About 40 per cent of Africa's population currently live in towns and cities, according to the UN Population Division estimates, and over 50 per cent will do so by 2025. While the overall population growth is expected to be 2.1 per cent a year between 2000 and 2030, despite declining fertility and the impact of HIV/AIDS, the urban population is expected to grow initially at 3.3 per cent. This conceals a considerable variation, with some urban centres increasing from 4 to 5 per cent per annum, while others grow at slower rates.1 Although presently Africa is the least urbanized region, by 2030 its urban population is expected to exceed the total population of Europe. Over half of the continent's urban population lives in urban centres comprising fewer than half a million people. Beyond these estimates, it is extremely hard to be precise; lack of resources, conflict and political difficulties mean that census data is patchy and unreliable. Some large countries like Nigeria and the Democratic Republic of the Congo have no reliable population figures. For example, although we think that Lagos and Kinshasa are among Africa's largest cities, we do not know how many people live there. In addition, economic crisis, conflict and drought lead to fluctuations in urban populations that are often not captured by census figures.

Conventional wisdom has it that the incidence of poverty and its severity and depth are greater in rural areas. Urban residents, it is alleged, benefit from the opportunities offered by urban labour markets, a greater availability of utilities and better access to education and health services. In the period of economic growth and public sector expansion in the 1960s, that was true in some countries: import-substitution industrialization policies led to growing employment opportunities; the public sector expanded; utilities installed during the colonial period had not yet broken down under the pressure of rapid urban growth; and committed post-independence governments found it easier to deliver free social services in accessible urban areas. However, the oil price increases of the 1970s, followed by the debt crisis and the adoption or imposition of structural adjustment policies (SAPs) in the 1980s, soon changed the situation.

Urban people, inevitablymore integrated into the cash economy than those living in rural areas, were hard hit by inflation and private-sector stagnation, even before the policies typical of the SAP "short sharp shock" treatment, such as price decontrol, job shedding by public and private sectors alike, reduced expenditure on services, reduction or elimination of subsidies and increased user charges. Driven by a hypothesis that "urban bias" in policy and investment was an important contributor to the failure of African countries to develop, the supposedly privileged urban population was to have its advantages eliminated. SAPs were, therefore, deliberately anti-urban in their effects. The extent to which there really was urban bias, whether it persists or has an adverse impact on economic development, remains controversial. Some allege that it continues, but the limited research undertaken has been unable to demonstrate the magnitude of the effects of urban bias or their implications for growth and poverty, since investment returns may also vary between urban and rural areas, being potentially greater in the former.1 Unfortunately, in the 1980s, SAPs were generally blind to equity implications. Even when attempts to collect poverty data were launched, these were based on national sample surveys and could rarely be disaggregated sufficiently to permit intra-urban analysis.

Average urban incomes are no doubt higher than average rural incomes. The proportion of people living in poverty is smaller and the standards of service provision per person are higher in urban than in rural areas. However, available figures indicate that income and consumption inequality is generally greater within urban areas. The presence of services does not mean that the poor can access them, and rates of impoverishment were significant in some urban areas during the 1980s. Many analysts believe that price decontrol, public sector downsizing and subsidy reduction impoverished large sections of the urban population, whereas rural people could always produce their own food even though they never had good access to subsidized services. The incidence of poverty is generally higher in secondary cities than in the largest urban centres, but where data are available they show an increase in both, for example in Côte d'Ivoire between 1985 and 1995.2 Urban people who were adversely affected included the already poor and the so-called "new poor"-those who lost their jobs or whose declining real incomes were no longer sufficient to support a family.

Although economic growth has picked up in some African countries since the 1990s and national trends show some decrease in the incidence of poverty, many urban dwellers continue to live in extreme poverty. Official figures show that half or more urban residents are poor. Where they show a lower proportion, this may reflect a failure to allow for higher cost of living in urban areas, where residents need to pay for rent, transport, water and sometimes the use of public toilets.3 It is argued that if urban life realities are taken into account, the situation of the poor is as severe as in many rural areas. Poor urban people have incomes insufficient to meet basic needs, lack assets and secure tenure, live in poor and overcrowded housing conditions, have inadequate access to infrastructure and services, have no social safety net and are powerless to influence decision-making.

It is estimated that about 60 per cent of Africa's urban labour force are in the informal sector, many earning low and irregular incomes. For many, the need to diversify income sources to improve security hinders investment in their enterprise to increase its scale and profitability. In many urban centres, households' economic and social links straddle urban and rural areas. In some cities, for example, many are split households, in which the husband migrates for work while the wife continues to farm. Elsewhere, especially when successive generations have been urban dwellers and chronic poverty has eroded the ability of poor households to maintain rural ties, claims to rural land and kinship links have declined.

