urban and Regional Research,vol. 3 (1979), pp. 419-441.

Marroquin, E., "Politica urban y explotación: las vecindades del centro de Puebla", in Contreras, E., and others, (eds.),Investigaciones Universitarias de Urbanismo. Memona de la Primera Mesa de Trabajo Sobre InvestigacionesUniversitarias de Urbanismo (Instituto de Ciencias de la Universidad Autonoma de Puebla, 1986), pp. 159-180.

Massolo, A., "Que el gobierno entidenda, to primero es is vivienda!", Revista Mexicana de Sociologia, tomo 48(1986), pp. 195-238.

Matthey, K., "Microbrigadas in Cuba", Habitat International, tomo 12 (1988), pp. 55-62.

Page 18

Mayo, S., "How much will households spend for shelter?", Urban Edge, vol. 9 (1985), pp. 4-5. Mayo, S.K., Malpezzi, S.,and Gross, D.J., "Shelter strategies for the urban poor in developing countries", The World Bank ResearchObserver, vol. 1 (1986), pp. 183-203.

McCallum, D., and Benjamin, S., "Low-income urban housing in the third world: broadening the economicperspective", Urban Studies, vol. 22 (1985), pp. 277-287.

Mendez Rodriguez, A., "Situación de la vivienda en arrendamiento y su impacto en el nivel de vida de lostrabajadores", Instituto de Investigaciones Eeonómicas, Universidad Nacional Autónoma de Mexico (1987)(mimeo.).

Mexico, Secretaria de Hacienda y Credito Publico, Programa Financiero de Vivienda (Mexico City, 1964).

Misra, R. P., (ed.), Million Cities of India (New Delhi, Vikas, 1978).

Moitra, M.K., and Samajdar, S., "Evaluation of the slum improvement program of Calcutta bustees", in Skinner, R., andothers, (eds.), Shelter Upgrading for the Urban Poor. Evaluation of Third World Experience (Manila, IslandPublishing House, 1987), pp. 69-86.

Muoghalu, L.N., 'he urban poor and accessibility to public housing in Nigeria", in Makinwa, P.K., and Ozo. O.A..(eds.). The Urban Poor in Nigeria (lbadan, Evans Brothers, 1987), pp. 161-171.

Muttagi, P.K., "Rehabilitation in Bombay", Habitat International, vol. 12 (1988), pp. 71-87.

Necochea, A., "El allegamiento de los sin tierra, estrategia de supervivencfa en vivienda", Revista Latinoamericana deEstudios Urbano Regionales, vol. 13-14 (1987), pp. 85-100.

Nelson, J.M., Access to Power. Politics and the Urban Poor in Developing Nations (Princeton University Press,1979).

Nelson, K.W., "Choices and opportunities: low-income rental housing in Indonesia", Urban Institute Working Paper3780-12 (1988).

Nientied, P., and van der Linden, J.J., "Evaluation of squatter settlement upgrading in Baldia, Karachi", in Skinner, R.,and others, (eds.), Shelter Upgrading for the Urban Poor: Evaluation of Third World Experience (Manila,Island Publishing House, 1987), pp. 107-126.

O'Connor, A.M., The African City (London, Hutchinson, 1983).

Okpala, D.C.I., "Rent control reconsidered", in Onibokun, P., (ed.), Housing in Nigeria (lbadan, National Institute forSocial and Economic Research, 1985), pp. 139-159.

Onibokun, O., (ed.), Housing in Nigeria (lbadan, National Institute for Social and Economic Research, 1985).

Padron, M., and Calderon, J., "Some contradictions about the tugurios: two case studies of development anderadication of two slums in Lima, Peru", in Bruno, E., and others, (eds.), Development of Urban Low incomeNeighbourhoods in the Third World (Darmstadt, Archimed Verlag, 1984), pp. 175-189.

Payne, G., Informal Housing and Land Subdivisions in Third World Cities: A Review of the Literature, Reportprepared for the Overseas Development Administration (Oxford Polytechnic, 1989).

Peake, L., "Government housing policy and its implications for women in Guyana", in Peake, L., and Moser, C.O.N.,(eds.), Women, Human Settlements, and Housing (London, Tavistock Publications, 1987), pp. 113-138.

Peil, M., "African squatter settlements", Urban Studies, vol. 13 (1976), pp. 155-166.

Peil, M., Cities and Suburbs: Urban Life in West Africa (s.l., Africana Publishing House, 1981).

Peil, M., and Sada, P.O., African Urban Society (Chichester, John Wiley, 1984).

Perez Perdomo, R., and Nikken, P., "The law and home ownership in the barrios of Caracas", in Gilbert and others,(eds.), Urbanization in Contemporary Latin America (Chichester, John Wiley, 1962), pp. 205-230.

Page 19

Perlman, J., "The failure of influence: squatter eradication in Brazil", in Grindle, M.S., (ed.), Politics and PolicyImplementation in the Third World (Princeton University Press, 1980), pp. 250-28a.

Perlo-Cohen, M., Estado, Vivienda y Estructura Urbana on of Cardenismo: El Caso de la Ciudad de Mexico (MexicoCity, Instituto de Investigaciones Sociales, Universidad National Autonoma de Mexico, 1981).

Perlo-Cohen, M., "Politica y vivienda en Mexico 1910-1952", Revista Mexicana de Sociologia, tomo 41 (1979), pp.769-835.

Peru, Instituto Nacional de Estadistica, Peru: Compendio Estadistico 1985 (Lima, 1986).

Philippines, Ministry of Human Settlements, Philippine Shelter System and Human Settlements (Manila, 1983).

Portillo, A.J., "El arrendamiento de vivienda en la Ciudad de Mexico", Cuadernos Universitarios No. 5 (Mexico City,Universidad Autonoma Metropolitana, 1984).

Pugh, C., "Housing in Singapore: the effective ways of the unorthodox", Environment and Behaviour, vol. 19 (1987),pp. 211- 230.

Pugh, C., "The World Bank and urban shelter in Calcutta" (Sheffield Polytechnic, 1988) (mimeo.).

Rakodi, C., "The production of housing in Harare, Zimbabwe: components, Constraints and policy outcomes", Trialog, vol. 20 (1989), pp. 7-13.

Rakodi, C., "Upgrading in Chawama, Lusaka: displacement or differentiation?", Urban Studies, vol. 25 (1988), pp.297-318.

Ray, T., The Politics of the Barrios of Venezuela (s.l., University of California Press, 1969).

Renaud, B., "Financing shelter", in Rodwin, L., (ed.), Shelter, Settlement, and Development (London, George Allen andUnwin, 1987), pp, 179-203.

Reviez, E., and others, Poder e Informacion (Bogotá, Universidad de los Andes, 1977).

Riofrio Benavides, G., Se Busca Terreno para Proxima Barriada: Espacios Disponibles en Lima 1940, 1978, 1990(Lima, DESCO, 1978)

Rivas E., "Estudio analitico de un submercado de vivienda: arrendamiento de pieza" (Centro de Estudios Urbanos yRegionales 1977) (mimeo.).

Robben, P.J.M., "Measurement of population dynamics following squatter settlement improvement in Ashok Nagar,Madras", in Skinner, R., and others, (ads.), Shelter Upgrading for the Urban Poor: Evaluation of Third WorldExperience (Manila; Island Publishing House, 1987), pp. 87-106.

Rodas, F., and Sugranyes, A., "El inquilinato on la ciudad de Guatemala: et caso de la zona 8", Boletin de MedioAmbiente y Urbanizacion, tomo 7 (1988), pp. 3-18.

Sachs, C., "The growth of squatter settlements in Sao Paulo: a study of the perverse effects of the state's housingpolicies", Social Science Information, vol. 415 (1983), pp. 749-775.

Sanchez-Leon, A., and others, Tugurización an Lima Metropolitana Lima, DESCO, 1979).

Sandbrook, R., The Politics of Basic Needs: Aspects of Assaulting Poverty in Africa (London. Heinemann, 1982).

Sands, G., "Into the nineties: review of Anthony Downs' Rental Housing in the 1980s", Growth and Change, vol. 15(1884), p. 56.

Sanyal, B., "A critical look at the housing subsidies in Zambia", Development and Change, vol. 12 (1981), pp. 409-444.

Sarin, M., Urban Planning in the Third World: The Chandigarh Experience (Salem, N.H., Mansell, 1982).

Saunders, P., "The meaning of "home" in contemporary English culture", Housing Studies (1989), pp. 155-166.

Page 20

Scobie, J., Buenos Aires: Plaza to suburb 1870-1910 (London, Oxford University Press, 1974).

Skinner, R., "Self-help, community organization and politics: villa El Salvador, Lima", in Ward, P.M. (ed.), Self-helpHousing: A Critique (Salem, N.H., Mansell, 1982), pp. 175-208.

Smith, D.M., (ed.), Living under Apartheid: Aspects of Urbanization and Social Change in South Africa (London,George Allen and Unwin, 1982).

Soto, J., "El acceso a la vivienda de los sectores mas pobres de la región metropolitana" (Universidad Catolica, 1967)(mimeo.)

Stedman-Jones, G., Outcast London (Oxford University Press, 1971).

Sternlieb, G., The Tenement Landlord (Rutgers University Press, 1969).

Strassman, W.P., "Types of neighbourhoods and home-based enterprises: evidence from Lima, Peru" (Ann Arbor,Michigan State University, 1985) (mimeo.)

Stren, R. E., Housing the Urban Poor in Africa: Policy, Politics, and Bureaucracy in Mombasa (Berkeley, Institute ofInternational Studies, University of California, 1978).

Stren, R.E., "Underdevelopment, urban squatting and the state bureaucracy: a case study of Tanzania", CanadianJournal of African Studies, vol. 16 (1982), pp. 67-91.

Stren, R.E., and White, R., (ads.), African Cities in Crisis (Boulder, CO, Westview, 1989).

Sudra. T., "Low-income housing system in Mexico City", Unpublished doctoral dissertation (Cambridge, MS,Massachusetts Institute of Technology, 1976).

Sundaram., P.S.A., "India", in Ha, S.-K., (ed.), Housing Policy and Practice in Asia (London, Croom Helm, 1987).

Sunil, K., and van den Eerenbeemt, F., "Renting - its place in the housing market of developing countries", Institute forHousing Studies Bulletin (Rotterdam, 1988).

Tanphiphat, S., "Recent trends in low-income housing development in Thailand", in Young, Y.M., (ed.), A Place toLive: More Effective Low-income Housing in Asia (Ottawa, International Development Research Centre, 1983),pp. 103-120.

Taschner, S.P., "Diagnosis and challenges off housing in Brazil". Paper presented to the Conference on Housing,Policy Urban Innovation, Amsterdam, 27 June-1 July, 1988.

Temple, F.T., and Temple, N.W., "The politics of public housing in Nairobi", in Grindle, M.S., (ed.), Politics and PolicyImplementation in the Third World (Princeton University Press, 1980), pp. 224-249.

Tipple, A.G., "The development of housing policy in Kumasi, Ghana, 1901 to 1981" (University ofNewcastle-upon-Tyne, Centre for Architectural Research and Development Overseas, 1988).

Tipple, A.G., The History and Practice of Rent Controls in Kumasi, Ghana (Washington, D.C., World Bank, WaterSupply and Urban Development Department, 1988), Working paper No. 88-1

Tipple, A.G., and Willis, K.G., "Who owns, who rents: tenure choice in a West African city" (University ofNewcastle-upon- Tyne, 1989) (mimeo.)

Tuncalp, S., 'he housing programme in Saudi Arabia", The World Planning Review, vol. 9 (1987), pp. 345-60.

Tung, F., "Public-housing management in Hong Kong's new towns", in Yeung, Y.M., (ed.), A Place to Live: MoreEffective Low-cost Housing in Asia (Ottawa, International Development Research Centre, 1983), pp. 199-214.

Turan, M., "Poverty, prudence, and place-making: strolling through gecekondus", Habitat International, vol. 11(1987), pp. 77-102.

Turner, B., (ed.), Building Community(s.l., Building Community Books. 1988).

Page 21

Turner, J.F.C., "Barriers and channels for housing development in modernizing countries", Journal of the AmericanInstitute of Planners, vol. 33 (1967), pp. 167-181.

Turner, J.F.C., "Housing priorities, settlement patterns, and urban developing in modernizing countries", Journal ofthe American Institute of Planners, vol. 34 (1968), pp. 354-363.

United Nations, Compendium of Housing Statistics 1975-1977 (New York, 1980).

United Nations, Compendium of Human Settlement Statistics, 1982-1984 (New York, 1985).

United Nations, Review of Rent Control in Developing Countries (New York, 1979).

United Nations Centre for Human Settlements (Habitat), Case Study of Sites and Services Schemes in Kenya: Lessonsfrom Dandora and Thika (Nairobi, 1987).

United Nations Centre for Human Settlements (Habitat), Global Report on Human Settlement 1986 (Oxford UniversityPress, 1987).

United Nations Centre for Human Settlements (Habitat), Rehabilitation of Inner City Areas: Feasible Strategies(Nairobi, 1986) (HS/79/85/E).

United Nations Centre for Human Settlements (Habitat), Renovación Habitacional Popular en of D.F. LaReconstruccion de Vivienda an of Contro Historico de la Ciudad de Mexico despues de los Sismos deSetiembre de 1985 (Nairobi, 1987).

United Nations Centre for Human Settlements (Habitat), Shelter for the Homeless: The Role of Non-governmentalOrganizations (Nairobi, 1988).

United Nations Centre for Human Settlements (Habitat), Shelter, Infrastructure and Services for the Poor inDeveloping Countries: Some Policy Options (Nairobi, 1987) (HS/OP/87-22).

United Nations Centre for Human Settlements (Habitat), Upgrading of Inner-city Slums (Nairobi, 1984)(HS/CONF/84.1E).

United Nations, Department of Economic and Social Affairs, Non- profit Housing Associations: Organization,Financing and Structural Integration (New York, 1975).

Urrutia, M., "Latin America and the crisis of the 1980s", in Emmerji, L., (ed.), Development Policies and the Crisis ofthe 1980s (Paris, Organisation for Economic Co-operation and Development, 1987), pp. 56-69.

Valladares, L., "Working the system: squatter response to settlement in Rio de Janeiro", International Journal ofUrban and Regional Research, vol. 2 (1978), pp. 12-25.

Villa, M., "La politicización innecesaria: el regiment politico mexicano y sus exigencias de pasividad dudadana a losdamnificados", Estudios Demograficos y Urbanos, tomo 2 (1987), pp. 27-52.

Violich, F., Cities of Latin America (New York, Reinhardt, Holt and Winston, 1944).

Von Hayek, F., and others, Verdict on Rent Control: Essays on the Economic Consequences of Political Action toRestrict Rents in Five Countries (s.l., Institute of Economic Affairs, 1972).

Wadhva, K., Rental Housing in India: Compulsion or Choice?, Papers presented to Expert-group M eating on RentalHousing in Developing Countries, held at IHS, Rotterdam, 9-13 October 1989.

Wahab, E.A., "The tenant market of Baldia township: towards a more general understanding of tenancy in squattersettlements", Free University Urban Research Working Paper No. 3 (1984).

Ward, P.M., (ed.), Self-help Housing. A Critique (Salem, N.H., Mansell, 1982).

Wegelin, E.A., Urban Low-income Housing and Development (Rotterdam, Martin Nijhoff, 1978).

Westerlund, L., "The housing allowance controversy and the Swedish example", Built Environment, vol. 14 (1988), pp.

Page 22

168-176.

WOW, J.W., and Perkal, A., (eds.), Statistical Abstract of Latin America 23 (Los Angeles, UCLA Latin AmericaCenter Publications, 1984).

Wohl, A., "The housing of the working classes in London, 1815- 1914", in Chapman, S.D., (ed.), The History ofWorking-class Housing: Symposium (London, David and Charles, 1971), pp. 13-54.

World Bank, Shelter, Poverty and Basic Need (Washington, D.C., 1980).

Yujnovsky, O., Claves Politicas del Problema Habitacional Argontino (s.l., Grupo Editor Latinoamericano, 1984).

Page 23

ABOUTStrategies for low-income shelter and services

development:The rental-housing option

HS/172/89 EISBNE 92-1-131533-6 (electronic version)

Text source: UNCHS (Habitat) printed publication: ISBN 92-1-131092-X (published in 1989).This electronic publication was designed/created by Inge Jensen.

This version was compiled on 2 January 2006.Copyright© 2005-2006 UN-HABITAT.

All rights reserved. This publication has been reproduced without formal editing by the United Nations.The description and classification of countries and territories in this study and the arrangement of the material do notimply the expression of any opinion whatsoever on the part of the Secretariat of the United Nations concerning thelegal status of any country, territory, city or area, or of its authorities, or concerning the delimitation of its frontiers orboundaries, or regarding its economic system or degree of development.Excerpts from the text may be reproduced without authorisation, on condition that the source is indicated.

UN-HABITAT publications can be obtained from UN-HABITAT's Regional Offices or directly from:

UN-HABITAT,Information Services Section,

G.P.O. Box 30030,Nairobi 00100, KENYA

Fax: (254) 20-7623477 or (7624266/7)E-mail: [email protected]

Web-site: http://www.unhabitat.org/

Page 24

mailto:[email protected]

http://www.unhabitat.org/

http://www.unhabitat.org/


Introduction

Hundreds of millions of tenants live in cities in developing countries. There are millions of other householdswhich share accommodation, have it provided by their employer, look after other people's houses or stay with theirrelatives. In many respects, there is a profound ignorance of who this largo mass of people is and of the conditions inwhich they live. Almost nothing is known about those who provide rental accommodation. This paper attempts tosummarize what is known and what is needed to be known about the demand for and supply of rental housing. It alsomakes some suggestions about rental-housing policy, about the feasibility of improving the quality and increasing thestock of rental accommodation and, finally, about the significance of tenure in housing choice.

Page 25


IntroductionA. Contribution of rental accommodation to housing the urban poor

1. How many tenants?

Figures on housing tenure in cities of developing countries are not very reliable. Often they are less than recentand, in some places, are lacking altogether. Therefore, it is impossible to make very many specific statements with totalconfidence. All that is clear is that most cities in developing countries contain large numbers of tenants and that sometenants make up the majority of inhabitants. At least 80 per cent of China's 330 million urban dwellers are tenants ofsome kind, (Badcock, 1986; Kojima, 1987) approximately hall of India's 159 million urban dwellers are tenants, andaround one quarter of Brazil's 110 million urban dwellers do not live in their own house. (Taschner, 1988) Howeverdubious some of the definitions and "ever ancient some of the data, the sheer numbers of tenants suggest that highpriority should be given to a study of their situation.

Page 26



2. Variations betw een countries and betw een cities

While there are huge numbers of tenants in the cities of developing countries as a whole, there are alsoenormous variations in the proportions of tenants and owners between cities. Table 1 shows that, while three out offour urban Venezuelans own or are buying their own house and two out of three Colombians, less than one in twoIndians, Koreans and Thais are in a similar position. In general, most people in cities of West Africa, China, India andthe Republic of Korea rent accommodation, while a majority of people in the large Latin American and West Asiancities tend to be owner-occupiers.

Tenure mixes vary not only between cities in different countries but also within countries. Table 2 shows that76 per cent of the population in Calcutta rents or shares accommodation, compared with only 47 per cent in Delhi. InEgypt, less than one household in three owns a house in Cairo but three out of four in the city of Beni Suef. InColombia, the proportion of tenants in Bogotá is twice as high as that in the city of Barranquilla. In Venezuela, one inthree households rents in Caracas but only one in six in Maracaibo or Valencia. Clearly, levels of urban houseownership are highly variable even within countries.