Most poor and many not-so-poor urban dwellers live in informal settlements, known as "slums". According to the United Nations Human Settlements Programme (UN-HABITAT), these are characterized by the lack of basic services, inadequate housing, high densities and overcrowding, unhealthy living conditions, insecure tenure, poverty and social exclusion. In 2001, it was estimated that 72 per cent of Africa's urban population lived in slums and the annual average growth rate of the slum populations was at 4.5 per cent between 1990 and 2001, nearly double the total urban population growth rate (2.7 per cent per annum).4 These living conditions are reflected in indicators of access to services and health outcomes, which need to be disaggregated within cities and by city size. Overall in sub-Saharan Africa, 28 per cent of urban households lack safe drinking water, and about 50 per cent or more do not have adequate sanitation.

The average proportion of urban people lacking access to three basic services-potable water, improved sanitation and electricity-has been estimated at between 44 and 63 per cent of the poor and 34 per cent of the non-poor. Health indicators demonstrate the impacts of poor living environments. For example, under-five mortality and morbidity rates are higher in Nairobi's slums than in the entire city and equal to or higher than in Kenya's rural areas-11 per cent of children living in slums die before they reach the age of five, compared to less than 1 per cent in non-slum areas.5 High mortality and morbidity rates and poverty are exacerbated by the HIV/AIDS epidemic, the prevalence of which is invariably higher in urban than rural areas. In high-prevalence locations, it affects every family directly or indirectly through illness, death or an increased burden of care for orphans and the sick. The vicious cycle of poverty and HIV/AIDS is reinforced by risky sexual behaviour. It has been found that poor women in Nairobi's slums engage in sexual activity at an earlier age and with a larger number of partners than elsewhere due to high unemployment and low and unstable incomes.6 In addition to gross inequalities within cities, provision in smaller cities and towns lags behind that in the largest cities.

Life for the urban poor is unpredictable: piped water is unreliable, informal settlements are susceptible to flooding, illness and death may affect people at any time, and eviction, crime and violence are prevalent. Many believe that privations of slum life have a destabilizing effect on households, leading to a vicious cycle of breakdown and impoverishment. The prevalence of slum conditions is a testament to the inability of poorly-resourced local government and public-sector agencies to keep pace with rapid urban growth.

The first step in addressing these challenges is to improve understanding of the realities of life in African cities: how the urban poor understand poverty and deprivation, how they attempt to develop secure and adequate livelihoods and what holds them back; and, in conjunction with other urban actors, how they have evolved ways of delivering and accessing goods and services they need. In contemporary African cities, only a minority of workers are in formal employment. In practice, households support themselves by a mixture of income-generating activities, mostly in the informal sector, and self-provisioning, for example urban agriculture, building on the resources to which they have access, such as labour power, education, health, land and housing, basic services, transport, savings and credit. Their assets and ability to undertake various livelihood activities are differentiated by individual characteristics-age, gender, religion, ethnicity-and the composition of households.7 An understanding of how people use their limited resources to cope with adversity or improve their well-being, and how wider economic changes and policies can erode people's resources and hinder their ability to develop secure livelihoods, can point to appropriate policy approaches.

First and most important is to avoid damaging policies and regulatory practices. The availability of livelihood opportunities depends above all on economic growth. However, experience demonstrates that if the distributional effects of economic growth are not considered, its potential benefits in terms of employment opportunities and poverty reduction may not be felt, while particular policies may actively hurt the already disadvantaged. It is clear that relying solely on the market or user charges to provide services exacerbates poverty. Falling school rolls and deteriorating health indicators in the 1980s led to recognition that without State intervention, access by the poor will fall, and that regulatory environments not conducive to business will hinder investment and economic growth. Similarly, at the local level, pricing policies for services may exclude those most in need, and damaging regulatory policies, such as harassment of informal-sector enterprises, evictions and demolitions of slum settlements, exacerbate deprivation.

Second, there is scope for policies to support economic growth and the livelihoods of the urban poor. The public sector has a key role in facilitating economic growth by ensuring macro-economic stability, instituting a more conducive regulatory environment for business and improving service delivery. Policies can provide modest support to livelihood strategies of the poor and enable them to improve their well-being and accumulate resources that can increase their resilience in the face of shocks and stresses. The scope of households for investing in human, physical and social capital is determined by the educational, health and environmental services available, security of tenure and the right to organize and exercise political voice. For example, even if water is charged for, more efficient provision resulting in decreased prices releases resources to meet other needs. Also, access to financial institutions can enable people to borrow, and environmental hazards can be reduced by working with informal entrepreneurs and developers to improve practices and standards.