Page 27



3. Is the proportion of tenants increasing or decreasing through time?

There is a divide in tenure between urban and rural areas; country dwellers are much more likely to live in theirown houses than are city dwellers. In the Republic of Korea, 43 per cent of urban dwellers own property comparedwith 80 per cent of country dwellers. In Sri Lanka, 80 per cent of country dwellers own property compared with only 57per cent of urban dwellers. It will not be wrong to say that there are no exceptions to this rule; such a generalizationapplies to rich countries and poor, to capitalist economies and to socialist.

The major shift of people from the countryside to the city, that has occurred over the past 30 to 40 years,implies that there has been an important change in the housing-tenure structure in most countries. In Latin America,for example, the change from a rural to a predominantly urban population means that a much higher proportion of thepopulation than before now rents. This has been partly balanced by the trend in most cities towards widespreadownership. If, in the past, most urban dwellers were tenants, in many cities, house-owners are now the majority. InMexico City, only 22 per cent of families occupied their own houses in 1060 but 53 per cent in 1980; in Lima, 11 per centof dwellings were owner-occupied in 1940 compared with 49 per cent in 1981 (Gilbert and Varley, 1988; RiofrioBenavides, 1978:58; Peru, INE, 1986). Inmost developing countries, the process of self-help construction, combinedwith the development of middle-class suburbs, has helped the change for the majority of households from beingrenters to owners.

What is uncertain is whether, currently, there is such a strong movement towards house ownership. In somecities, at least, there are reasons to believe that the commercialization of land is making self-help ownership difficult.(Ward, 19132) Reports from cities as diverse as Bogotá, Cairo, Karachi and Nairobi suggest that the poor are havingincreasing difficulty in buying land, because of the rising costs of land and buildings. (Abt Associates, 1982; Amis,1984; Gilbert, 1981; Payne, 1989; Wahab, 1984)

Elsewhere, the effects of economic recession might be placing ownership, even of self-help housing, beyondthe means of many poor households. In some African and Latin American cities, real wages have declined sodramatically that the purchase of land is beyond the reach of most budgets (Gilbert, 1989). However, even if it besuspected that economic recession is slowing the transition to ownership in some cities, it is not a worldwidephenomenon. Some national economies continue to grow rapidly while others decline, and State policy is highlydiverse: housing policy in. say. China is very different from that in, say, Chile.

The need for caution over discerning general patterns is certainly indicated by a glance at the relationshipbetween levels of national affluence and urban house-ownership. Cities in such affluent countries as Canada, NewZealand, and the United States of America have a majority of house-owners, whereas the cities of other such affluentcountries as France, the Federal Republic of Germany, and Switzerland are dominated by tenants. There is similarvariation in less affluent countries: Indian and West African cities contain high proportions of tenants, whereas thoseof Indonesia and Turkey are dominated by house-owners. Clearly, national affluence is not the main determinant oftenure patterns; the form of land and housing markets, the nature of urban transport systems and, most important ofall, the attitude of the State are critical. (Daunton 1987; Harloe, 1985)

Page 28


IntroductionB. Position of rental housing in national housing policy

During the past two decades, rental housing in developing countries has been badly neglected both bygovernments and in the housing literature. The problem of accommodating large numbers of tenants has simply notfound a place in most national plans or in most housing policies. In some countries it is true that housing has oftenfigured prominently in recent national plans, yet rental housing has never been discussed. In some, nationalgovernment agencies have invested heavily in housing programmes, but such programmes have been intended forhouse-ownership while rental housing has received little in the way of funding and has been left in the hands ofsubnational agencies. In the national plans of most developing countries, tenants and landlords have become"invisible".

The first reason for this is that self-help housing has come to dominate the political agenda, partly because thedramatic growth of this form of housing has clearly worried many politicians and administrators: the thought thathundreds of thousands of people were living in flimsy accommodation on the peripheries of their cities generated fearsabout the threat to public order and health. Recently, however, self-help housing has also dominated the politicalagenda for the opposite reason: as a result of the arguments of writers, such as B. Turner (1967; 1968), self-helphousing has been viewed as a possible panacea for the housing crisis and self-help house ownership has been viewedas a means of pacifying the poor and of providing a cheap method of increasing the size and, sometimes, the quality.of the housing stock. The "slums of hope" were increasingly viewed as a potentially fruitful field for governmentalinvolvement.

The second reason is that most governments are unenthusiastic about dealing with the problems of inner-city"slums of despair". such areas are often in an advanced state of physical decay and contain a high proportion offamilies with severe economic problems. In addition, many occupants are newly arrived migrants who, manygovernments still hope, will return to their homes in the countryside. If this were not enough reason to neglect rentalhousing, undesirable political side-effects might ensue from any attempt to tackle the rental problem: attempts toimprove conditions in inner-city areas might come into conflict with powerful real-estate interests. Elsewhere,antipathy to the rentier class has discouraged governments from introducing programmes which might increaselandlords' profits rather than improve the position of tenants.

It is only very recently that there have been signs of a change in attitude. International attention has onceagain. begun to focus on the rental-housing market. Within the Word Bank, for example, several officials have begunto recognize that rental-housing programmes offer opportunities for improving living standards. (World Bank, 1980,Keare and Parris, 1982; Lemer, 1987) While this argument will take a long time before it is translated into action on theground, the movement has begun.

Page 29


Chapter III. Landlord-tenant relations

The image of landlord-tenant relations derived from inner-city areas of the United Kingdom of the nineteenthcentury is hardly a positive one. On the one hand, exploitative landlords are depicted as constantly trying to evicttenants who cannot pay extortionate rents; oh the other, tenants are frequently "flitting" or are destroying theproperty of the owners. (Daunton, 1987; Englander, 1983; Foster, 1979; Kemp, 1987) Such bad relations were animportant ingredient in the eventual demise of private rental housing in the United Kingdom.

How valid is this view of landlord-tenant relations in the cities of developing countries? Are relations betweenlandlord and tenant generally conflictive or relatively benign? What are the sources of tension between tenant andlandlord, and how are those conflicts resolved? How are tenants selected, and how often are they evicted? Suchquestions are important from a policy perspective, even if we know very little about the answers. If landlord-tenantrelations are generally and consistently bad, it makes little sense to encourage low-income rental housing. It, however,relations are relatively benign, there might be a point in encouraging this form of tenure.

Page 30


Chapter I. The supply of rental housingA. Public-sector housing for rent

Until the 1960s, most housing built by the public sector was rented to occupants. Since then, there has been agradual shift towards selling publicly built housing. In practice, this shift in policy has had little effect on tenurepatterns, since relatively few governments have built housing in large quantities. Only in a limited number of places,such as China, Saudi Arabia, Singapore and Hong Kong and to a lesser extent, Chile, Côte d'Ivoire and India haspublic-sector construction contributed significantly to the production of urban houses. (Grimes, 1976; Tuncalp, 1987)

The early policy of renting out public-sector housing was usually a continuation of colonial practice. This wasvery much the case in former British colonies, where the tradition had been for government to provide houses forprofessional expatriate staff. It was not easy for newly independent governments to change that policy. In Nigeria,"when indigenous government officials replaced their colonial predecessors, this luxury housing was passed on tothem, again as a perquisite of office. It continues to be the main type of housing for the privileged, elite sectors of thepopulation". (Barnes, 1982:6) As O'Connor (1983:175) puts it, "once the local elite have the privilege of highlysubsidized housing of high quality they are reluctant to give it up".

In parts of Africa, public rental housing also derived from a further legacy of colonial rule: the attempt toprevent Africans from migrating permanently to the cities. A plank in this policy had been to prohibit Africans frompurchasing housing. Thus, African workers in Harare. Lusaka, Nairobi, and Johannesburg were required either to livein workers' hostels or to rent government-built homes; by 1963, almost half of Nairobi's population lived in governmenthousing. (O'Connor, 1983: 176) In most African countries, house-ownership was permitted immediately afterindependence, even if renting from State or parastatal organizations continued to be the preferred option of mosthouseholds in a few countries, notably in Zambia. (Sanyal, 1981:419)

Elsewhere in developing countries, a handful of countries were Ideologically committed to producing rentalhousing. In China, for example, most accommodation since 1949 has been built for rent. Since private ownership wasprohibited during the Cultural Revolution, and private renting banned after 1957, the vast majority of urban Chineserent their houses from a municipality, from a collective or from some kind of State enterprise. In Cuba, the urban reformof 1960 turned all rental housing over to the State and forbade private renting: (Gutierrez and others, 1984:257) sincethe revolution, most new houses have been built by state enterprises. (Matthey, 1989)

Finally, special conditions have forced several governments to construct housing for rent in Hong Kong andSingapore, the influx of refugees after the Second World War led to serious housing problems which the authoritiessought to defuse through large-scale public building programmes. For a number of years, most of the new housingwas rented to the occupants. In 1970, 25 per cent of housing in Singapore was rented from the public sector, only 7 percent was made up of public-sector sales. (ISHOC, 1987:57) In Saudi Arabia, a severe housing shortage, combined withsubstantial wealth stimulated the adoption of a massive programme of construction, including both public and privatehousing for rent. (Tuncalp, 1987) Elsewhere, governments, building new cities and satellite towns, have sometimesrented them out, for example, in Chandigarh, Dodoma and Tema. (Sarin, 1982)

With these few exceptions, there has generally been little public-sector housing for rent. In many countries,indeed, there has been little such housing at all. In Jakarta, it made up only 3.7 per cent of the housing stock in 1985; inTurkey, it contributed less than 1 per cent of residential construction in most years since 1963; (ISHOC, 1987:70) and,in Lagos, only about 1 house in 100 has been built by the public sector. Even when public-sector housing has beenbuilt on a large scale, it has generally been sold rather than rented. In the Republic of Korea, most State-developedhousing is offered for sale: "The number of formally rental units produced in Korea over the last decade amounts to ..less than half of one per cent of the total number of approximately 3,000,000 households currently renting". (Ha,1987:105) Despite a recent shift in policy (see chapter IV), rented public-sector housing makes up only 2.4 per cent ofSeoul's total housing stock. (ISHOC, 1987) Similarly, in Latin America and the Caribbean, few governments have builtpublic housing for rent. In Mexico, the State has built or financed large numbers of houses in recent years, but only18,000 units have ever been built specifically for rent. (Garza and Schteingart, 1978; Gilbert and Varley, 1988) InColombia, the National Housing Agency built a few houses for rent during the 1940s but soon decided to convertthem to ownership - a policy that has never been reversed. (Laun, 1977) In Venezuela, public-sector housing has beensold ever since the failures with the superblock programme in the 1950s. (Carlson, 1961)

Even in countries traditionally committed to rental housing, attitudes have recently changed. In China, theGovernment decided to begin sales of public housing, in order to obtain funds to build new accommodation. (ISHOC,1987: 21) Cuba has followed a similar line; since 1985 it has been possible to buy rental property from the State and for

Page 31

houseowners to let their property. Guyana also decided, in 1982, to sell public housing to tenants. (Peake, 1988:138)

Page 32


Chapter I. The supply of rental housingA. Public-sector housing for rent1. Access to public-sector housing

A critical problem with most public-sector housing programmes in developing countries, whether for rental orsale, is that the poor often have limited access to the accommodation. Certainly, this was proved to be the case inAbidjan, (Cohen, 1974: 45) Bogotá, (Laura, 1977) Kampala, (Pell and Sada, 1984:302) Kingston, (Klak, 1989) KualaLumpur, (Wegelin, 1978; Hai. 1983:72) Lagos, (Aradeon, 1978) Lusaka, (Sanyal, 1981:434) Mexico City, (Cornelius, 1975)Manila, (Hollnsteiner, 1974) Mombasa, (Siren, 1978) Nairobi, (Temple and Temple, 1980) Nouakchott, (Iliffe, 1988:241)Rio de Janeiro, (Valladares, 1978) San Salvador, (Grimes, 1976) and Sao Paulo. (Batley, 1983) Only in Singapore can theopposite be claimed. (Field, 1987:156)

The poor who are not excluded by the high cost of the accommodation are eliminated by the nature of theallocation system. Sometimes, this is manipulated politically, often favouring supporters of the government. In oneLatin American country, party affiliation has been a key ingredient in entry to public-sector housing projects eversince the establishment of a democratic system in 1958. In Nigeria, party patronage was the principal basis forallocation during the Second Republic. (Muoghalu, 1987:166) Elsewhere, different vested interest groups pressure theauthorities to allocate a high proportion of houses to their members. In Nairobi, "when medium-cost estates came upfor allocation. council employees often constituted a powerful internal lobby". (Temple and Temple, 1980:246) Even inChina, Badcock (1986:164) notes that "the housing delivery system in China appears to be no less susceptible tomalpractice, nor any more sensitive to needs, than state bureaucracies elsewhere"; there are many examples of bribery,abuse of privilege and incompetent officials.

Even where bureaucratic procedures try to be fair, they often operate against the interests of the poor. In HongKong, small households do not fare well in the allocation system, even it they are very poor. (Drakakis-Smith, 1979:156)In Sao Paulo, the financial constraint on one governmental housing agency required it to establish selection criteriawhich would exclude most of the poor: "while ... is required to maintain a programme for groups defined as unable toattain officially approved housing standards in the regular market, it is nevertheless also required to recoup the costsof the programme from its purchasers". (Batley, 1983:149) in many cities, poor families lose out because public-sectorhousing allocation is linked to employment. In China and Cuba, for example, State employers provide much of thehousing for their workers. In Manila, public-sector workers were particularly well placed to receive houses, and, in onescheme, the Mayor insisted that city workers be ranked first in the allocation process. (Laquian, 1969) Similarly, "inPort Harcourt .. high ranking public servants inhabit highly-subsidized, good quality housing, even though many ownproperty both in the city and elsewhere". (Ekpenyong, 1989:46) Not only does an employer-dominated delivery systemmean that the poor are often excluded, it can also create other kinds of problems. In China, "changing jobs is morepractical and easier than changing houses". (China Daily, reported in Badcock, 1986)

In order to avoid favouritism, malpractice or corruption, some governments have resorted to lotteries. TheColombian housing agency allocated houses in this way between 1982 and 1986, and the same method has been triedin Nigeria. In the latter, however, it hardly improved the position of the poor. As Muoghalu (1987,166) notes,high-income groups obtained as many houses as the poor in two estates in Enugu, and middle-income groupsproportionally more than either.

Even when the authorities manage to get the poor into public housing, the tenants do not always benefit fromtheir supposed good fortune. Thus, in one particular scheme in Durban, "many tenants are pushed into poverty bytheir rental payments". (Corbett, 1982:194) Elsewhere, poor households are forced to sell or rent their houses tohigh-income groups. This has frequently been the case in Bogotá, where the National Housing Agency reports thatsome 35 per cent of apartments in the Ciudad Kennedy scheme are rented out. "The apartments owners were payingthe ICT an average of 632 pesos per month in mortgage installments but were renting their apartments to tenants for anaverage of 1,152 pesos" (Handelman, 1979:14). Similar processes have been reported in resettlement schemes in Rio deJaneiro, (Valladares, 1978) and, in Bangkok, residential rights in highly subsidized rental accommodation are often soldby poor families who cannot afford the rents or the fares to and from work. (Tanphiphat, 1983:112; Drakakis-Smith,1981:125) in Kenya, too, poor tenants often sublet public housing to the better-off (Stren, 1978:223).

Page 33


Chapter I. The supply of rental housingA. Public-sector housing for rent2. Managing public-sector housing

Many governments have experienced considerable problems in managing their rental-housing stock. Indeed,the same problems are constantly reiterated in the literature. A common problem is the limited revenue generated byrents. In China, for example, rental income "is insufficient to cover replacement costs, let alone helping to underwriteexpansion of the national housing programme". (Badcock, 1986:162) In some places, low revenues are due to thedifficulty of collecting the rents. In the 23 de enero superblocks in Caracas, for example, some of the settlers movedforcibly from the ranchos could not afford to pay the rents and maintenance charges. This led to the accumulation ofmassive rent arrears of $5 million and put the whole programme into debt. (Carlson, 1961; Dwyer, 1975:129) Problems ofcollecting rents have also been reported in Colombia. (Laun, 1977:311) In Côte d'Ivoire, there was a rent strike in largepublic-sector housing estates in the late 1960s. (Cohen, 1974:87)

Elsewhere, the main financial problem arises from the low rents charged. In Guyana, minimal rents ingovernmental housing finally convinced the Ministry of Finance that the property must be sold to the tenants (Peaks,1987:138). In Mexico City, both of the agencies operating rental housing charge very low rents. In 1977, the annualincome of one agency represented less than half the costs of running the programme. (Garza and Schteingart, 1978:220-1) Low rents are often the outcome of reluctance on the part of the authorities to take the unpopular step of raisingthe level of payments. In Nairobi, whenever rent rises have been mooted, "the political opposition of tenants has beenresounding". (Stren, 1978:223) In Hong Kong, low rent levels continued in resettlement estates well into the 1970s,partly due to the reaction to the urban riots of 1967. (Drakakis-Smith, 1979:47) However, sometimes, low rents are theresult of inadequate accounting systems: in China, where rents represent only from 4 to 6 per cent of householdincome, "the basis for determination of rents was just the upkeep, maintenance, and management costs. No accountwas taken of depreciation, taxes, land costs, interests or profits". (Kojima, 1987:44) The outcome is obvious: "rentalincome in China at present is insufficient to cover replacement costs, let alone helping to underwrite expansion of thenational housing programme". (Badcock, 1986:162)

The problem of limited rent receipts is among the factors which have convinced many governments to withdrawfrom rental housing. In Mexico, the deficits accumulated by two housing agencies soon convinced the State that itshould never again become a social landlord. (Gilbert and Varley, 1988) In Colombia, the National Housing Agencysoon reversed its rental policy, in part because of tenants' reluctance to pay rent and the agency's inability to recoupthe debts through legal means. (Laun, 1977:311) In Nairobi, "the spectre that haunted officials in the City Treasurer'sDepartment was that a low-cost scheme would become obsolescent before the end of the repayment period". The CityCouncil would then be left with an uninhabitable estate and, therefore, no rents to pay off a massive debt. (Temple andTemple, 1980:238)

While there have been problems in managing rental property, the decision to sell property has more often thannot been based on political grounds. Such a factor was dearly influential in Colombia, when the President declared, in1948, that "the day that a citizen becomes the owner of a house, when he realizes that the walls protect his wife andchildren and that those walls will protect them when he is dead, he is totally transformed and becomes at one withsociety". (Laun, 1977:311) In Singapore, where the Government has given generous incentives to its tenants to buytheir houses, "the political role of the housing programmes has been considered ... carefully and the spread ofhome-ownership throughout the middle echelons of society has played a crucial role in their political castration".(Drakakis-Smith, 1981:128) An ideological element has also operated in the Republic of Korea, where "housing officialsargue that there are not enough long-term capital funds available to produce rental housing. They emphasized that thesale of rental houses produces savings in public expenditure and aids housing authority finance". (Ha, 1987:106)Presumably a governmental commitment to a budget deficit in the housing sector would have solved that problem.

Page 34


Chapter I. The supply of rental housingB. The private rental sector

1. Who are the landlords?

If most rental housing is in the hands of the private sector, it is imperative to know the characteristics of thelandlords. Are landlords predominantly large-scale or small-scale operators? Do they find the activity profitable and,therefore, continue to invest in rental housing? Without understanding the characteristics of landlords andlandlordism, it is difficult to devise an appropriate policy towards the rental housing-sector.

In Latin America, it appears that the large landlord is generally a figure of the past. In Mexico, few large-scalelandlords are reported in Guadalajara, Mexico City or Puebla. (Coulomb, 1989; Gilbert and Varley, 1988) Similarly, inColombia, landlords in poor areas are mainly small-scale operators (Edwards, 1982; Gilbert, 1983), and, in Bogotá, eventhe average high-income rental-property owner owns only one property. (Jaramillo, 1985:115) Even in inner-city areas,where the average number of households per dwelling is quite high, there are relatively few large-scale rentallandlords. Recent research in Bucaramanga, (Edwards, 1981) Caracas, (CPU, 1989) Guatemala City, (Rodas andSugranyes, 1988) La Paz, (Beijaard, 1986) Guadalajara, Puebla, (Gilbert and Varley, 1988) and Mexico City, (CENVI,19139) is explicit in this regard. Certainly, the affluent in Latin America no longer invest in rental housing in the waythey did in the past. Up to the 1940s, the creation or adaptation of housing for rent was regarded as sensible financialmanagement. Investment in rental housing also brought social kudos. (Armus and Hardoy, 1984:40; Scobie, 1974:154)In nineteenth century Santiago de Chile, it was such a respectable business that the Church was an importantlandlord; in Mexico, post-revolutionary leaders invested heavily in urban property. (Violich, 1944; Perlo-Cohen, 1979)

As Latin American cities have grown and the rental population has begun to live in consolidated self-helpsettlements, the characteristics of the typical landlord have changed. The majority of owners seem increasingly to bedrawn from the same strata of society as do their tenants. Thus, in Mexico City, "landlords of the urban periphery are... labourers, employees, traders, bricklayers or artisans"; (Coulomb, 1985a:52) in Bogotá, many landlords arethemselves poor and are renting out one or two rooms in their own houses to supplement their incomes. (Gilbert, 1983)In Puebla and Guadalajara, most landlords are drawn from similar groups to the tenants and own very little in the wayof property. (Gilbert and Varley, 1988) Few landlords seem to own more than a couple of properties.

A similar pattern seems to be characteristic of many parts of Asia and North Africa. In Indonesia, the "familyentrepreneur", who provides only a handful of units, is very widespread. (Nelson, 1988:25) In Karachi, renting In thelow-income suburbs is "imperfect, small scale, illegal and capitalistic". (Wahab, 1984:9) In Ankara and Istanbul,self-help owners soon start to rent out accommodation in the gecekondu, gradually extending their property to createrooms for tenants. (Turan, 1987) In Cairo, "... the owner of a plot may have a ground floor built by a mason room byroom, and occupy one room himself, renting out the others. After saving for several years he will have the first floorbuilt ... this process can extend over eight to twelve years, and allows three storeys to be built". (El Kadi, 1988:31)

In the cities of sub-Saharan Africa, there is evidence of a powerful landlord class, even i most landlords "ownonly one or two buildings". (O'Connor, 1983:191) However, in Lusaka, where admittedly the number of large-scalelandlords is increasing, the average rental house has only between three and four rooms. (Rakodi, 1988:309) Similarly,in Lagos, Barnes (1987:55) comments that "... by concentrating intensively on payment for a houseplot and notconcentrating initially on financing the structure which was to be built on it, people of relatively modest means wereable to enter the ranks of property owners." For Nigeria generally, Okpala (1985:154) argues that "big commercialdevelopers are still relatively few on the scene. Many of today's landlords are still as poor as many tenants."

If most African landlords own few buildings, those properties can still hold many tenants. In a large low-incomeneighbourhood in Lagos in the early 1970s, "an average house had forty-five residents living in thirteen rooms, ofwhich ten to eleven were let out". (Barnes, 1987:58-59) In Kibera, the average house has over 10 rooms, and "theaverage landlord lets over 12 rooms". (Amis, 1987:246) In Latin America, too, there is evidence of some landlordshaving many tenants: in Puebla, one inner-city tenement contains 64 households. While there are some substantialproperties In consolidated self-help areas, the average size is small. (Gilbert and Varley, 1988)

Naturally, the scale of landlordism has some influence on where the owners live. This is an important issueinsofar as n might affect how they behave; many writers have argued that resident landlords tend to behave moresympathetically than those who live somewhere else.

In Africa, Pail (1976:162) suggests that most landlords are resident. particularly small-scale landlords who needto live close by to supervise their property. This seems to be the pattern in Lusaka where Rakodi (1988:309) found that

Page 35

"in 1978 only 15 per cent of buildings were owned by absentee landlords, most of whom (11 per cent) lived elsewherein Chawama...." However, there is plenty of evidence of absenteeism. In Nairobi, the United Nations Centre for HumanSettlements (Habitat) (1987a:61) reported that many owners in the Dandora sites-and-services schemes had rentedtheir houses and moved to irregular settlements nearby. Similarly, in Kibera, Amis (1988:249) notes that "... a high levelof absentee landlordism was found among persons who let out more than twenty units; 64 percent lived elsewhere inNairobi or Kenya, while 14 per cent lived elsewhere within the settlement. This leaves only 22 per cent living with theirtenants as resident landlords". In the shanty towns of Port Harcourt, half of the 93 per cent of households with morethan one household have an absentee owner. (Ekpenyong, 1989:48)

In Latin America, the situation is highly variable. In the centre of La Paz, 30 per cent of owners lived on thepremises while many others left someone in the building in charge. (Beijaard, 1986:66) In the self-help settlements ofBogotá, most landlords owned little property, and more than half lived in the same house as their tenants. (Gilbert,1983:470) In Bucaramanga, "most landlords live on the same lot as their tenants in every type of low-incomesettlement". (Edwards, 1982:147) Similarly, in Santa Cruz, three out of five landlords in self-help settlements live on thesame plot. (Green, 1988:213) While few landlords live with their tenants in Guadalajara, almost one quarter of tenantshave a resident landlord in Puebla. (Gilbert and Varley, 1988:148)

Page 36


Chapter I. The supply of rental housingB. The private rental sector

2. How profitable is rental housing?

The profitability of renting naturally varies from city to city. It is affected by public policy, notably rent control,and by the possibility that the poor can gain access to cheap land on which they can become owner-occupiers. InLatin America, renting no longer seems to be highly profitable, and few businesses now invest in rental housing. InMexico, most landlords claim that renting is unprofitable, and many in Guadalajara and Puebla denounced it as a"business for widows". (Gilbert and Varley, 1988) By contrast, some landlords in Mexico City think that it is profitable,even if they are a minority operating in a consolidated self-help area well located to attract tenants. (Coulomb, 1989)Most Latin American writers accept that public policy, together with a growing range of investment opportunities, hasreduced the attractiveness of rental housing. A few commentators present a different view. In central Lima, althoughlandlords "obtain a relatively low rent for each room, it adds up to a very high sum in terms of rent per square metre".(Padron and Calderon, 1984:178) Similarly, in Sao Paulo, Kowarick and Ant (1988:22) argue that "the intense pressureof demand in these inner-city areas has made tenements an excellent business; if the buildings in question were let forother purposes, they would produce only one quarter of the income from their present tenants'. However, even theyare not saying that it warrants further investment.

In Africa, reports seem to differ on the profitability of rental housing. In Nairobi, there seems to be little doubtthat it is an excellent investment. (UNCHS (Habitat), 1987a; Kabagambe and Moughtin, 1983) in Mathare Valley, "thereturn on capital investment in this tenement housing was spectacular - in the region of 50-100 per cent per annum. It isthis profitability that has transformed the entire provision of low-income housing in Nairobi. ... The transformation thatthe housing companies have effected in terms of housing provision ... has been remarkable". (Amis, 1987:259) In onearea of Lagos, however, renting is clearly not very remunerative; if a few landlords in Mushin became prosperous, "themajority of the owners remained in the low-income class". (Barnes, 1987:61) Similarly, in Ghana, although the powerfulinvest heavily in housing for rent, rent controls mean that it is only profitable to let to high-income groups. For themajority of landlords, subject to rent controls, renting is unprofitable. (Tipple, 1988)

In Asia, there is little information, although, in Indonesia, Nelson (1988:24) argues that "rental units seem to bea good investment, and one in which is a variety of actors are participating. The conversion of a room or constructionof a qubuk is attractive for even the small-scale owner, while the high demand and high rate of return makes itattractive for the larger-scale owner." Even so, it is "very uncommon for individuals to go out and purchase land orunits explicitly to get involved in the rental market". (Nelson 1988:27)

Looking at renting as a business, however, is following a form of logic that might not be shared by everylandlord. In sites-and-services schemes, many poor families rent as a method of repaying the cost of purchase (Keareand Parris, 1982: ix; UNCHS (Habitat), 1987a) Similarly, in Mexico, Coulomb (1985a) argues that there is a stratum oflandlords who are engaged in "domestic renting". Their rationale is not to make profits per se but to supplement theirincome, however minimally. Some landlords in Guadalajara and Puebla favour renting, as it constitutes "a source ofincome that doesn't end" - there will always be tenants requiring accommodation. Such a rationale is especiallycommon among the older generation; the prospect of rent providing an income for one's old age seems still to be animportant consideration not only in Mexico but also in Indonesia and Venezuela. (Nelson, 1988:27; CEU, 1989) Thisgroup of the population knows little about banks and interest rates and does not calculate profits in the same way asthose in business. For many, it is the only "investment" that they know anything about: for older people, oftenmigrants from the countryside, putting their money into "bricks and mortar" to create extra housing space is a logicalprocess.

They might also have other reasons for extending their property. In Guadalajara and Puebla, many small-scalelandlords bulk accommodation for their children: when the children left home, they rented out the spare rooms. Otherswere renting rooms in the expectation that older children would return to live with them. The rationale behind rentalinvestment among the majority of small-scale landlords is clearly very different from that of most people in business.

Page 37


Chapter I. The supply of rental housingC. Low-income rental-housing conditions

1. The public sector

In those countries where public-sector housing is occupied by middle-class tenants, housing conditions are farbetter than those found in most parts of the private rental sector. However, in the few countries where public-sectorhousing has been built an masse for the poor, conditions are often very crowded, and houses lack adequate services.Nowhere is this better demonstrated than in Shanghai, where 90 per cent of housing is publicly owned, 12 per cent ofhouseholds had less than 2 square metres living space per person, and a further 26 per cent between 2 and 4 squaremetres. (Badcock, 1986:161-162)

The way in which public-sector housing is treated by the tenants and the lack of maintenance by the publicauthority are frequently causes of deterioration. In Singapore, some of the earliest flats had sunk into almost slum-likeconditions within 20 years of being bulk. (Hardoy and Satterthwaite, 1981:253) In the early public schemes inVenezuela, conditions long continued to be a problem. (Dwyer, 1975:127) Handelman (1979:15) has commented that lifein the 23 de enero complex, built in the 1950s, "is not pleasant. The elevators are constantly broken or vandalized ....Garbage piles up near many of the houses bringing flies and rats ... and '23' is reputedly one of Caracas' mostdangerous-areas." In Beijing, there are now complaints in high-rise blocks that lifts are out of action and that waterdoes not reach the top floors. (Dwyer, 1987:486) Only in Hong Kong do reports comment favourably on themanagement of estates and the degree to which the inhabitants are satisfied with the housing. (Tung, 1983; Leong andothers, 1983)

Part of the problem with public-sector housing maintenance concerns subletting. Despite being prohibited bymost authorities, this is a very common practice in most estates. Often subletting is a cause of overcrowding - asituation reported in Abidjan, Bogotá and Nairobi. (Stren and White, 1989:56; Laun, 1976:310-311; Temple and Temple,1980:247)

Page 38



2. The private sector

Many private tenants in African, Asian and Latin American cities live in conditions of severe overcrowding. InKumasi, in 1980, the mean occupancy rate in tenements was 3.9 people per room, and more than one family in four wasliving at a density of more than five people per room. (Tipple, 1988:60) Among the ultra-poor tenants of Port-au-Prince,dwellings are "quite small, providing an average of two square metres of habitable area per person". (Fass, 1987:196) inthe inner-city areas of Mexico City, households occupying a single room in 1980 were living at densities ranging from3.5 to 4.4 people per room. (Garcia, Peralta, 1986:289)

If tenants suffer from severe overcrowding, there can also be little doubt that much rental accommodation is ofvery low quality. Often, this is because the owners constructed accommodation as cheaply as they could. In theinner-city areas of most Latin American cities, former elite residences were often converted into multiple-family rentalaccommodation with little or no investment. Even when tenements were originally intended for rent, the aim was tobuild as cheaply as possible. Rental accommodation typically consisted of windowless rooms opening on to a centralpatio area containing only a few taps and a toilet - services intended for all the tenants. This kind of building was thetypical form of construction until the 1940s. Today, it is not common, but similar kinds of tenement are still beingproduced in the consolidating peripheries of large cities. In Guadalajara, new forms of vecindad are appearing inseveral settlements, (Gilbert, 1987) and, in Sao Paulo, houseowners build so-called "sausage" dwellings by adding on"rows of cubicles served by a single toilet, a single washtub and, often, by a single water well". (Kowarick and Ant,1988:23)

Arguably, the worst form of rental accommodation, however, is found where the poor construct their ownflimsy dwellings on rented land. This is a common feature of housing in Lima, Port-au-Prince and a number of Indiancities, including Madras. (Fass, 1978:167; Dietz, 1980:61; Desai and Phillal, 1970) On occasion, such rental plots haveeven been found in the central areas of large cities. In Mexico City, prohibitions on the construction of vecindadades,during the 1940s, encouraged land owners to rent out vacant areas in the inner-city. In Bangkok, too, rented plots havetraditionally been an important source of land for housing, even in the inner-city areas. (Angel and Pornchokchai,1989; Crooke, 1983)

Rental-housing conditions seldom seem to improve over time, mainly because so few landlords or tenants seemto maintain the property. In Kumasi, "since few house-occupants have any financial stake in the house and rentpayments do hot begin to cover maintenance costs, little regular repair is done". (Tipple, 1958:81) It is often argued,indeed, that some owners are hoping that their property will fall down, so that they can realize the value of the land onwhich the tenement is falling. Evidence of this is apparent by the behaviour of landlords in Guadalajara and Puebla.(Gilbert and Varley, 1988; Marroquin, 1986) In Bombay, too, rent controls produced such a level of deterioration that"between 1967 and 1979 as many as 4328 houses collapsed, resulting in 3,114 casualties, of which 382 were fatal".(official Bombay report cited in Muttagi, 1988:74)

However, there is considerable variation within the private-rental sector. Rich tenants obviously live in muchbetter conditions than do poor. Often within the same settlement, some families live in decently serviced apartmentswhile others subsist in severely overcrowded tenements. In the rental tenements of the inner-city areas of mostdeveloping countries, there is much more overcrowding and much more deterioration than in the consolidatingself-help settlements.

Page 39



3. Housing conditions of tenants compared w ith those of ow ner-occupiers

It many tenants live in bad housing conditions, it should not be assumed that they always occupy worseaccommodation than other households. In some cities, too much attention should not be directed towards theproblems facing tenants, because all poor households tend to live in poor conditions. Thus, in Lagos, three out of fourhouseholds occupy one room, and the average occupancy is 4.1 people per room. (Okpala, 1985:146) In Bombay, in1971, 77 per cent of households lived in one room, at an average density of 5.3 people. (Misra, 1978:375-378)

However, it is probably true that tenant households occupy less space than do most owner-occupiers. Thus, infour settlements in Bogotá, 62 per cent of tenant households occupied only one room, compared with 27 per cent ofowners. (Gilbert, 1983: 470) Although tenants had slightly smaller families than owners, this does not compensate forthe difference. In Guadalajara and Puebla, the pattern is by no means as clear, since the age of the settlement is animportant influence on house size. Thus, owners in old settlements occupy much more space than tenants, but ownersin new settlements have much less space than owners in older areas. (Gilbert and Varley, 1988:121-122) In Kumasi,many owners seem to have more space than most tenants: however, owners who have many kin staying with them arescarcely better off than most tenant households, and, in any case, the differences between tenure groups are nogreater than the differences between different kinds of housing. In Cairo, the expected relationship between tenure andcrowding simply does not hold; owners live 1.8 people per room compared with the 1.9 of tenants. (Abu Associates,1982:121)

In terms of services, there is also often very little difference between owners and tenants. In Kumasi, although"house owners tend to enjoy better levels of service than tenants and subtenants... the differential attached toownership is not very great". (Tipple, 1987:73) In Bogotá, Hamer (1981) found that tenants had almost identical accessto electricity, water and sewerage with that of owners, and Lamer (1987-30) reports a similar finding for Cali.

It is also, sometimes, the case that owner-occupiers have less access to services than do tenants. Tenants tendto move into areas which have been in existence for some years: not only are the buildings in such areas relativelyconsolidated but they also tend to be serviced. By contrast, marry recent owner-occupiers occupy plots on peripheralland which has not yet been supplied with services. In Guadalajara and Puebla, the owners of houses in new peripheralsettlements lacked piped water and, sometimes, electricity - services which nearly all tenants possessed in other partsof those cities. (Gilbert and Varley, 1988) Lemer (1987:30) presents data for Cairo and Manila which show that, at similarlevels of income, tenants generally have much higher access to water than owners.

It would be unwise, therefore, to make too much of the distinction between the housing conditions of ownersand tenants, because of the huge diversity within each category. As Tipple (1988) observes for Kumasi, the city sectorin which the housing is located "is a stronger determinant of servicing levels than tenure". No tenure group in eitherthe tenement or indigenous housing areas has a level of service provision approaching that or any tenure group livingin governmental or high-cost housing.

Table 3

Figure 1

Page 40


Chapter II. The occupants of rental housingA. Are tenants poorer than owners?

Tenants are drawn from every level of society, and few generalizations are satisfactory. Not all tenants are poor,not all tenants are poorer than their landlords, and not all tenants live in similar kinds of accommodation. Figure 2 ismerely a reminder, based on data from several cities, that tenants are to be found among every income group: there arerich tenants as well as poor.

It is also clear from figure 2, however, that a high proportion of poor households rent accommodation - ageneralization that normally holds whether the society is affluent or poor. In the United States of America for example,"in 1980, 67,6 per cent of all renters - compared with 37.1 per cent of all home owners - had household incomes below$15000". (Dreier, 1984:261) At the other end of the income scale, in Calcutta, rich households tend to own while verypoor households are likely to rent. (Pugh, 1988) In cities in other developing countries, the pattern is similar. In Harare,"the remaining types of accommodation for rent bridge the divide between legality and illegality and are moreimportant in meeting the needs of the poorest than owner occupation" (Rakodi, 1989:10). In Santiago, Soto (1987:106)notes that roughly two thirds of those who rent a room are in the bottom two quintiles of the income distribution. InKumasi, "owners have higher household expenditures than tenants ... roomers - who are just a majority of households... include many of the poorest and least advantaged". (Tipple and Willis, 1989: 28)

Even if there is a tendency for the number of tenants to include a higher proportion of the poor compared withthat of house-owners, it certainly cannot be assumed that all tenants are poor. In Nairobi, Amis (1988:250-251) pointsout that "... it would be a mistake to assume that all the tenants are on the breadline; some hold down middle-rankingclerical jobs but choose to live in such areas to minimize their expenditure on housing". Clearly, there is a great deal ofheterogeneity within the tenant population. As Tipple and Willis (1989:17) have observed, "... differences... are morepronounced in Kumasi between households with different access to services than between owners and tenants."Similarly, Gilbert and Varley (1988) have found that many tenants in inner-city areas are more prosperous than mostowners in recently established self-help settlements. In Bucaramanga, Edwards (1982:143) has found many examples ofwhat he calls "upwardly mobile tenants" households that reject the feasible option of house-ownership.