Third, for supportive and regulatory policies to be designed and implemented, urban governance systems appropriate to African circumstances of limited financial and administrative capacity, as well as limited state legitimacy and private sector development, are needed. These need the capacity to foster economic development and provide services to the urban populations. Larger cities have the most diversified economies and are best placed to attract non-primary domestic and foreign investment. They have great potential to raise a significant proportion of their revenue needs, provided they have the legal basis and political autonomy to do so. National governments have traditionally let their fear of alternative political and administrative power bases outweigh their willingness to encourage the development of accountable and effective city management, leading most urban governments to have unfunded mandates. They depend on central government for funds and therefore lack wide support and legitimacy in the eyes of urban residents and businesses. The economic health of smaller cities and towns depends on their functions in surrounding regions, such as provision of financial and business services, higher level of education and health services, administration, the supply of agricultural inputs and consumer goods, markets for agricultural produce, agro-processing, etc. While the scope for local revenue generation in small centres is probably less than in larger cities, they are no less in need of predictable central-local financial transfers and democratic local government structures.

It is clear that without effective urban management in Africa's towns and cities, the Millennium Development Goals will not be achieved-especially the MDG targets to attain universal primary education, reduce child mortality, combat HIV/AIDS, malaria and other diseases, halve the proportion of people without sustainable access to safe water and improve the lives of at least 100 million slum dwellers by 2020-unless they are met in urban areas. Indeed, tackling the issues in urban areas has advantages because, although concentrations of people and waste in the absence of infrastructure increase environmental health risks, economies of scale and proximity can also facilitate cost-effective service provision.8

What is the way forward? It is clear that if economic policies deliver their anticipated benefits and if economic growth resumes, we can expect increased urbanization, since activities associated with urban areas are likely to play important roles in economic recovery and diversification. Also, if this positive scenario comes about, Governments should aim to enable poor urban residents to take advantage of economic opportunities by equipping them with education, skills, good health, a secure residential base and adequate public transport. Increasing their assets, analysts suggest, enables them to improve the adequacy and security of their livelihoods. Human capital improvements, in particular, require better basic infrastructure and services, such as water, sanitation, solid waste management, public transport, land registration, education and vocational training, and health care and financial services. Redesigning approaches to urban management and supporting improvements to infrastructure and services require competent governance. Research demonstrates that competent governance also requires an organized civil society, supporting residents' efforts to exercise their political rights, a political commitment at central and local levels, and adequate resources from a combination of local revenue and central-local transfers.9

On the face of it, few African countries and urban centres have seen much progress towards developing responsive and accountable political and administrative structures or a capable bureaucracy. However, resourceful residents, public-sector employees and local officials have evolved ways that often work on a large scale and are widely understood and accepted, including systems of informal residential land delivery, retailing and market trade, social networks based on kin, religion or neighbourhood, water vending and governance of neighbourhoods and markets. These evolving "home-grown" systems should not be regarded as "illegal" or the cause of urban problems or barriers to their solution. Instead, they and their relationships with State structures should be understood and built upon, and their weaknesses mitigated.

Notes

1. Montgomery, M.R., Stren, R., Cohen, B. and Reed, H.E. (2004) Cities Transformed: Demographic Change and its Implications in the Developing World, London: Earthscan.

2. Grimm, M., Guénard, C. and Mespleé-Somps, S. (2002) "What has happened to the urban population in Côte d'Ivoire since the 1980s? An analysis of monetary poverty and deprivation over 15 years of household data", World Development, 30(6), 1073-95.

3. Satterthwaite, D. (2004) The Under-estimation of Urban Poverty in Low- and Middle-income Nations, London: International Institute for Environment and Development, WP on Poverty Reduction no. 14.

4. UN-HABITAT (2004), The State of the World's Cities 2004/5, London: Earthscan.

5. African Population and Health Research Centre (2002), Population and Health Dynamics in Nairobi Informal Settlements, Nairobi: APHRC.

6. Zulu, E.M., Dodoo, F. N. A. and Chika-Ezeh, A. (2002), "Sexual risk-taking in the slums of Nairobi, Kenya", Population Studies, 56, 311-23.

7. Rakodi, C. with Lloyd-Jones, T. (eds) (2002), Urban Livelihoods: A People-centred Approach to Reducing Poverty, London: Earthscan.

8. Bolnick, J., Kayuni, H.M., Mabala, R., McGranahan, G., Mitlin, D., Nkhoma, S., Oucho, J., Sabri, A., Sabry, S., Sattherthwaite, D., Swilling, M., Tacoli, C., Tambulasi, R.I.C. and van Donk, M. (2006), A Pro-Poor Urban Agenda for Africa: Clarifying Ecological and Development Issues for Poor and Vulnerable Populations, London: International Institute for Environment and Development, Human Settlements DP Series, Urban Change-2.

9. Devas, N., Amis, P., Beall, J., Grant, U., Mitlin, D., Nunan, F. and Rakodi, C. (2004), Urban Governance, Voice and Poverty in the Developing World, London: Earthscan.