In a few cities, the affluent are more likely to rent accommodation than the poor. In Cairo, for example, richhouseholds are less likely to own than poor households: "... the first quartile of the income distribution has 38 per centowners while the highest quartile has only 29 per cent". (Abt Associates, 1982:125) Similarly, in Nigeria, the tenants ofpublic housing tend to be drawn from the professional groups, (Poll and Sada, 1984) and, even in the private sector,"many landlords are still as poor as many tenants... many tenants are richer than their landlords". (Okpala, 1985:154)

There is a possibility that house-owners are poorer than tenants In those cities where the very poor are forcedinto squatting by their inability to afford rent. (Sachs, 1983) While this is far less common than many writers claim, it isyet another deviation from the expected income/tenure relationship. Perhaps, it is merely part of the same phenomenonas the poor who are forced to sleep on the streets because they cannot pay rent.

In sum, therefore, income is clearly an important influence on tenure but is in no sense the determining factor.There is great diversity among both tenants and owners; there are rich and poor tenants and rich and poor owners.Equally, at similar income levels, some households rent while others own.

Page 41


Chapter II. The occupants of rental housingB. Other socio-economic characteristics of tenants

If tenants are heterogeneous in terms of their incomes, it is perhaps not highly surprising that they show equalvariation with respect to their other socio-economic characteristics.

First, tenants do not seem to be drawn predominantly from the ranks of migrants; indeed, there seems to be noclear relationship linking tenancy with migration. Admittedly, there is evidence that the Turner (1968) bridgeheadermodel still applies broadly to migrants who have recently arrived in the city. In La Paz and Ciudad Juarez, Hoenderdos,Lindert and Verkoren (1983:381) have concluded that "where the migrants' first accommodation is concerned, rentaland free accommodations were of utmost importance". Similarly, in Bogotá and Mexico City, Gilbert and Ward(1982:137) found that very few tenants acquired their own house immediately on arrival, the vast majority renting orsharing with kin. In Bolivia, Côte d'Ivoire, India and Indonesia the higher proportions of recently arrived migrants wereto be found among tenants. (Green, 1988; Cohen, 1974:46; Gupta, 1985:49; Nelson, 1988:18-22.) As time passes,however, migrants tend to become house-owners. In Bogotá and Mexico City, a majority of self-help owner-occupiersare migrants, and, in La Paz and Ciudad Juarez, Hoenderdos, Lindert and Verkoren (1983:381) argue that "the longerone has lived in the city, the better the dream of owning a house can be realized!".

There is even the possibility that migrants move into house-ownership in peripheral settlements more readilythan city residents. In Guadalajara and Puebla, for example, owners in self-help settlements are more likely than tenantsto have been born outside the city and, in particular, to have been born in the countryside. (Gilbert and Varley,1988:133)

Migration, therefore, seems to have an influence on original tenure in the city but gradually diminishes inimportance over time. It is also clear that migrant newcomers no longer settle predominantly in the inner-city, aspredicted by the original Tumor model. Today, most move into accommodation in the consolidated periphery of thecity. (Bahr, 1986; Conway and Brown, 1980; Gilbert and Ward, 1982; Hoenderdos, Lindert and Verkoren, 1983)

Secondly, age is an unreliable indicator of household tenure, although it is often true that young families rentmore frequently than old families. Thus, in Bogotá, owners in five peripheral settlements were nine years older thantenants and, in Mexico City, they were seven years older. (Gilbert and Ward, 1985) Similarly, in Kumasi, owneroccupiers are older than tenants (Tipple and Willis, 1989:17). In seven towns in the United Republic of Tanzania, 56 percent of those over 60 years-of-age owned a house, compared with only one quarter among the 35-44 year age group.(Poll and Sada, 1984:296) It young people tend to rent, however, many old people can also be found among the tenantpopulation, particularly in old inner-city areas. Such a pattern has been found in Buenos Aires, Bucaramanga and, to alesser extent, in La Paz. (Rivas, 1977; Edwards, 1982; Beijaard, 1986)

Thirdly, although there is some evidence that size of family is linked to tenure, it is less important than mostwriters would predict on the basis of developed-country experience. As Jackson (1973:89) notes in his study of MexicoCity, "... one of the most significant findings of this study is that the life cycle changes account for little of the mobilityin the sample ... increase in family size were not related to mobility." On the whole, however, households of only one ortwo people tend to rent, and large families tend to be owners. This might be linked, of course, to other factors, such asthe age or income of the household head, but it is not wholly surprising that, in cities where self-help ownership offersmore space than rented accommodation, large households move into ownership. Thus, in Bogotá and Mexico City,owners' households contained one person more, on average, than those of tenants. This is also true in Kumasi,although there the reason is very different; household size and overcrowding both rise because migrants from thecountryside join their owner-occupying kin in the city. (Tipple and Willis, 1989)

Fourthly, there is some evidence that female-headed households are highly concentrated in rentalaccommodation. In one Lusaka shanty town, "a much higher proportion of female than male household heads in theupgraded area were tenants (76 per cent compared to 57 per cent)". (Rakodi, 1988:309) In Mexico, Chant and Ward(1987) have argued that women-headed households tend to be excluded from ownership in State housing schemes. InLa Paz, Beijaard (1986:50) found that about half of the tenant households he visited were headed by women, and, inGuadalajara and Puebla, the higher proportion of female-headed households was found among tenants (Gilbert andVarley, 1988). In Africa, Peil and Sada (1984:297) claim that women are unlikely to become house owners. Certainly, inLagos, "women do not purchase real estate in proportion to their numbers; less than one in six Mushin houses isowned by a woman". (Barnes, 1982:18)

Finally, there is some suggestion that ethnicity is a factor in tenure choice. In cities of West Africa, such as

Page 42

Kano, Ibadan and Zaria, indigenes tend to own property, and migrants from different racial groups tend to rentaccommodation in separate areas (O'Connor, 1983:171-172). In many parts of eastern and southern Africa, of course,the relationship between race and tenure became accentuated during the colonial period; Europeans owned theirhouses, and Africans were not allowed to. Today, the distinctions are still present but much less marked than before.In Latin America, there is little information, although, in Santa Cruz, Indian migrants tend to rent, whereas nativemestizos tend to own. Whether this is due to racial discrimination, differences in income or recency of arrival, however,is uncertain. (Green, 1988:347)

Clearly, therefore, there are few general rules. Tenants can be found among all income groups, at any age,among different kinds of household and among large families of all sizes. We also find different kinds of tenants indifferent kinds of rental property.

Page 43


Chapter II. The occupants of rental housingC. Methods of explaining tenure choice

1. Choice versus constraint

Why is it that, in most African. Chinese, Indian and Korean cities, the vast majority of people rentaccommodation whereas the majority in West Asian or Latin American cities tend to own? In the literature, there aretwo classic approaches to explaining household tenure. The first is based on household choice, emphasizing thathouseholds with different characteristics choose different kinds of accommodation. The second tends to emphasizethe constraints facing households in their choice of accommodation and use the structural features of the city toexplain household tenure. In practice, both approaches are valuable.

It is generally argued that the nature of the local land market is a critical influence oh access tohouse-ownership in self-help settlements. (Gilbert, 1981; 1983) As a general rule, cities where land is expensive ordifficult to occupy (because of, say, police action) have higher levels of tenancy than those where land invasions arecommon or very cheap plots are available. This appears to be the pattern in Colombia (Gilbert, 1981; Edwards, 1982)and in Mexico. (Gilbert and Varley, 1988) Colombian cities, such as Bogotá and Medellin, where land in illegalsubdivisions is quite expensive, have much lower rates of ownership than other Colombian cities, where landinvasions tend to be common. Similarly, in Mexico, high levels of tenancy in Guadalajara and Puebla, where landinvasions are rare, contrast markedly with the situation in northern cities, such as Chihuahua, Culiacan and Hermosillo,where invasions are much more frequent.

In Santiago de Chile, the prohibition on invasions since 1973 has had a clear impact on housing tenure andliving conditions in general. In a city where invasions became the dominant form of low-income land acquisition duringthe 19605 and early 19705, the prohibition had a great impact on the structure of housing. (Soto, 1987) Bahr andMertins (1985) note how many families in that city share accommodation with kin, and Necochea (1987) contrasts thesituation in 1973, when most families in Santiago had a house or plot, with that in 1985, when possibly 42 per cent offamilies were sharing houses.

There is also evidence in support of this kind of argument from other continents. In Egypt, Abt Associates(1982:125) argue that "patterns of ownership appear to be strongly influenced by housing market conditions,especially housing costs. Cairo, which has higher costs of ownership than Beni Suef (mainly because of higher landcoats) has comparatively fewer owners". Similarly in Nairobi, Amis (1987:260) links shifts in tenure patterns withchanges in government policy towards land acquisition. "The administrative limitation of self-help housing has had acrucial impact in Nairobi by forcing individuals into the housing market ... The commercialization of unauthorizedhousing combined with the decline in employee housing means that since the mid-1970s rent has been the soledeterminant of housing eligibility or entitlement."

Ideally, it should be possible to check some of these ideas by comparing the cost of house and land purchasewith the cost of renting in different cities. In practice, there are too few data published for different cities to be able todo this. However, data collected by Gilbert and Ward (1985) for Bogotá, Mexico City and Valencia suggest that there isa clear negative relationship between land prices and levels of house-ownership in those cities. Similarly, data on theimputed costs of renting and ownership, produced by Malpezzi and Mayo, (1987a) provide at least prima faciesupport for such a relationship.

Table 4 shows the rent-to-income ratios of owners and tenants in a number of different cities. It shows thatowners spend much more of their income on housing than do tenants in Bangalore, Busan, Cairo, Manila and Seoul.By contrast, they spend about the same in Bogotá and Cali, and much less in Davao. While differences in the desirefor ownership certainly explain some of the variation, it is likely that the ratios strongly reflect differences in relativecost of ownership in each city. Where house-ownership is very expensive relative 10 rents, the ratio will be high andfew families will be owners. Certainly, it cannot be coincidence that the proportion of owners to renters is very low inBangalore, Busan, Cairo, Manila and Seoul and high in Bogotá, Cali and Davao.

The difference between the two Philippine cities is particularly revealing. In Manila, where the expenditure ratiois 3.0, only 33 per cent of households are owner-occupiers. By contrast, in Davao, where the ratio is 0.5, 73 per cent ofhouseholds own their own house.

Page 44



2. Is there a general preference for ow nership?

What this argument does not allow for, however, is the possibility that most tenants are content to remain astenants. Certainly, the literature has pointed out that, where circular migration is common, many town dwellers do notwish to own. In Indonesia, for example, many migrants have no intention of staying or owning in the city - in fact,many already have a house or land in their village. (Nelson, 1988:18) Similarly, most interpretations of the situation inAfrica explain the high incidence of renting in terms of migration. Thus, O'Connor (1983:185) argues that "far moreurban dwellers than in Latin America or south-east Asia prefer to rent accommodation, because they do not intend tostay permanently, and because entrepreneurs have made this available, sometimes at very low rents." This certainlyappears to be the situation among well-housed tenants in Kumasi where "... renters are remarkably similar to ownersand appear to find their tenure a valid alternative to owning even for the highest income groups". (Tipple and Willis,1989:28) They rent because "tradition demands that everyone who can should build a house in his village rather thanin the city where he works. While rents are low, few households will regard the large amounts needed to build andmaintain a house in Kumasi to be a worthwhile alternative to virtually free tenancy". (Tipple, 1987: 82-83)

In Nairobi, however, a similar desire to return home is combined with high rent levels. In this case, there can belittle doubt that many of the poor would be keen to build their own self-help house if only they could gain access toland. There is certainly no reason to suppose that the eventual desire to return to the countryside is incompatible witha strategy of owning property in the city. A property in the city could be sold when the household wished to return"home". Such a strategy is strongly discouraged in Nairobi, however, by both the difficulty of squatting and the costof purchasing land.

Whether or not most households south of the Sahara wish to own in the city, there is no doubt that the vastmajority of urban dwellers elsewhere express a strong desire for house-ownership. In Egypt, 91 per cent of ownerhouseholds in Cairo and 100 per cent in Beni Suef "would prefer to own rather than rent", and, among tenants, thefigures are 75 percent and 82 per cent, respectively. (Act Associates, 1982:123) In Karachi, the poor prefer to own,even if it is seldom possible because of rising land prices. (Wahab, 1984) In Mexico, 96 percent of tenants interviewedin Guadalajara and 93 per cent of tenants in Puebla said that they would like to own a "home". (Gilbert and Varley,1988:144) In Santa Cruz, 95 per cent of tenants thought that owning was preferable to other options. (Green, 1988:252)Similarly, "there can be no doubt that all Brazilian workers aspire to buy their own home". (Kowarick and Ant, 1988:65)

Of course, general statements about how nice it would be to own must be taken with some caution. The factthat households wish to own does not mean that they will become owners even when they have the opportunity.Thus, in Cairo, "ownership status, despite being preferred by households, is not positively related to income". (AbtAssociates, 1982:125)

Clearly, housing choice is rarely simple, and the trade-offs are often very complicated. In practice, while incomeis an important ingredient in household choice, families spend money in different ways.. Also, not every family is inthe same position to take up a particular option. Thus, self-help ownership might be available to every family above agiven level of income, but many might not be prepared to live in settlements without regular services. Others might beill-equipped to undertake the process of self-help construction, particularly female-headed families and families with noexperience of building. As Jackson (1973:99) has argued, "it could be that the colonia proletaria life-style offersspecial attraction for those with more rural backgrounds". Much also depends upon the locational factors involved. Ifthe choice lies between ownership in a distant, peripheral self-help settlement and renting in an inner-city location, theoption of ownership might be refused. This is certainly a common pattern among some tenants in central Montevideo,where Benton (1987:49) argues that "residents' attachments to the city center were not only economic but also werebased on strong sentiments of neighbourhood loyalty and tenement solidarity ... these families did not view squattingas an acceptable alternative." In Buenos Aires, Rivas (1977:81) claims that tenants living in the inner-city areas areculturally averse to living in self-help housing. Similarly, in Guadalajara, Puebla and Mexico City, there is plentifulevidence of people wishing to remain in the inner-city area rather than locate in the distant periphery. Many of thesepeople have always lived in the inner-city, and an alternative location is simply not acceptable. (CENVI, 1989; Gilbertand Varley, 1988)

Such reluctance to move to the periphery might be linked to feelings of attachment to the neighbourhood andto a well-established network of friends. However, preference for the inner-city might be based equally well onstraightforward common sense. As one tenement dweller In Sao Paulo put it, "The advantage here is that you can walkto work. Since I moved here eleven years ago, I've never had trouble with getting to work late and having my paydocked. And I don't get home worn out either. You know, when I moved I'd already started making payments on a

Page 45

piece of land..,". (cited in Kowarick and Ant, 1988:67)

A further factor in the decision to rent might be that rent levels relative to incomes are very low. Such asituation might arise because of rent controls, because there is an excess of urban accommodation, becauseaccommodation for certain groups is subsidized or because housing conditions are inadequate. In Zambia, employeeswith high salaries remain in rental housing because of generous subsidies. "The subsidized housing is a uniquebenefit: not only does it increase in absolute magnitude with increases in income, but also as a proportion of income,its increase is substantial". (Sanyal, 1981:436) Rent controls can also maintain the attractiveness of renting by creatingvery low rent-income ratios; in Kumasi, the vast majority of households paid less than 10 percent of their income onrent. (Tipple, 1988: 40) Poor housing conditions can also have the same effect. Thus, in Kisumu, Kenya, most of thepoorest group spent less than 15 per cent of their income on rent and only 8 per cent were forced to spend more than35 per cent. (Macoloo, 1988:167) In Guadalajara and Puebla, a combination of factors, including rapid inflation, meantthat rent-income ratios among low-income families averaged only 13 per cent. (Gilbert and Varley, 1988:157) Whateverthe precise cause of relatively low rents, limited rental payments might convince some families to continue in rentalaccommodation despite overcrowding and poor housing conditions, rather than facing the difficulties and/or expenseof becoming self-help owner-occupiers.

What this evidence suggests is that tenure is merely one factor in household choice. What might be vital in thefinal analysis is not the choice between owning and renting but the choice between the kind of renting and the kind ofownership. Some tenants might only opt to take up a particular kind of ownership.

Page 46



3. Why is there a w idespread desire for house ow nership?

Although some households continue to rent when they could afford to own, in most cities there is a strongdesire to own a house. Indeed, the desire for house ownership is probably on the increase.

First, a house purchase can be a good investment. In the United Kingdom, and in some parts of Latin America,middle-class house owners can credit mortgage payments against their annual tax bill. As house and land valuesgenerally rise faster than inflation, house purchase offers a secure financial investment. In recent years, buying a homein the United Kingdom has been a better investment than purchasing shares. Even self-help ownership creates certainbenefits in terms of capital accumulation; land and house prices increase particularly as service levels improve.Without a plot of land. many tenants feel that their rental payments are simply lost. Tenants throughout the worldfrequently express the view that rent is just "money down the drain". (Gilbert and Varley, 1988; Saunders, 1989)

Secondly, buying a house allows families to increase their total income. Families can easily run a business fromhome. Indeed self-help settlements are full of grocers, cafes, workshops and home businesses. (Eckstein, 1975; Gilbert,1988; McCallum and Benjamin, 1987; Sarin, 1982; Strassman, 1985) House-owners can also rent out parts of then'accommodation to tenants. As Peil (1976:162) argues, in all parts of Africa, migrants aspire to build or purchase one ormore houses for subsequent rental".

Thirdly. ownership seemingly offers households security and autonomy. Tenants are relieved from the fear thatthey will have to move if they cannot pay the rent. Once a house is owned, loss of employment is much less of a threatto the family; a temporary crisis can be weathered without the constant threat of eviction. Ownership also provides agreat sense of stability, because most owners live in their houses longer than most tenants. Thus, in Santa Cruz, Green(1988:253) argues that "everyone wants to own to escape from wandering from house to house like a gypsy, ...Ownership is a resource which aids survival, and therefore brings with it a degree of peace of mind". There is also thebenefit that comes from not having a landlord checking on what one is doing to the property.

Finally, personal identity is inherently linked in many societies to the tenure of the house. "The desire to ownone's home is overwhelmingly strong in all classes and all regions of England, and this cannot simply or evenprincipally be explained by financial considerations. The home is the core of most people's lives, and to own that homeis at the centre of most people's aspirations and values". (Sounders, 1989:191) increasingly, house-ownership is animportant ingredient in establishing social rank and status. In many parts of Africa, for example, Peil and Sada(1984:284) note that "house-ownership is a sign of success wherever the location and whatever the condition." As oneNigerian interviewee put it. "Where a man rents he considers himself small". (cited in Barnes, 1987:69)

Increasingly, too, "the home is not only itself an object of consumption, but it is also the container withinwhich much consumption takes place". (Sounders, 1989:177) Ownership of furniture or television is set against theenvironment in which it is placed: even the appearance of a car is improved when it is set against a house with agarage and a driveway. Such feelings are boosted by the general attitude in most capitalist societies that it is up to theindividual to house his family adequately. "Such thinking receives constant reinforcement from the government, fromprivate builders, from advertising in the mass media, and from other sources, until the individual comes to believe thatunless he possesses his own house (and lot), he has not provided sufficiently for himself and his family, and that hehas in some vague way, not done his proper share for himself or for his country ...". (Dietz, 1981:40)

If most house-owners gain many advantages over tenants, the degree of advantage is highly variable betweensocieties and even within the same society through time. Thus, as Kemp (1982:4) points out, "to simply categorizeoccupiers as either 'owners' or 'tenants' is to ignore an important element of complexity and variations. Hence, tenuresshould rather be viewed as bundles or configurations of property rights and obligations, the precise mix of which isliable to some variation, albeit within limits. These configurations are not immutable but vary over time and space."Society and, in particular, State policy determines the sets of advantages and disadvantages attached to differentforms of tenure. As a result, in certain societies, many of the advantages which normally accrue to house-ownersapply equally well to tenants.

With respect to security of tenure, for example, chapter III shows that tenants in some cities have lived for 20years or more in the same accommodation. In some countries, the customary advantages of ownership, in terms ofsecurity, simply do not apply. In China, for example, "households renting from a Municipal Housing Bureau or acollective enjoy substantial property rights and security of tenure. They may continue to occupy a dwelling after

Page 47

retirement, and with the death of the household ... (the right to tenure) is normally transferred to next of kin". (Badcock,1986:162) in other societies, it is by no means certain that ownership offers complete security. While families arepaying off a mortgage, there is always the danger that household income will fall, so that repayment becomes difficult.In Sao Paulo, this was a problem among the purchasers of public housing during the mid-1970s; numerous familiesabandoned the property to escape their accumulated debts, and others were simply evicted. (Batley, 1983:147)Similarly, in the United Kingdom, mortgage defaults have risen dramatically in recent years; in 1988, more than 16,000houses were repossessed, compared with only 2500 in 1979. (Booth, 1989)

The virtues of house purchase as a form of investment, therefore, depend greatly upon what is happening toland and house prices and upon the relative subsidies given to owners and tenants. 1n the past, council-housetenants in the United Kingdom received generous subsidies, but, in recent years, most of the State subsidy has beenshifted to house-buyers, through increased tax relief on mortgages. In China, however, "rents are heavily subsidizedby the State ... The housing subsidy varies from city to city, but can be up to 60% of the rent". (Badcock, 1986:162)The advantages of house-ownership are reduced in consequence.

What determines the benefits to be derived from ownership or tenancy? The general answer is State policy. Insome societies, the balance of advantage is deliberately sipped towards owners and elsewhere it is Tipped towardstenants. The precise balance depends on the ideology and practice of the State, but, in general, socialist societiesfavour tenants in public housing, and capitalist countries favour individual house-ownership. In China and Cuba,private house-ownership has only been encouraged very recently. By contrast, in Peru, "almost no effort goes intodeveloping communal housing... Rather, public and private authorities - and their various clienteles - all direct theirenergies toward the provision of individually and privately owned dwelling units, often the single family type". (Dietz,1981:40)

Page 48


Chapter III. Landlord-tenant relationsA. The selection of tenants

How tenants are selected gives a clue to many of the sources of conflict between landlords and tenants. In fact,remarkably little is known about the process of selection.

In low-income areas of Guadalajara and Puebla, landlords seem to be very fussy about the kinds of tenant theyare prepared to accept. They do not like big families, households with animals, single people or anyone with a badreputation. Many, however, admit that their preferences are limited by the fact that some tenants lie about thecharacteristics of their family. In practice, too, there is considerable variation in landlord behaviour. Some checkcredentials very carefully and demand the name of a guarantor who will pay the rent in case of default; others merelygive the tenant "a quick look over". Opinion is also very mixed about the virtues of taking in a tenant who iswell-known to the owner. Some believe that a close personal acquaintance inhibits a commercial relationship; others,however, like someone who comes with a recommendation from a friend or an existing tenant. Perhaps, the situation isbest summed up by Coulomb's summary of attitudes in Mexico City; the best tenant is "an acquaintance but not adose acquaintance". (CEU, 1989:61) in Indonesia, tenants are usually known to the landlord and "in many cases ...originated from the same village or city". (Nelson, 1988:33) This familiarity between landlord and tenant is regardedvery favourably and usually eliminates the need for any other kind of vetting procedure.

Ethnicity seems to be a factor of no importance in some cities but to be very important in others. In Karachi, forexample, people of similar ethnic origin tend to live together, and landlords tend to look for tenants from among thepeople they know. They are encouraged to do this, because they believe that the tenant will be likely to followcommunity rules about acceptable behaviour - rules which usually act in favour of the landlord. (Wahab, 1984:12) Bycontrast, certain African landlords follow the opposite course, looking for tenants from other ethnic groups whocannot mobilize communal pressure for Improved housing conditions. (Poll and Sada, 1984:300) in Nairobi, Amis (1984)notes that landlords choose people from different ethnic groups, and, in Lagos, 82 per cent of the houses in Mushinhave tenants from more than one ethnic group. "Private owners prefer to rent to members of ethnic groups other thantheir own because ... it is easier to collect rents from those to whom one is not close". (Barnes, 1982:16)

Page 49


Chapter III. Landlord-tenant relationsB. Mobility of tenants, security of tenure and eviction

In the United Kingdom of the nineteenth-century, working-class tenants moved frequently and slum tenants"perpetually". (Kemp, 1987:8) In London, one third of working-class families moved every year. (Foster, 1979)Unfortunately, not very much is known about the situation in the cities of most developing countries. The informationthat is available indicates a great deal of variation.

In African cities, such as Lagos, Lilongwe and Nairobi, relatively frequent moves are common, with tenancies inthe Kibera area of Nairobi averaging 2.6 years. (Amis, 1984; Poll and Sada, 1984:300) in Cairo, however, rent control hasencouraged long tenancies, on average over 12 years. In Indonesia many landlords prefer tenants to stay for longperiods because of trust and friendship. (Nelson, 1988:35)

In Latin America, the situation seems variable. In Bogotá, Bucaramanga and Santa Cruz, tenancies seem to berelatively short, averaging between 1.9 and 2.5 years. (Gilbert, 1983; Edwards, 1982; Green, 1988) in Caracas, however,mobility in the consolidated self-help settlements is limited, and some tenants remain for 20 years or more. Whiletenancies are fluid in the inner-city, with some tenants only slaying a few months, many have lived for 15 years or morein the same house. (CEU, 1989:19-20) Similarly in Guatemala City, 60 per cent had lived more than five years in the sameroom. (Rodas and Sugranyes, 1988:8) in Guadalajara and Puebla, the average tenant does not move frequently; theaverage stay in the current house is eight years, in the previous house over four years. Among tenants living in theinner-city, occupancies are even longer than that, averaging 11 years. (Gilbert and Varley, 1988:183) Great stability isfound in central Mexico City, where rent control has discouraged many tenants from moving. In the Guerrero district,60 per cent of one sample had lived more than 10 years in the same vecindad, and as many as half had lived there formore than 20 years. (CEHAP, 1986:180) Similar levels of immobility have been found in other surveys in the centralareas of that city; in one set of vecindadades, the average tenancy was 18 years, in another, it was 22 years. (CENVI,1989) it is quite clear that most Mexican tenants do not move frequently.

Similarly long stays have been observed in the central areas of cities in other parts of the world. In the slums ofOld Delhi, for example, 40 percent of tenants have been in their present dwellings for more than a decade, "27 per centfor more than 15 years". (Desai and Pillal, 1970:209) it is possible, therefore, that inner-city tenants cling to theiraccommodation for longer periods than those in peripheral settlements. This might be linked to rent-control legislationor to the advantages of inner-city location.

Page 50


Chapter III. Landlord-tenant relationsC. Conflicts between tenants and landlords

Landlord-tenant relations seem to vary considerably from city to city. The literature certainly records evidenceof hostility in many places. Thus, in Baldia township, Karachi, Wahab (1984:19) speaks of the lack of consideration oflandlords and the "tense relations between them and their tenants". However, in Indonesia, landlord-tenant relationsare frequently amicable, being based on family-like principles of helping one another. (Nelson, 1988:32) incommercialized settlements in Nairobi, "landlord and tenant relations are generally hostile". (Amis, 1988:251)

Relations between owners and tenants seem good in Latin American cities. In Guadalajara and Puebla,landlord-tenant relations are not generally conflictive (Gilbert and Varley, 1988:192) Certainly, very few tenants reportthat they left previous houses because of bad relations with the landlord, a finding supported, of course, by longaverage stays. Some tenants even speak well of their landlords. In Caracas, too, the rental relationship is frequentlyamicable: "what most characterizes relations between owners and tenants, especially in the self-help settlements, arethe informal legal arrangements, the 'solidarity', and the mutual trust, which contribute to the 'good' relationship,'without problems'". (CEU, 1989:25) in Santa Cruz, relationships are generally easy but the reasons are rather unique.Green (1988:225) argues that, because hyperinflation in the mid-1980s undermined renting as a business transaction,"46 per cent of landlords claim that the principal motive for continuing to rent is not economic. Of these, 62 per centsay that companionship is more important than money." Loneliness and fear of burglary are why some continue totake in tenants.

Of course, the relationship might be different according to whether the landlord lives on the premises or riot. InCaracas, CEU (1989) suspects that large, commercially-oriented landlords are the most grasping. In Nairobi, theopposite seems to be the case: "paradoxically, it seems that the smaller-scale landlords ... are the harshest ... But forthese landlords, their tenants' rent is often their primary source of livelihood". (Amis, 1983:251) in Karachi, relations aretense, in large part because the small-scale landlord fears that he will be found to be renting out illegally.

Where conflict occurs, it is most likely to arise over three main issues - the physical state of the building, thelevel of rent and threats of eviction. In Kibera, Amis (1988:251) claims that "... landlord-tenant relations center on themonthly rent payment, for which the ultimate sanctions for nonpayment are physical violence and immediate eviction.Examples of both are fairly common, and horror stories of such behaviour abound". Similarly, in Santa Cruz, Green(1988:241) argues that, "in the main, conflict revolves around the tenant's failure to pay punctually at the end of themonth." At the same time, there are also conflicts which arise from living in the same house; landlord and tenant mightsimply get on each other's nerves. The issue of repairs can also be a source of conflict. In Guadalajara and Puebla, onethird of tenants interviewed reported current problems with the physical state of the house - leaking roofs, damp wallsetc.; unfortunately, only one in five complaints seemed to produce any action from the landlord. (Gilbert and Varley,1988:201)

Page 51


Chapter III. Landlord-tenant relationsD. Contracts and legality

Tenants often receive no contract from the landlord. Few landlords issue contracts in low-income areas ofBogotá, Bucaramanga or Santa Cruz, (Edwards, 1982; Gilbert, 1983; Green, 1988) although, in Mexico City, the situationseems to vary considerably. In two peripheral settlements, CENVI (1989:58) found that more than half the tenants had acontract, while, in two others, only one in six had signed one. Elsewhere in Mexico, contracts seem to be quitecommon; three out of five tenants have a contract in Guadalajara, and four out of five in Puebla. (Gilbert and Varley.1988)

Whether having a contract is of benefit to either party is an interesting issue. In Karachi, Wahab (1984) claimsthat landlords deliberately keep the rental situation illegal, so as not to have to pay taxes. In Mexico, however, it isclaimed that the tenant has far more rights without a contract, which is why most owners in Guadalajara and Pueblainsist on issuing one. Some tenants also sign contracts that contain illegal clauses and, in the absence of legal advice,follow the disadvantageous rules laid down. Even when a legally binding contract is signed, there is still the problemof how to apply it; in most countries, the legal system is extremely slow, and delays can rebound badly on a tenantwho is being asked to vacate the property. Yet delays can work to the tenant's advantage; in Colombia, Mexico,Turkey and many other countries, landlords complain bitterly about the time required to evict a tenant. (Gilbert andVarley, 1988:189; Jaramillo, 1985:110)

Page 52


Chapter III. Landlord-tenant relationsE. Landlord and tenant organizations

Much has been written about social movements in developing cities, but relatively little urban protest seems tohave been the work of tenant organizations. Indeed, most protests seem to stem from trades unions, squatterassociations, neighbourhood groups or ethnic and migrant organizations. If there are cases of active tenants'organizations in a few cities, the vast majority of tenants do not seem to belong to any association. In South-EastAsia, Evers (1984:493) comments that, "so far, strangely enough, there is no 'land to the urban tenant' movementcomparable to a 'land to the tiller' movement in rural areas." Virtually no-one belonged to an association in twoself-help settlements surveyed recently in Mexico City, and, in a third settlement, the average was one in 10. (CENVI,1989) Similarly, in Guadalajara and Puebla, tenant organizations have been very ineffective, and, in a survey of tenantsin the city, few members of these associations were found. (Gilbert and Varley, 1988:206) Similarly, in Buenos Aires,Cuenya (1986:6) recalls that, with the exception of one rent strike early in the century, tenants had been inactive untilvery recently. Only during the 1980s have they begun to engage in street marches, organize national congresses andlobby the authorities.

Tenants generally seem only to have mobilized in certain circumstances. Organizations have tended to becommon in public housing estates (as in Abidjan and Mombasa), (Cohen, 1974; Siren, 1982:152-153) and tenants havealso organized demonstrations when they have come under some kind of threat, as in inner-city areas subject to urbanrenewal programmes. In Bogotá, in the early 1970s, for example, the threat of a new motorway led to widespreadprotests during the early 1970s among low-income communities along the proposed route. (Reviez and others, 1977)Even when threatened, however, tenants might not mobilize. In Montevideo, for example, the removal of rent controlsand a campaign of urban renewal, during the early 1970s, failed to stimulate a tenant movement; fear of theauthoritarian military regime precluded such action. (Benton, 1987:46) in Sao Paulo, tenants who lost their housesbecause of the construction of a metropolitan railway could not mobilize, because the authorities did not tell them whatwas happening. By the time they knew they would be evicted, it was too late. (Batley, 1983)

In this context, the events in Mexico City after the 1985 earthquake seem to be very exceptional. As a result ofthe damage to vecindadades in the inner-city area, many tenants feared that they would lose their tenancy rights -especially as landlords seemed to be using the emergency to demolish already deteriorated rental property. The localneighbourhood organizations, therefore, organized a series of marches and established a co-ordinating association toplan their protest. Their campaign brought a change in official policy, with the Government building low-incomeaccommodation in the inner-city area, an almost unprecedented reversal of the general policy of urban renovation andslum demolition. (Massolo, 1986; Connolly, 1987; Villa, 1987)

In most cities, however, few tenants belong to tenant organizations, and most are also much less involved thanowner-occupiers in neighbourhood associations. In a general review of political organization among the poor, Nelson(1979:253) notes that "neighbourhood associations are rare in older, more central, heavily rental low-incomeneighbourhoods." in Bogotá and Mexico City, participation rates in settlement petitioning were certainly far lessmarked among tenant households than among owner-occupiers. (Gilbert and Ward, 1985:205) in Lagos, Barnes(1987:66) also found that "owners were more involved in community associations than were tenants". Despite theirnumbers tenants are not even influential politically; thus, "in both Mathare and Dagoretti, the interests of residentlandlords were fundamental to election politics. Although a large majority of the population were tenants, the votes ofthe poorest could be obtained with the large sums available for election contests". (Peil and Sada, 1984:347)

From this last comment, it is clear that landlords are more likely to wield influence than are tenants, especiallywhere it is the rich and powerful who let accommodation. In Nigeria, both the affluent and the well-connected are likelyto become landlords, a situation which derives from the time of independence when government-supplied housing"filtered down to top civil servants and politicians who succeeded the white man ...". (Muoghalu, 1987:166) Clearly,such landlords have the power to influence urban policy. No demolition affected the crowded squatter settlements ofPort Harcourt, for example, because "many landlords ... have connections with people In top positions". (Ekpenyong,1989:47) in Ghana, "the members of ruling regimes have consistently invested in housing and have often usedwhatever gains they made in office to build large houses". (Tipple, 1988:44) This elite has invested in the top end ofthe rental market which has been excluded from blanket rent controls.

While power can help a person become a landlord, some people become powerful because they are landlords.As Barnes (1987:54) points out, "the people who were successful in securing property were a heterogeneous groupbut they were similar in one critical respell. Ownership set them apart from tenants to the extent that they occupied aprivileged position in the urban social hierarchy".

Page 53

With respect to political power, the Latin American situation seems rather different from others. The vastmajority of landlords lack political influence and, often, they are little better organized than the tenants. In Guadalajaraand Puebla, none of the 47 landlords interviewed by Gilbert and Varley (1988) was a member of a landlord association;in Mexico City, only one out of 4b resident landlords belonged to such an organization. (CENVI, 1989) In Bogotá,many middle-class landlords only pined a landlord association when new rental legislation threatened to reduce theirrents. Since the legislation was modified, however, their organization has been rather inactive.

In general, therefore, political competition and conflict tend to be organized along lines other than tenure. Class,employment, race or religion seem to be the most significant motives behind political mobilization.

Page 54


Chapter IV. Public policyA. Rent control

1. The introduction of rent controls

Faced by a mass of poor families and an inability to provide sufficient public housing, most governments havefelt the need to introduce some kind of rent control. Few governments in developed countries had failed to introducecontrols by the late 1940s, and, even in developing countries. "... rent control is not a new or novel concept". (UnitedNations, 1979:2) Rent controls have been adopted widely because they are cheap to introduce; they are popular withtenants who outnumber landlords many times; they help to control inflation, both directly by keeping rents low andindirectly by discouraging demands for wage increases; and they are relatively simple to administer.

There is a large literature on this experience, and accounts on individual country experiences are included in thefollowing sources:

Argentina: Rivas (1977)Brazil: Kowarick and Ant (1988), Taschner (1988)Chile: Lozano (1975)Colombia: Fedelonjas (1982), Jaramillo (1985), Jaramillo (1987)Egypt: Malpezzi (1986)Ghana: Tipple (1988)India: Muttagi (1988), Sundaram (1987)Kenya: Stren (1978)Mexico: Coulomb (1985b). Perlo-Cohen (1979), Portillo (1984)Nigeria: Onibokun (1985)OECD countries: MacLennan (1986)

Most forms of rent control have been brought in during periods of rapid inflation, particularly during wartime.In the United Kingdom, rent controls were introduced during the First World War, and similar measures were taken insome of the then colonies (including India, Kenya, Malaysia and Singapore) during or immediately after the War. InLatin America, rapid inflation during the 1920s led to legislation in Argentina and Chile. (United Nations, 1979; Rivas,1977) Controls were relaxed during the 1930s but were re-established as a result of the Second World War. Mostdeveloped countries "introduced rent controls at the outset or in the aftermath of the 1939-1945 war", (MacLennan,1986:22) and, in many British colonies (including the later Ghana, India, Kenya, Malaysia, Sri Lanka, Uganda and theUnited Republic of Tanzania), the authorities took similar action. (United Nations, 1979; Stren, 1978; Tipple, 1988)There was, also, a rash of rent-control legislation in Latin America during the 1940s (for example, Argentina in 1943,Brazil in 1942, Chile in 1941, Colombia in 1943, and Mexico in 1942). If the slowing of inflation was a common motivebehind rent-control legislation, the timing and form of that control were heavily influenced by political considerations.Electoral considerations seem to have been at play in Brazil when, in 1964, President Goulart "wanted immediately tosign the decree pegging rents. He felt ... that his announcement that the rent indexation measure was in the pipelinehad caused much more enthusiasm, even wild cheering, than the other reforms of which he spoke, including the landreform... in Rio, Sao Paulo, Recife and the other big cities, the rents measure was the talk of the town". (Jurema, 1964)

Page 55



2. Which rents have been controlled?

In Cuba and China, rent controls covered all kinds of property, but, elsewhere, coverage has been highlyvariable. Many governments have refused to apply rent controls to new property, so as not to discourage rentalinvestment, the period of exemption for such building varying from three to 20 years. (United Nations, 1979:6)Elsewhere, legislation has been applied only to property above or below particular levels of rent. In Harare, controlscover only high-income property and, in Bangalore, they exclude all low-income accommodation. (Rakodi, 1989)Typically, however, controls have been applied only to low-income and middle-income rentals (United Nations,1979:6); in Colombia, Ghana, Kenya and Mexico, rents in high-income accommodation have rarely been affected bycontrols. (Jaramillo 1985:123) Where controls are limited to specific rent groups, coverage tends to diminish throughtime, because the level of rents below which controls apply are rarely raised in line with inflation. Thus, the KenyanGovernment, up to the middle 1970s, "was progressively limiting the coverage of the legislation... by maintaining theceiling above which the rent restriction legislation did NOT apply". (Stren, 1978:156) Similarly, in Mexico City, thecontrols introduced in 1948 were affecting only 22 per cent of property by 1961 and only 1 per cent by 1976. (Mexico,Secretaria de Hacienda, 1964)

However, the most relevant question Is whether rent controls are actually applied in practice. The evidence onthis seems to be very mixed. Controls do seem to have had an effect in the central areas of Buenos Aires (Rivas, 1977),Sao Paulo (Kowarick and Bonduki, 1988) and Mexico City. They have also been highly effective in Harare and inKumasi. (Rakodi, 1989:12: Tipple, 1988:47) Elsewhere, however, rent controls seem to have had little impact. In Nigeria,"rents actually paid by tenants have risen and remain higher than edict-stipulated levels in many urban centres".(Okpala, 1985:150) In Guyana, Peaks (1987:135) claims that the effects of rent controls are "minimal".

Typically, however, rent controls affect different parts of the rental-housing stock in different ways. Certainly,legislation is least likely to operate well in low-income settlements. In Bogotá, for example, Jaramillo (1985:124) foundthat only 17 per cent of tenants in illegal subdivisions had a written contract, and the rental legislation was understoodby only one in four landlords and one in five tenants. In Mombasa, there appear to be different views on theeffectiveness with which legislation brought benefits for the poor. Stren (1978:152), for example, argues that, during thelate 1960s, the poor made regular use of rent tribunals and often succeeded in getting their rents reduced. By contrast,Ghai and McAuslan (1970:289-290) claim that "the urban poor have little knowledge of the content of rent controllegislation ... and they have practically no contact with the legal profession in its civil capacity. They are thereforetotally unable to make use of the facilities provided for them."

Another general problem is that landlords are wont to find ways of avoiding the legislation. Frequently, theyintroduce new methods of charging; key money might be extracted from new tenants, for example. (Malpezzi, 1986;United Nations, 1979.23) in Nigeria, "prospective renters are subjected to heavy advance payments without legalprotection by way of receipts and where receipts are issued, they carry the standard rent fixed by the edicts". (Okpala,1985:153) In India, rent control has also led to illegal transactions and a parallel market in premium and subletting.(Sundaram, 1987:65) in the inner-city areas of Mexico City, rents during the 1980s seem to have been rising much morequickly than the legislation permits. (Portillo, 1984; Mendez Rodriguez, 1987) In both Lima and Rabat, inadequatelyformulated rental legislation seems to have rebounded on tenants, making their situation worse rather than better.(Padron and Calderon, 1984:184-185; Keles and Kano, 1987:154)

Page 56



3. Effects of rent controls

Insofar as rent controls have been applied effectively, have they helped the tenant population, and particularly,the poor? in the short term, controls often benefit tenants by cutting rents. In Cairo, Kumasi and Mexico City, thesebenefits have continued to apply over long periods to any tenants who have stayed in the same property: rent-incomeratios for many households in those cities are very low. As time passes, however, the side-effects of rent controlbecome apparent and undermine most of the advantage for tenants.

Few economists seem to doubt that rent controls discourage investment and reduce the size and quality of therental-housing stock. (MacLennan, 1986:22) in cities in developing countries, rent controls consistently "reduce therate of return on housing, thereby, discouraging investment in shelter". (Renaud, 1987:187) There is, certainly, plentifulevidence in support of this contention. In Harare, Rakodi (1989:12) argues that the effect of rent controls since 1982"has been to inhibit new construction for rent, and to encourage the sale of rented property". In Sao Paulo, the rentfreeze of 1942 helped "to discourage the building of rental houses ... especially that for low-income families".(Taschner, 1988:15, 17)

Not only does rent control reduce investment but it often encourages landlords to shift property out ofresidential uses. Thus, the introduction of rent controls in Bogotá, in 1983, led to middle-income housing being put upfor sale or modified for office and commercial use. (Jaramillo, 1985:125) Similar shifts in property investment have beenobserved elsewhere. "In Lagos, there is an increasing tendency to build for rent flats than single rooms to avoid rentcontrols". (Peil and Sada, 1984:300) In Buenos Aires, Rivas (1977) notes how the rent control legislation in the early1920s slowed tenement construction and encouraged landlords to develop the hotel-pension as a means of avoidingthe legislation.

Rent controls also seem to have a negative influence on maintenance, accelerating property deterioration. Oneview on the United States of America experience is that "next to bombing, rent controls seems in many cases to be themost effective technique so far known for destroying cities". (Lendeck, 1972:39) in member countries of theOrganization for Economic Co-operation and Development, "rent controls ... have had a pervasive negative effect onproperty condition". (MacLennan, 1986:71) In Bombay, Muttagi (1988:74) argues that rent controls gave landlords "noincentive to keep the building in a good state of repair" and, in Montevideo, rent controls in the inner-city led tolandlords responding "to depressed rents by foregoing maintenance and repairs...". (Benton, 1987.37) in Mexico City,cheap vecindadades were poorly maintained. Rents were often too low to pay for repairs. More important, however,was the lack of land market controls. Its price went up so quickly that land with a rent-controlled vecindad is nowvalued less than vacant lots. About the only way to evict the tenants, who are protected by law, is through thecollapse of the house.

On top of this, the distributive effects of rent controls are not necessarily those desired by the legislators. InCairo, the benefits to long-time renters were far higher than those received by recent or frequent movers. As a result,there is little link between those in need and those who gain from rent controls. Similarly, in Bangalore, there are caseswhere the benefits of rent control were clearly accruing to the wealthy at the expense of the intended beneficiaries. Notonly may rich tenants gain while poor tenants receive no benefit but sometimes rich tenants may gain at the expense ofpoor landlords.

Page 57



4. Removing rent controls

In the light of all these problems, should rent controls be removed? On the right of the political spectrum, ofcourse, there is total agreement on this issue. As a foremost right-wing economist puts it, "... rentcontrol/restriction/regulation has done much more harm than good in rental housing markets - let alone the economy atlarge". (von Hayek and others, 1972) This conclusion is based on a series of arguments, notably that rent controlperpetuates shortages, encourages immobility, swamps consumer preferences, fosters deterioration, erodesproduction and incentives, and distorts land-use patterns, On the political left, the opposite view is sometimes given,in Mexico City, rent controls have at least maintained cheap accommodation in the inner-city areas. (Coulomb, 1985b)In Chile, too, it has been argued that the 1970 rent-control legislation brought important benefits for the poor. "Thenew maximum rent was to be computed as 11 per cent of the assessed value of the property for real estate taxpurposes. Since the assessed values generally had been considerably lower than the market value, the new rent levelswere, in most cases, much lower than before. One result was that many rental houses and apartments were sold totenants at attractive prices, and cooperatives and condominiums were organized to manage the acquired housing".(Lozano, 1975:183)

There is also considerable debate about whether rent controls have been as destructive as many on the righthave charged. In Mexico, for example, it is argued that the expanding range of investment options has contributedmuch more than rent controls to the undermining of rental investment. The World Bank's research on rental housing inBangalore, Cairo, Kumasi and Rio de Janeiro also shows that rent control emerges as merely one among several factorsdiscouraging investment in housing. Even stronger than this is the view expressed by the United Nations Centre forHuman Settlements (Habitat): (1984:12) "comparing statistics concerning private sector production of rental housingprior to and after the imposition of rent control suggests that the impact of rental legislation in this respect may havebeen marginal vis-à-vis a series of other factors, such as the rising prices of land and materials, increasing labour costsand the emergence of other attractive options for capital return, which combined to divert private sector investmenttowards more lucrative ventures within the housing market."

There are also some economists who argue in favour of carefully designed forms of rent control. AsMacLennan (1986: 22) has argued, housing shortages, although not necessarily chronic and long-term, do tend torecur as a phenomenon of market adjustment (with demand always likely to expand more rapidly than supply). Sinceshortages disproportionately impact on low-income groups, some form of control must be introduced, but the controlmust be set in order not to preclude re-investment. Such rent controls would certainly not attempt to cover all kinds ofrental housing but should be "discrete and selective" with respect to type and age. (Okpala, 1985:139)

On the whole, however, the current consensus favours decontrol, although most agree that existing controlsshould not be removed too quickly, if they are dismantled too quickly, the results can be highly damaging for tenants.In Montevideo, in 1973, "with housing already scarce and rents artificially depressed for decades, the results of thispolicy were immediate and dramatic, especially for the poor. Housing quickly became the category of goods to risemost steeply in price". (Benton, 1987:40) The answer is surely to phase rent controls out gradually - a floating up andout, as Malpezzi terms it.

Page 58



5. What is the best method of controlling rents?

If rents are to be controlled, what is the best method of doing this? The United Nations (1979) provides a goodreview of the different methods which include: freezing rents, roll-back formulae, rates linked to square metres of space,rents based on a fixed rate of return, rents based on cadastral appraisement and indexation.

Probably, the easiest system to apply is that based on cadastral appraisement. Annual rents are permitted up tosay 12 per cent of cadastral value. In theory, this gives the owners a reasonable return on their capital, while limitingrents. The problem, however, is that cadastral values are often much lower than market values. While this gives anincentive for owners to have their properties revalued, the efficiency of most valuation offices is rather low, and a greatdeal of administrative difficulty is involved in changing the taxable value, (Jaramillo, 1985:120) There is also a difficultywhen other taxes are linked to cadastral value, In Bogotá, where water charges and several local taxes are levied on thisbasis, raising the cadastral value has led to massive increases in a variety of taxes and charges for some owners.

Perhaps, the greatest problem with respect to low-income tenants, however, is that many self-help housingareas have still not been regularized, and therefore, lack cadastral values. While efforts have been made in recentyears, in cities such as Bogotá and Mexico City, to update the tax base, this means that a considerable amount ofrental property is excluded.

A few governments have tried to tie rents to tenants' incomes. This was tried in Nigeria in 1977, when Federalguidelines laid down that the rents of low-income and middle-income tenants were not to exceed 20 per cent of theirincomes, and those of high-income tenants, 25 per cent. (Okpala, 1985:141) The difficulty involved here is that tenantsin the same kind of property and, even, in the same house could be paying very different rents according to theirincomes. In addition, it is difficult for landlords to know what the income of their tenants actually is. While theprinciple of fixing rent on the basis of ability to pay is excellent, the problems of application seem to beinsurmountable.

Whatever method is chosen, it should not be too complicated for the administrative capacity of a localauthority to apply it. Also, it should not encourage bribery and corruption, or be too difficult for landlords and tenantsto understand.

Page 59


Chapter IV. Public policyB. Increasing the size of the rental housing stock

Should more people live in rental housing? Since 90 per cent of the Chinese rent accommodation, and 90 percent of Bangladeshis own it, the question can only be answered sensibly in a specific context. Perhaps a validquestion is whether every society should offer a wide choice of tenures: surety, the answer to that is in the affirmative.Since the range of household needs is very diverse, a similarly wide array of housing options is to be recommended.Certainly, rental accommodation seems to have an important role to play; it seems particularly suitable for newlyestablished households, for transitory groups, such as temporary workers or students, and for those without or withother uses for savings. It is also suitable for those households, especially the old, who simply do not want theresponsibilities that go with ownership.

Arguably, there are also other advantages. Lemer (1987:35) points out that rental accommodation provides aquality of housing equal to or better than owner-occupation, but the "demand for rentals is met with smaller units.denser development, and with possibly greater private sector participation". As such, governments can achieve agiven increase in the housing stock with reduced resources. Rental housing also offers a way of increasing theincomes of poor households. In sites-and-service schemes, "it is now realized that the leasing of rooms may be one ofthe most effective ways of increasing incomes of the plot holders... this source of profit is an important factor makingplots affordable to participant families". (Keare and Parris, 1982:ix) Similarly, rents create resources which can be usedfor investment; landlords can use their profits in other directions. Perhaps, even more important is the possibility thattenants can use their savings to set up a business rather than tying up their money in a house. (Edwards, 1981; Keareand Parris, 1982; Sunil and van den Eerenbeemt, 1988)

In general, therefore, it is difficult to argue against the case that some kind of rental accommodation should beavailable. Where it is now lacking, the only question is how best to produce it. Past experience suggests that thepublic sector can rarely provide it satisfactorily. Even where governments have traditionally constructed mosthousing, attitudes have recently changed; in China and Cuba, encouragement has been given to the private sector.Elsewhere, the general trend has also been away from public housing construction and ownership; in Hong Kong,Singapore and most parts of Latin America, public housing is being sold to tenants. Whatever the precise reason, thetendency of most governments at the moment is to encourage the private sector to supply rental housing: thequestion is whether this can be done.

Page 60



1. Incentives to the formal sector

Numerous governments around the world have tried recently to stimulate formal-sector involvement in therental sector. In order to encourage investment, they have dismantled rent controls and have offered generousincentives to private companies. In Saudi Arabia, subsidized loans encouraged private builders to produce multistoreyaccommodation for expatriate professional personnel; some 17,100 units were built between 1975 and 1980, and afurther 7700 between 1980 and 1985. (Tuncalp, 1987:353-354) Similarly, the Government of the Republic of Koreaintroduced highly-subsidized loans for rental-housing construction as well as generous exemptions from capital gainstax. (Byong Key, 1987) The most attractive incentive, however, was the offer of serviced land at a 10 per cent discount.Between 1982 and 1986, some 1.5 million square metres of land were provided by the Korean Land DevelopmentCorporation, sufficient for 34,000 rental units. Unlike the Saudi Arabian programme, this ambitious scheme hasprovided housing for low-income families. Whether, given its high level of subsidy, it could be replicated in manyother countries is questionable.

The Mexican Government has also offered generous incentives to private companies to build housing for rent.During a period of rapid inflation, it offered loans at very low Interest rates, providing that the accommodation berented for a minimum period of 10 years. In 1985, the minimal rental period was reduced to five years, and tax relief onpart of the investment was introduced. Official figures suggest that the scheme has been successful, but mostcommentators take the opposite view. Many investors think that l is simply more profitable to put their money in thebank or to speculate on the stock exchange (Gilbert and Varley, 1988:57) and it will take a great deal of governmentaleffort to change this attitude.

Page 61



2. Incentives for the informal sector

By contrast with the attitude of many professional builders and financiers, the small-scale landlord continues tolet rooms. Some, indeed, believe that renting is still a good business. Among 40 landlords in two self-help settlementsin Mexico City, three out of five thought it to be at least a reasonable business. (CENVI, 19139:39) fn Guadalajara andPuebla, although there was disagreement among landlords that renting was a good business, many were still buildingnew accommodation. The most satisfactory explanation of this behaviour is that small-scale landlords are following adifferent rationale from that of commercial landlords and rental administrators. Small-scale landlords in theconsolidating self-help settlements are unlikely to compare the returns from tenants with those from share dealing orforeign-currency transactions. In Mexico, "a mixture of motives, revolving around the family, old age and a lack ofperceived alternatives, seems to characterize the small-scale landlord". (Gilbert and Varley, 1988:230) it could well bethat in most developing-country cities, there is a majority of landlords who will continue to build accommodation andto rent it out.

If self-help landlords are continuing to build, either because it is the only business they know or because theyeventually want the extra space to accommodate kin, perhaps they should receive assistance. As Hansen and Williams(1988:316-317) argue, "increasing the supply of rental units by encouraging homeowners to rent out rooms and addrental units may be the most efficient way to increase shelter in third world cities. Two factors underlie this argument.First, adding rental units to existing housing means that no additional land costs, which can account for 50 per cent oftotal dwelling costs, are incurred. Labour and materials are the only major inputs. Second, as a result of higher-densitydevelopment, homeowners and renters combined on the same lot can better afford basic services such as water. Thusthe prospects for cost recovery of infrastructure investment are greater than in less dense, low-income communities."Since, small-scale landlords probably produce the bulk of the accommodation, they should receive any incentives thatthe State makes available to landlords.

The advantage of giving incentives to this group is that they are generally not much better off than theirtenants. In this light, the policy is not socially regressive. While it would probably increase inequalities withinlow-income settlements, it is much more equitable than most normal governmental housing policies.

What might also encourage further investment by small-scale landlords is the regularization of self-helpsettlements. Indeed, regularization seems to be a highly effective method of increasing the amount of rental housing,operating both on the demand and on the supply side. Improvements in transport provision and the availability ofelectricity, water and drainage attract tenants to a neighbourhood. In Bogotá and Mexico City, upgraded settlementshave quickly attracted tenants; (Gilbert and Ward, 1985) expansion in the number of rental rooms also followedregularization and servicing programmes in Madras, Manila and Rabat. In the Tondo improvement programme, eventhough reblocking reduced the average plot size, rental housing space per unit soon rose by 10.5 square metres andthe number of families letting property increased by 10 per cent. (Keare and Parris, 1982:9) In Madras, an improvementprogramme in Ashok Nagar led to a near doubling in the number of rooms available for rent. (Robben, 1987:104) Whilethere have been exceptions - for example, Baldia township in Karachi (Nientied and van der Linden, 1987) -regularization and servicing appear to be a highly effective method of encouraging rental housing for the poor.

Evidence certainly suggests that the growth of rental accommodation occurs without the need to offer full legaltitle. Indeed, landlords seem to invest whatever the legal niceties. fn Indonesia, "lacking the proper legaldocumentation has not stopped people from investing in building rental units. In some cases it even seems toencourage their building, because with a greater population, the neighbourhood becomes less likely to be the target ofeviction". (Nelson, 1988:28-29) Similarly, the illegal subdivisions of Bogotá and Bucaramanga and the unauthorizedejido settlements of Mexico City, Guadalajara and Puebla are full of landlords and tenants. (Coulomb, 1981: Edwards,1981; Gilbert 1983; Gilbert and Varley, 1988)

What is not clear is whether regularization damages the interests of existing tenants. Certainly, there isevidence of rents rising rapidly after service improvements. In Calcutta, a post-improvement rent survey showed that"between 1974 and 1980, the average rent increase had been 43% in improved bustees and only 16% in unimprovedbustees". (Moitra and Samajdar, 1987:77) Rent rises were also observed after upgrading projects in Karachi, Manila,Madras and Rabat. (Nientied and van der Linden, 1987; Keare and Parris, 1982; Robben, 1987; Keles and Kano, 1987)The rents rose both because owners were required to pay for the new services and because the improvement attractednew tenants to the settlements. Certainly, many of the new tenants were drawn from high-income groups. Whether thisled inevitably to the displacement of existing tenants, however, is not entirely clear. In Calcutta, Moitra and Samajdar(1987:77) claim that self-employed workers were gradually replaced by high-income white-collar workers. In Madras,

Page 62

however, although rents rose, landlords neither evicted their existing tenants nor sought to raise the rent greatly.(Robben, 1987:99)

If settlement upgrading has a generally positive effect on the supply of rental housing, so, on occasion, doesthe establishment of a sites-and-services programme. In Rabat, somewhere between one and eight rental units arecreated on every plot sold. (Keles and Kano, 1987:164) in Kenya, the conversion of owner-occupied,sites-and-services schemes into rental settlements seems to be even more dramatic. In one project in Nairobi, threequarters of the households are now tenants; in another, in Thika, nearly all housing Is rented. In both schemes, manyowners "preferred to sublet the whole house and remain in lower quality housing areas". (UNCHS (Habitat), 1987a:8)The lesson seems to be that, in many cities, the creation of any form of new housing will rapidly increase the supply ofrental accommodation. The policy implication is that opening up new self-help areas creates opportunities both forownership and for renting. Keare and Parris (1982:ix) suggest that this both expands the housing supply and helpsowners to repay their mortgages.

An alternative method of increasing rental housing is to shift help away from owners and potential landlords tothe tenants. If giving incentives to the private sector is not always effective, what about offering income supplementsto the poor? On the basis of experience in the United States of America, Dasso (1968) argues that rent supplementsrepresent a least-cost means of ensuring adequate housing for all, with minimal social cost to low-income families.Similarly, Duncan (1977) supports the idea of housing subsidies to the tenant on the basis of Swedish experience;however, even the effects of this long-established programme have been inadequately studied, and, in any case, theprogramme is currently subject to review. (Westerlund, 1988) Bradbury and Downs (1981) present a generalintroduction to the whole issue of housing allowances in developed countries.

It is quite dear that, given the poverty of the governments of most of the developing countries and theconstraints which face them in raising revenues, too little in the way of funding will be available to give housingallowances to any more than a few tenants. Even if the funds could be found, it is by no means certain that tenantswould spend the subsidy on rents; very poor families often have more urgent needs than housing. It is, also, notcertain that landlords would increase the housing stock available, they might just increase rents.

Page 63


Chapter IV. Public policyC. Environmental standards, slum demolition and urban renewal

The most serious accommodation problems in rental housing are usually found in old, inner-city areas. Theproblems are most serious because the property has usually deteriorated, owing to lack of maintenance. Since therising value of the site often means that the building and the future flow of rents are worth far less than thedevelopment value, many landlords cease to maintain the properly. Indeed, they hope that it will soon fall down, sothat tenants will leave and they can sell.

The traditional answer to slum housing has been to demolish it. Either the private landlord has been permittedto sell off his property to developers, or the public sector has introduced urban-renewal or relocation schemes.Unfortunately, the basic result of these schemes has usually been to displace tenants from their houses. Worse still,what begins as a policy of improving housing usually leads to the removal of tenants and their being moved to poorhousing elsewhere. (Morris, 1979; Hardoy and Satterthwaite. 1981: UNCHS (Habitat). 1984) Where tenants have beenmoved to new housing estates, relocation has rarely been a success. In Malaysia, relocation schemes offeredimproved living conditions, but the displaced families did not appreciate either the new location or the high cost ofaccommodation. (Wegelin, 1978:115) In the disastrous squatter-relocation schemes in Rio de Janeiro, many families gotbadly into arrears on their mortgage payments, while many others sold their new houses and moved back to the centreof the city. (Perlman, 1980; Valladares, 1978)

This is hardly a novel finding, because it was a common experience in nineteenth-century London. The effortsto improve sanitation and housing conditions merely led to displaced tenants moving into the nearest remainingslums. Increased overcrowding was one result of efforts to improve sanitation. (Stedman-Jones, 1971) Massivedemolition schemes and tough building codes were important causes of worsening housing conditions. Even in situimprovement could be counterproductive: "as an answer to the housing problem, the effects of the (Octavia Hill)scheme were insignificant. Landlords who exerted themselves to improve their properties, by the normal laws of themarket, ejected poor tenants, put up the rents, and attracted another class of occupier. (Stedman-Jones, 1981:195)

Similar efforts to improve the quality of housing in developing countries, have produced identical results.Experience in Lima showed that, "... if improvements were made, rents automatically rose". (Dietz, 1981:63) A study inSao Paulo found that renters are the most vulnerable group as regards displacement associated with urban-renewaland slum upgrading projects; they leave because of rising rents. (Bogus, 1981)

Schemes specifically designed to rebuild housing for low-income groups have also been unsuccessful. Twoprojects in central Lima demonstrate that most redevelopment schemes destroy more homes than they create and alsostimulate a process of gentrification. Rehabilitation in Lima led to much of the population of the Manzanilla and ElPorvenir districts being displaced to peripheral neighbourhoods. (Padron and Calderon, 1984:186) in Mexico City, theTepito Plan resulted in the construction of 761 flats and the destruction of 1160 rental units; most of the displacedfamilies were forced to remain in supposedly temporary housing. (CENVI, 1986)

Even efforts to prevent the construction of bad housing have often proved counterproductive. In Mexico City,tough planning codes were introduced at the end of the Second World War to discourage the building ofvecindadades. "The response of the commercial sector ... was the growth of the so-called ciudades perdidas". (Sudra,1976:84) Owners would rent out very small plots of land on which tenants would build shacks. "The price of such aplot ... was well above the rent for a dwelling in the rent-controlled vecindad and proximate to the economic rent for avecindad room". (Sudra, 1976:84)

The basic dilemma is that, without subsidies, very poor tenants can only afford to live in the inner-city i1housing conditions are unsatisfactory. Any attempt to improve deteriorated property will lead to high rents and,therefore, to the eventual removal of poor tenants. If subsidies are available, however, the chances of displacementmight be lowered. Certainly, the efforts made in Mexico to repair the damage caused by the 1985 earthquakes seem tohave been highly successful. In an 18-month period, some 3121 damaged properties were expropriated, and 48,800houses were built or reconstructed. (UNCHS (Habitat), 1987) Of course, here the conditions forced the authorities totake radical action. Expropriation on such a scale was unprecedented in Mexico, and a 50 per cent subsidy wasprovided to the inhabitants. The interesting question is whether the 12,000 families excluded from the initial phase ofthe programme will fare as well under the terms of the unsubsidized second phase.

If large-scale schemes rarely match the achievements of the post-earthquake efforts in Mexico, perhaps theanswer is to undertake limited programmes. Certainly, there is evidence that legislation can help improve the

Page 64

conditions of inner-city properties, at least in the short term. Even when landlords would prefer deterioration andcollapse, methods have been devised which preserve properties. Thus, in Oporto, "tenants can undertake essentialrepairs after 120 days and offset costs against rent payments". (MacLennan 1986:23) Another method of improvingmaintenance is to establish repair funds. In Sri Lanka, a national fund for repairs is financed by a tax on landlordsowning expensive property. (United Nations, 1979:37) A variation of this kind of scheme has been recommended inMexico City, where Coulomb (1985:45) suggests that tenants contribute 5 per cent of their rent to a repairs fundcontrolled by the local authority. In Bombay, a tax on buildings is collected by the municipal authorities, and theproceeds are spent on the rental housing most in need of repair. (Sundaram, 1987:65-6; Muttagi, 1988) This programmehas permitted groups of tenants to undertake the repairs themselves (UNCHS (Habitat), 1984:28) and has encouragedthe setting-up of tenant associations which were responsible for repairs to 2000 buildings, roughly one repair out ofevery four made during the 1969-1984 period. (UNCHS (Habitat), 1986:17) Of course, too much should not be expectedof such schemes: they are unlikely to repair more than a fraction of the deteriorated property. Even when they workreasonably well, as in Bombay. the problem is that "as some dilapidated buildings are being reconstructed, ... newshanty towns have been appearing". (Muttagi, 1988:85) In fact, no scheme is likely to be successful on its own. AsSternlieb (1969:xvii) argues on the basis of experience in the United States of America, "there is no single one-shotpanacea, be it code enforcement, financing, or tax relief, which will substantially improve the maintenance of slumtenements or induce owners to rehabilitate their parcels".

The real danger is that public measures might make matters worse. As the United Nations Centre for HumanSettlements (Habitat) (198:26) points out, "any form of intervention, be it from the public or the private sector, ininner-city areas seems to have the potential to disrupt the delicate equilibrium that safeguards the right to stay of thelow-income groups" The general answer to this danger seems to be to leave the property much as it is. The need ofmany poor people working in the inner-city areas is accommodation nearby. They will tolerate poor living conditions inorder to cut their journey to work. Given this preference for a good location, some official tolerance of poor but notdangerous conditions is to be recommended. Certainly, governments should not permit families to live inaccommodation that is physically dangerous; action should be taken if families are likely to be injured or be killed byfalling masonry or the like. However, in such circumstances, their choice is relatively straightforward: force thelandlord to remedy the faults or close down the property.

Page 65


Chapter IV. Public policyD. The role of public rental housing

It is difficult to make a general case that the State is either a better or a worse landlord than the average privatelandlord. The evidence from around the world suggests that there is so much variation that any generalization isbound to be of dubious value. In some societies, the behaviour of most private landlords seems to be quitereasonable, and there is no obvious sign of generalized landlord-tenant conflict. Elsewhere, the behaviour of someprivate landlords is appalling. Similarly, the performance of the public sector as landlord is highly variable. Somegovernments have been good landlords, insofar as they have given tenants complete security of tenure, even whenthe latter have failed to pay the rent. They have also rarely raised rents in line with prices. As a result of these implicitsubsidies, waiting lists for governmental rental housing have always been long. Yet few governments have been ableto maintain an affective programme for long, since neither resources nor management capacity have generally beenadequate.

In the final analysis, however, the view of the desirability of the government's acting as a landlord depends onthe level of trust which people have in the State. In some parts of the world, the State is trusted. Elsewhere, thegovernment can be so inefficient that the majority of tenants would not want to welcome it as landlord. As Okpala(1985:158) puts it, "The state as landlord ... could simply put power into the hands of a few in a. manner which towersabove the petty powers of individual landlords, and unlike in private ownership by a landlord, the state as landlordallows no peaceful alternative."

In any event, it is less than clear whether most governments want to take on the role of social landlord. Whilethe Governments of Saudi Arabia and the Republic of Korea have recently built some rental housing, mostgovernments. Including the two above, have relied principally on the private sector. Indeed, despite a realization inrecent years that rental housing is necessary, a number of governments around the world have eschewed the taskthemselves. Certainly, in Mexico, the Government has consistently preferred to offer incentives to the private sectorrather than construct rental housing itself.

Governments have turned their attention away from public rental housing because they have found itsmanagement particularly trying. Handling maintenance problems, reducing levels of tenant dependence and collectingsufficient in the way of rent have proved beyond them. Yet, there is evidence of governments which have managedtheir housing effectively. In Hong Kong, maintenance proved to be difficult in the early years of the public-sectorhousing programme, (Drakakis-Smith, 1979:163) but, today, it seems as if housing is well-maintained and operated. Inthe new towns, management has been improved by appointing a housing manager for each estate. "A better service isprovided to the tenants, and problems are dealt with speedily on the spot". (Tung, 1983:203) Hong Kong also showsthat rents in governmental property can be raised to "economic" levels, even if this means that housing goes tohigh-income groups. (Drakakis-Smith, 1979)

Even if public-sector housing can be run efficiently, that is not an argument for every government's engaging inlarge-scale construction programmes. Indeed, there is probably every reason why most governments should not try.After all, even where success has been achieved, it has been permitted by very special conditions. In Hong Kong, forexample, the authorities do not have to buy land at the market price: all land belongs to the Crown and can only beleased. "This has not only enabled the government to subsidize public housing projects by allocating sites at reducedprices, but has also permitted funds to be raised by auctioning off portions of cleared land to private developers atconsiderable profit". (Drakakis-Smith, 1979:164) Many governmental housing programmes elsewhere would work moreefficiently than they do, given similar conditions.

In Singapore, where two thirds of the population lives in housing built by public bodies, the general conditionsare hardly replicable in most cities. As Hardoy and Satterthwaite (1981:253) have asked, 'How many other third worldnations can claim rapid and sustained economic growth rates with very little immigration from other areas? in addition,how many major third world cities can build on the extensive reserves of well-located public land Singapore had in theearly 1960s?"

In most developing countries, public housing agencies are forced to operate in difficult environments. In most,they have been starved of resources, and their programmes have constituted little more than cosmetic exercises. Theyhave been "intended to be visible symbols of governmental concern for the poor. Rather than meeting the real needsof the low-income groups, such schemes are designed to win approval and votes, or to impress overseas visitors".(Drakakis-Smith, 1981:124) in practice, they have seldom achieved even these limited goals.

Page 66


Chapter IV. Public policyE. The role of rental housing in national settlement strategies

Many governments have adopted urban decentralization policies. They have tried to slow the growth of theircities and to develop neglected or resource-rich regions. Such policies have employed a range of strategies, includingindustrial deconcentration, decentralization of government offices, provision of infrastructure and services inprovincial cities, tax-break and credit programmes, and establishment of new towns and cities. (Gilbert and Gugler,1982:chap. VIII)

Assuming such programmes to be in any way practical (not necessarily a self-evident proposition), rentalhousing has a role to play in accommodating newcomers to expanded urban centres. In practice, however, fewgovernments have built rental housing, preferring to encourage house-ownership. Only in a few cases, notablyBrasilia, Chandigarh (at least initially) and Tema (more permanently), has rental housing played an important role.(Sarin, 1982:89) Clearly, if governments do not want to act as landlords, they will continue to sell public-sectorhousing, and rental housing will have a small role in national settlement strategies. The potential advantages of suchhousing in absorbing migrant populations will be sacrificed, given the practice of most governments to encourageowner-occupation.

Page 67


Chapter IV. Public policyF. The roles of co-operative housing and non-governmental organizations

Co-operatives make a substantial contribution to the provision of housing in many developing countries.(United Nations, 1975; Lewin, 1981; Turner, 1988) They encourage community participation, establishsites-and-services programmes, stimulate settler organizations to produce building materials and, generally, increasethe ability of the poor to manage their own affairs. Similarly, non-governmental organizations have contributed greatlyto self-help housing. (UNCHS (Habitat), 1988)

Unfortunately, their important contribution does not seem to have included much in the field of rental housing.A review of the contribution of non-profit housing associations in the middle-1970s reported virtually nothing in therental field (United Nations, 1975), and a recent review by the United Nations Centre for Human Settlements (Habitat)(1988) omits any mention of rental housing at all. Both non-governmental organizations and the co-operativemovement seem to have been strongly committed to the extension of owner-occupation. The common ingredient inmost programmes has been to obtain land on which to organize a sites-and-services programme. Very few non-profitorganizations seem to have been concerned directly with tenants. The neglect of rental housing by non-governmentalorganizations and co-operatives, of course, is no more surprising than its neglect by governments and is explained bysimilar reasons. It is unfortunate, however, because both kinds of organization are capable of playing an important rolein improving rental-housing conditions.

First, co-operatives represent a means of replacing exploitative forms of landlordism, and co-operatives shouldbe active in establishing and managing rental housing. This is particularly appropriate in inner cities, where tenantshave often lived for many years in the same property. Where there is mutual trust between the occupants, they oughtto be encouraged to buy out their landlords and jointly administer their accommodation. Certainly, the collectiveownership of tenements seems more appropriate than individual buy-outs which are likely to lead quickly to sales tohigh-income groups. (UNCHS (Habitat), 1984:34 and 40) As CHF (1985:9) argues, tenant or limited-equity co-operativesretain most of the increase In property values, only distributing part of this gain to individual members. The advantageof such a system is that newcomers can be accommodated at affordable rents, when existing members leave. Inpractice, however, there seem to be few examples of such co-operatives or housing associations in operation; Indiaappears to he the only exception. (UNCHS (Habitat), 1984:34; CHF, 1985:9) Of course, it should not be thought thatco-operative ownership is the answer to all rental-housing problems. Co-operatives generally work best with amiddle-class membership and have proved difficult to organize among the poor. (Keles and Kano, 1987:75) in addition,co-operatives can be misused; in Kenya and in Turkey, some of them have been turned into vehicles forstraightforward commercial speculation. (Amis, 1984; UNCHS (Habitat), 1987a; Keles and Kano, 1987:74)

Secondly, non-governmental organizations should be active in providing information to tenants and inencouraging the development of tenant associations. Tenants frequently need advice about law and about how todeal with the tactics of unscrupulous landlords. An important role for non-governmental organizations would be toarrange for free or subsidized advice to tenants. This is an imperative, given the current lack of effective tenantassociations in so many cities. So far, however, few non-governmental organizations seem to have accepted such arole.

Page 68


Chapter IV. Public policyG. The role of the law and the courts

In many countries, the law relating to rental housing is generally logical and sensibly worded. What is less clearis whether it is actually applied in practice. There are certainly many landlords and tenants who are unaware of the lawor who are unable to apply it.

In most countries, court proceedings are difficult for both landlords and tenants. Cost is frequently a problem,but it is the slowness to make decisions that is often the principal concern. In Mombasa, "magistrates were so busywith other work that plaintiffs in rent and eviction cases had to wait many months for a hearing". (Stren, 1978:149) inmost of Latin America, "judicial systems are overburdened, inefficient and unjust, ... judges have spectacular caseoverloads ... and cases drag on interminably". (Urrutia, 1987:59-60) Frequently, the slowness of the court proceedingsactually distort the aims of the legislation. In Colombia and Mexico, for example, landlords generally win cases, but ittakes so long to win them that many do not resort to legal action at all. (Gilbert and Varley, 1988; Jaramillo, 1985)

In the light of this kind of experience, the United Nations (1979:53) argued that "the settlement oflandlord-tenant disputes and the enforcement of rent legislation should be entrusted as much as possible to rentcontrollers and similar local authorities in order to reduce the workload of the overburdened court systems of mostdeveloping countries." Unfortunately, while there is much to be said in support of that recommendation, there is noguarantee that it will necessarily work, for it is an unfortunate feature of many developing countries that theadministrative system is inefficient and, often, partisan and corrupt. In Latin America, there is often little confidence inlocal government. As Dietz (1981:63) comments with respect to Lima, "going to the authorities on any level generallyproduced delays or simply no response; local municipal inspectors ignored conditions or were bought off." Even inCaracas, where municipal arbitration seemed to be reasonably effective, "some decisions were made predominantly onthe basis of partisan political influence". (Perez Perdomo and Nikken, 1982:224)

One option is to leave arbitration to local neighbourhood councils. Since most urban communities haverepresentative bodies, small disputes might be resolved best by a committee appointed by such bodies. Both Edwards(1981) and Karst, Schwartz and Schwartz (1973) drew this conclusion on the basis of work in Colombia and Venezuela.Yet there is no certainty that the local community trusts its representatives. In most communities, landlords have muchmore clout than tenants, and political influence might prove a decisive factor in arbitration. (Gilbert and Ward, 1985;Ray, 1969) In this respect, the sample results of Karst, Schwartz and Schwartz are instructive: most people had moreconfidence in official bodies than in their neighbourhood associations.

Questioning how the law should be implemented, of course, begs the question of the appropriate scope of thelaw in rental matters. Frequently, laws actually distort the workings of the rental system. Rent control either leads todeliberate evasion or to distortions in the normal operation of landlord-tenant relations. Sometimes, the law simplyfavours one group at the expense of the other, or favours particular subsets of landlords or tenants to the cost of therest. Too often, indeed, the law attempts to tackle issues which cannot be resolved except by increasing the supply ofhousing or raising the incomes of the tenants.

Perhaps, all that can be said at this level of generalization is that the law should be as simple as possible. Itshould be understood by most landlords and tenants. It should reflect as far as possible the normal practice of themarket. As far as possible, landlord-tenant agreements should be laid out in a standard written contract, blank copiesof which should be available from the average stationer, Disputes should be resolved, in the first instance, byneighbourhood or local-authority arbitrators. Insofar as legal advice is necessary, landlord and tenant organizationsshould be encouraged to provide it.

Page 69


Chapter V. Recommendations for public policy

Since tenure patterns vary widely across the globe, few general recommendations in favour of one or anotherkind of tenure structure can be made. In any case. since owning and renting do not offer the same benefits in onesociety as in another, such an exercise would be futile. The only general recommendation that might be made is that awide range of housing options should be available to every family. Accommodation under different forms of tenure, indifferent locations and with a range of prices and quality are an essential prerequisite for satisfactory housing. It is theresponsibility of governments to widen the range of choice available to every income group. If there is no rentalhousing available for a given income group, there Is probably a need for some; it there is no form of ownership, it isprobably required. John Turner's argument that different kinds of families require different kinds of housing and thatfamilies are the best judges of their needs is undoubtedly true. Insofar as governments are able to influence thehousing situation, their priority should be to widen the range of choice of land and housing options, so that everyfamily can obtain something approaching the kind of housing that it needs.

Page 70


Chapter V. Recommendations for public policyRent controls

1. Rent controls have been in operation in many developing countries for a number of years. The results ofthese controls are highly variable but, in general, they have discouraged many landlords from investing in rentalhousing. In addition, many among the urban poor have been excluded from the benefits of control. Newcomers to thehousing market, whatever their income, have usually been forced to pay market rates. As a result, the currentconsensus favours the removal of controls. It is also accepted, however, that any form of decontrol must be gradual, ifit is not to create political tension and to cause undue hardship for those tenants currently benefiting from rentcontrols.

2. Insofar as some governments wish to continue controlling rents, the most appropriate methods for localconditions should be selected. The method chosen should be sufficiently simple for landlords and tenants tounderstand, should not require the establishment of a complicated administrative system and should, provide little inthe way of opportunities for graft and corruption.

Page 71


Chapter V. Recommendations for public policyPublic-sector housing supply

1. Few governments currently wish to build housing for rent; even fewer wish to continue as landlords. Theyhave adopted this attitude, partly because of their experience as landlords and partly because of the lack of financialresources available to them. There seem to be few good reasons why governments should act as social landlords, ifsuch housing is destined only for high-income groups or is going to he badly run. At the same time, there is noinevitability about either outcome. Both in Hong Kong and in Singapore, very large public-housing programmes havebeen generally successful. Of course, conditions in those two places were very unusual; nevertheless, their experiencedemonstrates that the State can be a successful landlord.

2. On the whole, however, governments are probably best advised not to build housing for rent. Beforecommencing on a rental-housing programme, they should consider carefully whether the conditions which havepermitted some degree of success in Hong Kong and Singapore can be replicated in their own societies. Given theirlack of resources, however, they would often be best advised to concentrate on cheap, and, therefore, more likelyreplicable projects, such as sites-and-services and slum-upgrading projects.

3. Insofar as there is far too much evidence of the 'wrong' people obtaining public-sector housing or of the rightpeople getting it only for them to immediately sell or sublet it, what is the point of subsidies? There is a good case thatexisting levels of subsidy should be gradually reduced. If subsidies are to be given, allocation systems shouldguarantee that it is the very poor who occupy the housing. Insofar as poor people gain access to public housing, it isdifficult for governments to prevent them subletting; indeed prohibitions are almost certainly counterproductive.

4. In the final analysis, whether the State should act as a social landlord depends on the trust which peoplehave in the State. Where the its bureaucracy is so inefficient that the majority of tenants do not trust it, the Stateshould not attempt to act as a social landlord.

Page 72


Chapter V. Recommendations for public policyIncreasing the supply of private rental housing

1. It might be difficult to persuade professional builders and investors to invest in rental housing in most cities.Other channels for investment are often more profitable than housing and, usually, cause fewer headaches. Evidencefrom the Republic of Korea and Saudi Arabia shows that investors can be tempted by offers of financial incentives,but the cost of such incentives is high. In most countries, governmental resources are probably best spent givingmoney directly to poor families, both landlords and tenants.

2. By contrast with the attitude of many professional builders and financiers, the small-scale landlord stillcontinues to rent out rooms. If it is the case that such landlords are continuing to build, either because this is the onlybusiness they know or because they eventually want the extra space to accommodate kin, direct assistance to thisgroup probably represents the best method of increasing the supply of private rental housing. Insofar as many suchlandlords are relatively poor, this strategy offers a way of distributing governmental resources to low-income groups,combining goals of growth with distribution.

3. The regularization of self-help settlements will almost certainly encourage investment by small-scalelandlords. If roads are improved, bus services provided, and water and electricity supplied to the settlements, newtenants will want to move into the area, and there is plenty of evidence that landlords are prepared to createaccommodation. The evidence suggests that there is little need for settlements to be fully legalized: landlords ofteninvest without title deeds. The only significant problem with a regularization strategy is that, insofar as owners arerequired to pay for the improved infrastructure-and services, housing costs and, therefore, rents are likely to rise.While improved services mean that living conditions are likely to improve, in some cases, poor tenants might begradually displaced by affluent newcomers. Given that increasing the supply of housing is the highest priority,displacement of some households should be viewed as a necessary, albeit unfortunate, side-effect. In any case, thereis some evidence that many poor tenants retain their accommodation in upgraded settlements.

Page 73


Chapter V. Recommendations for public policyHousing conditions

1. The best method of improving housing conditions in the private sector is to encourage construction andupgrading. This will allow tenants to move out of poor housing into good accommodation or from expensive to cheapaccommodation, as they think fit. Legislation seems to be rather ineffective in persuading landlords to improve theconditions of their property. However, regular publicity campaigns, aimed at embarrassing bad landlords andencouraging regular repairs and maintenance, can bring some improvement.

2. One measure that might help the position of the tenant is to allow tenants to pay for essential repairs out oftheir rents, if the landlord refuses to undertake them. Repair funds are a complementary method of bringing aboutimprovements. Another valuable method of helping to improve living conditions is to encourage tenants indemarcated properties to form a co-operative to buy out the landlord and to manage the property themselves (that is, iflandlords are prepared to sell). This is likely to work especially well in inner-city areas, where tenants have lived in theproperty for a long period.

3. Many poor people, working in inner-city areas, need accommodation nearby. They will tolerate poor livingconditions, in order to reduce the journey to work. Governments need to face the fact that, without subsidies, verypoor tenants can only afford to live in the inner-city, when physical conditions are crowded and, sometimes,unsanitary. Given this conflict, some official tolerance of poor but not dangerous conditions is to be recommended. Acase can be made that poor housing conditions should be tolerated In inner-city areas. Indeed, any attempt to improvedeteriorated property will lead to high rents and, therefore, the eventual removal of poor tenants. Since the poor needto live near their work, housing improvements might lead to worsening housing conditions for the poor.

Page 74


Chapter V. Recommendations for public policyThe limits to State policy

1. The effectiveness of housing policy depends greatly upon the efficiency and honesty of the State, both indeveloped and developing countries. There is little point in recommending sophisticated forms of intervention, wherethe State bureaucracy is inefficient, personalistic or partisan. There is little point in extending the scope of official rulesand regulations, if this will merely open up further possibilities for what many would regard as dishonesty andcorruption.

2. Few governments are prepared to recognize that, without structural reforms, their ability to influencerental-housing conditions is limited. Insofar as they can influence housing conditions, the tools at their disposalfrequently lie outside the limits of the Ministry of Housing or its equivalent. Frequently, it is factors, such as theshortage of cement or the attractions of foreign-exchange speculation, that are the main deterrents to housinginvestment. As Sands (1984) has argued, "given the complex nature of the economic, social and demographic factorswhich impinge on the rental housing market, an inability to provide a simple solution to the rental housing dilemma isperhaps unavoidable". In short, the limits to governmental policy have to be understood in a structural context.Governments should not seek to take action where no remedy is feasible.

3. When governments attempt to improve rental housing conditions, they often follow a purely legislativeroute. There has often been little relationship between rental-housing legislation and official housing policy. Theco-ordination between housing programmes and legislation should be improved. Also, housing programmes shouldcontain specific recommendations about policy towards the rental sector. Monitoring of the effectiveness of everykind of State action is essential.

Page 75


Chapter V. Recommendations for public policyThe role of the law and the courts

1. Legislation is often poorly understood by both tenants and landlords. The State should attempt to keep itslaws as simple as possible. It should attempt to legislate on as few matters as possible and to concentrate its efforts ona few high priority areas.

2. Too often, the court system impedes the working of the law and distorts the spirit of the legislation. Eithercourt procedures should be simplified and made inexpensive, or alternative methods of arbitrating disputes betweenlandlords and tenants should be developed. Such alternative forms of arbitration could include local authoritytribunals and neighbourhood committees.

3. Legislation will work best if it is seen to be fair to both landlord and tenant. It will work best if knowledge ofthe law is widely diffused. Landlord and tenant associations should be encouraged to diffuse information to theirmembers about current laws and regulations. Insofar as legal advice is necessary, they should be encouraged toprovide it. Non-governmental organizations would seem to be an excellent channel for encouraging the developmentof such associations.

Page 76


Chapter VI. Priorities for research

Despite a recent upsurge in research on rental housing, still, remarkably little is known about the mechanics ofhousing supply, the rationale behind housing choice, landlord-tenant relations, the effects of government policy andthe operation of the law. While there is some information about a handful of cities, there is very limited information onhow rental-housing markets operate in most countries. Given this situation, a series of issues would repay furtherresearch.

Page 77


Chapter VI. Priorities for researchTenure patterns

It is clear that tenure structures vary considerably between countries and between cities. Some information ontenure patterns has been gathered, and some hypotheses have been formulated about why variations have comeabout. What Is still lacking is detailed Information for a sufficient range of cities over time. This information isimportant to establish She determinants of tenure structure. What, for example, is the effect of land-price changes ontenure structure and what is the effect of a rise in the costs of building materials? What is the likely effect of aneconomic recession on housing tenure? At the moment, it is only possible to speculate on the tenure effects ofchanges in such key variables. Without understanding these interrelationships, governments will be incapable offormulating an effective housing policy; external influences will constantly undermine any attempt at reform.

A databank for a number of cities should be assembled which would contain temporal information on a series ofvariables, including housing-tenure structure, service levels, housing conditions, land prices, land-acquisitionmethods, building costs, prices of building materials, wage rates, general cost of living and levels of employment andunemployment. A databank would make it possible to offer informed comments on the changing state of housing indeveloping countries, and to establish whether ownership is becoming more or less widespread or whether levels ofovercrowding are improving or worsening. It is necessary to know whether economic recession encourages ordiscourages households from becoming house-owners. At present, research bodies tend to undertake studies of asingle city, generalizing from that experience for the rest of a continent or the rest of the world.

Page 78


Chapter VI. Priorities for researchHousing supply

Currently, far too little is known about who invests in rental housing. It is not known whether the level ofinvestment is increasing or decreasing. Nor is it known how rental housing is created: is it through new investment,the addition of rooms to existing homes, the conversion of old properties or what? Neither is it known whether themajority of landlords are commercial operators or families building an additional home for rent or self-help builders. It isdifficult to formulate policies to encourage rental housing, when it is not known who is producing rental housing,because it cannot be established who will respond to incentives. If it is not known how rental housing is created, itcannot be known whether to offer incentives for conversion or for new building.

It is also necessary to know the rationale underlying rental-housing investment. Examination should be made ofthe portfolio management of commercial companies, to see why they invest or fail to invest in rental housing. Still moreimportant is a study of the self-help landlord. Insofar as this group of landlords is responsible for most newaccommodation, an understanding of why they build or convert accommodation is essential. In particular, whatdifferentiates the self-help landlord who expands the property for rent from the ordinary self-help owner-occupier? ifthere is little in the way of a commercial rationale underlying self-help construction for rent, there might be little pointoffering incentives to this group of landlords.

Insofar as landlords respond to incentives, what is the best form that such incentives can lake? What evidenceexists on the effects of different kinds of incentives? is there evidence of self-help landlords being encouraged toinvest through the offer of credit? Do they want or require architectural advice or help with planning authorities?

Page 79


Chapter VI. Priorities for researchThe desire for house-ownership

It seems that large numbers of households wish to own their own houses. At the same time, there are a numberof very interesting exceptions which have not been fully explained. Why is there, seemingly, such variation betweenhousehold behaviour in West Africa and in most parts of Latin America? is there a cultural explanation, linked todifferences in the permanence of urban migration, or is there a straightforward economic explanation? Clearly, themeaning and significance of ownership in different societies is an important factor in influencing tenure preferences.What economic advantages does ownership offer in different societies? is house-ownership a good investment, evenin self-help settlements? Little is actually known about the lousing market in low-income settlements: how difficult it isto sell houses in the absence of housing finance and the economics of self-help ownership. Some analysis of thecapital gains to be derived from low-income housing investment would be valuable.

In addition, there should be an understanding of the non-economic advantages of ownership vis-à-vis renting.Is it the dislike of landlords, the desire for personal security, the social status of ownership per se, the lack of otheroptions or yet unidentified reasons that motivate so many households to invest in housing? A comparison of thesituation facing owners and tenants in different cities would yield considerable insight on this question.

Page 80


Chapter VI. Priorities for researchThe characteristics of tenants and non-owners

Data should be gathered on the characteristics of tenants and how they compare with those of owners. Istenancy a function of age, family structure, social background, income or culture? To what extent are the very poorpredominantly tenants? Which households remain in tenancy, and is this a matter of choice or the result of a lack ofoptions? Almost nothing is known about the answers to such questions in most cities.

If relatively little is known about the nature of tenants, still less is known about other kinds of non-owners. Themost important group would be those who share houses with friends or kin. In some cities, sharers can make up asmuch as one fifth of the total population. Why do they choose to share rather than rent? Why is sharing morecommon in some cities than in others? Is sharing culturally determined, say; by the nature of extended familyrelationships or is it an outcome of the lack of housing options? Are particular kinds of households forced to sharewith kin, because they are not wanted by landlords? Do sharer households contain a high proportion of the very pooror those with many children? Is sharing a temporary phenomenon by which newcomers can become accustomed to thecity or by which young families can save money before investing in their own house? Alternatively, is it an emergencyresponse to a hostile economic or housing environment, for example, the necessary reaction to falling real wages or toa clamping-down on informal methods of land acquisition? At present, there is virtually no knowledge on this topic.

Page 81


Chapter VI. Priorities for researchLandlord-tenant relations

If it is possible, even less is known about landlord-tenant relations than about most other aspects of rentalhousing. This critical issue would repay attention because it is a potential source of social conflict.

First, is there any relationship between the form of landlordism and nature of landlord-tenant relations? Arelarge-scale landlords more commercially aggressive than small-scale landlords or less so? Are rental administrators,where employed, harsher than the landlords themselves - a familiar tendency in nineteenth century Scotland?(Daunton, 1983)

Secondly, what determines the length of the average tenancy? Does it matter whether tenants move seldom orfrequently? Do tenants suffer from a lack, of permanency or are they stuck in bad accommodation afraid to movebecause of a rise in rent? In the United Kingdom of the nineteenth century, middle-class tenants tended to have verystable tenancies, and working-class tenants very brief ones. In Scotland, very poor tenants were tied for one year totheir accommodation and envied English tenants their freedom of movement. (Daunton, 1983) It is uncertain whethertenants prefer to move easily. when their circumstances change, or whether they require permanency. According tothe circumstances, either can be advantageous for the tenant. In the United States of America, "when housing choicesare abundant and vacancy rates are high tenants who do not like their apartments vote with their feet: they move".(Dreier, 1984:262) The converse of this dissatisfaction with housing conditions will often be non-mobility. (Jones,1979) Knowledge is required of the constraints and opportunities facing tenants. Of course, different kinds of tenantswill have different preferences; those with high Incomes will presumably favour regular movement, while those withlarge families or having other characteristics which make them unwelcome to many landlords will presumably valuestability. Since governments frequently legislate to increase security of tenure, this would seem to be an importantsubject for study. In practice, do tenants in developing countries vote with their feet? Does rental legislation affect thelength of the average tenancy?

Thirdly, knowledge is also required of the selection of tenants. In particular, what kinds of families havedifficulty obtaining private rental accommodation? Without knowing the answer, it cannot be established whichhouseholds have the greatest priority for State help. It is also necessary to understand the logic underlyingpublic-sector housing allocation systems. A register should be established, containing details of how public agenciesin different cities allocate housing and the problems they face in managing their allocation system.

Where landlord-tenant relations are generally bad, the idea of establishing tenant co-operatives to manage theaccommodation seems to have some appeal. However, the record of co-operatives in developing countries is verymixed, and there is very little documentation about co-operatives operating in rental housing. Are rental co-operativesa good idea, have they worked anywhere and are they better at maintaining property than private or public landlords?

Very little is known about tenant organizations and urban protest. Despite the burgeoning literature on socialmovements, very little is available on tenant associations and the conditions under which they operate. Likewise, littleis known about the organizations which represent landlords. Information about both kinds of organizations would beuseful in understanding the politics of housing. It would also be useful in policy terms. If effective tenant and landlordassociations existed, information about the law and its functioning could be diffused through them. and this wouldimprove the implementation of any new legislation.

Page 82


Chapter VI. Priorities for researchRent controls

Knowledge is needed concerning the relationship between rents, prices and incomes. Such an analysis couldbe done relatively easily and cheaply, because many countries already produce series of data which have simply notbeen collated and analysed. The main need is to investigate whether rents rise in line with inflation, to find out howthey are linked to income levels and the minimum wage, and to examine how they are influenced by rent-controlmeasures. Such an analysis has to be broken down into at least three different income- groups.

Page 83


Chapter VI. Priorities for researchMaintenance and repair of the housing stock

The maintenance of housing is a problem In most inner-city areas and is likely to be an increasing problem inconsolidated self-help settlements with large number of tenants. How can landlords best be encouraged to maintaintheir property? Are tenants the best people to undertake small repairs? What examples are there of successfulschemes? Do repair funds and the like actually work? Insofar as public housing is concerned, is there anything to belearned from experiences of Hong Kong or Singapore?

Page 84


Chapter VI. Priorities for researchThe role of the law and the courts

To what extent are rental laws effective? Do landlords and tenants sign contracts, and to what extent are thesecontracts legally valid? To what extent do rents comply with rent-control edicts? Insofar as there are importantdifferences between cities in terms of the effectiveness of the law, how can the variations be explained?

Studies of rental disputes in the courts would be useful to establish both the evenhandedness of the judicialsystem and the extent to which different groups make use of the law. Where several systems of arbitration exist, itwould be helpful to know how effectively those systems operate. Do local-authority rent controllers orneighbourhood-arbitration procedures actually work?

Page 85


Chapter VI. Priorities for researchA list of priorities

All of the above topics require considerable research attention. However, the following issues probablydeserve highest priority.

First, a serious attempt should certainly be made to seek to understand the nature of the small-scale landlordand the reasons why he or she invests. This is a high priority, if any effort is to be made to stimulate rental-housingproduction by this group.

Secondly, knowledge needs to be acquired concerning the tenure preferences of low-income populations. Mostimportant is to establish the reasons why so many prefer to he house-owners. Establishing the bases which determinetenure preference will open up new opportunities for governments to change the parameters delimiting thosepreferences.

Thirdly, there is a need to understand the causes of good and bad landlord/tenant relations and the role thatthe law plays in improving or worsening those relations. Indeed, the role of the law in establishing the nature of rentalarrangements should have high priority.

Fourthly, the role of co-operative organizations in rental accommodation should be considered carefully. Whyare there so few rental co-operatives, and to what extent is this a feasible way of improving housing conditions,particularly in inner-city areas?

Page 86

· D. Contracts and legality E. Landlord and tenant organizations IV. Public policy A. Rent control o 1. The introduction of rent controls o 2.

Documents