16 Conclusion

Carole Rakodi

Abstract

Ce dernier chapitre fait ressortir certains des principaux ths et des principales conclusions, en mettant d'abord l'accent sur les rapports entre mondialisation et urbanisation en Afrique, puis sur les caractstiques du processus d'urbanisation des plus grandes villes, et enfin sur les questions qui se posent et les expences acquises en mati de politique et de gestion. Les liens avec l'nomie mondiale par le biais des investissements des socis transnationales vent surtout constitupar les exportations de produits primaires, m si le peu que les transnationales investissent dans les manufactures et les services se concentre dans les villes. Le consumsme mondial est plus envahissant, favorisar le contrque les transnationales exercent sur les mas et il en va de m pour l'influence des agences internationales. C'est pourquoi l'autonomie nomique et politique de l'Afrique est handicappar sa position dans le monde. Les effets des conditions en mati de politique impos par les donateurs et les programmes de pr se font fortement sentir, directement et indirectement, dans les zones urbaines. Les conditions des relations entre les pays d'Afrique et l'nomie mondiale et la fa dont elles ont luvec le temps vent un des drminants essentiels des modes d'urbanisation, mais pas le seul.

Malgrn ralentissement du rythme de la croissance urbaine, notamment dans la plupart des villes les plus important es, au cours des dernis dnnies et en particulier durant les ann les plus sombres de la rssion, ces villes continuent de se propager n taux plus vue celui de la croissance naturelle. Les tendances de l'nomie urbaine, notamment la stagnation ou le rssement du secteur officiel de grande envergure au profit du secteur informer et du travail temporaire en m temps que la diversification et l'nouissement du secteur informer, vent assez mal comprises. La tendance ne prminance croissante de mnismes officieux, voire illux, pour la division et l'amgement des sols se fait sentir dans tout le continent. Face a rssion et 'augmentation des inlit les mges urbains ont ri de diverges fas, tirant parti des strates de migration, des liaisons entre villes et campagnes, de l'esprit d'entreprise, du travail des femmes et de la culture des sols urbains disponibles. La liblisation politique s'est traduite par une lution de la recherche, qui met moins l'accent sur les aspects ethniques en eux-ms, les classes, la nature de l'at, les rmes autoritaires et les relations de patron lient que sur la rrgence de la soci civile. Le chapitre dit, document 'appui, le rdes organizations non-gouvernementales, et autres associations de la soci civile, dans la vie et la gestion urbaine, quoiqu'il soit encore difficile de discerner si la proliftion de ces associations va avoir un impact ong terme et une signification politique majeure. Les mesures prises jusqu'rnt pour la gestion de la ville se vent aves incapables de relever le d de l'accration de la croissance. L'ec de la dntralisation, de la rrme des structures et procres administratives et de la rsion des politiques, exacerbar la rssion et les programmes d'ajustement structurel, a provoqune nette drioration de la qualite la vie, surtout chez les pauvres, et de la capacites organismes urbains. II va falloir changer les rmes politiques et administratifs des villes ainsi que les politiques et procres de gestion urbaine, en particulier en mati de crion de ressources, de dloppement nomique, d'apport d'infrastructures et de services, de logements et de sols, de planification des sols, de rementation et d'amoration de l'environnement. Enfin, il faudrait drminer queues nouvelles donn amor et queues nouvelles recherches de fond nous permettraient de mieux comprendre l'nomie et les socis urbaines, les marchimmobiliers, les mesures politiques et les effets des politiques et procres de planification et de gestion.

Introduction

The first part of the concluding chapter draws out some of the main themes and findings of this volume, concentrating first on the relationships between globalization and urbanization in Africa, secondly on the characteristics of the urbanization process and the largest cities, and thirdly on policy and management issues and experience. Finally, some of the outstanding research needs and priorities are identified.

Global forces and Africa's future

Whatever the terms and outcomes of its integration into the world economy, world politics, and the cultural-ideological world system, Africa is, like it or not, part of that system. On balance, as discussed by Rakodi (chap. 2) and Simon (chap. 3) in this volume (see also Sklair, 1991; Brown, 1995), Africa has in the past and continues in the present to lose more than it gains: for much of the continent, dependency, marginalization, and a lack of policy autonomy are the main outcomes of its incorporation. The organizational framework for the global economy, as discussed in this volume, is comprised of two interlocking sets of institutions: the transnational corporations (TNCs) engaged in exploitation of primary resources, manufacturing, and services, as well as the institutions that enable them to carry out their activities, such as the major stock exchanges; and the international agencies, particularly the International Monetary Fund IMF and the World Bank, pulling in their wake smaller multilateral and bilateral agencies. External forces, be they TNCs or international agencies, call the shots, heavily influencing patterns of investment, preventing the write-off of debts, and dictating particular economic policy packages. Africa's bargaining power is so limited and the developmental effects of global economic changes and international agency policy conditionality so negligible for much of the continent that reductions in dependency, participation in growing and prosperous sectors of the global economy, and dynamic economic growth and development seem remote possibilities.

Though large parts of the economies of African countries, both rural and urban, do not come under the control of TNCs, almost none is immune from the effects of donor conditionality and aid, and global consumerism has penetrated even urban slums and remote rural areas, thanks to the power of the media and advertising. Links with the global economy are particularly manifest in the cities, which are the locations within countries for much of whatever TNC investment in manufacturing has occurred, TNC administration and services, and the offices of international agencies and diplomats. In addition, their populations have easier access to imported goods and the media. Even in the cities, however, large parts of the economy, the spatial fabric, and the socio-political structure are not directly affected by the operations of TNCs, although many of these have been indirectly affected by the terms of the continent's integration into the world economy and culture.

At present, foreign direct investment continues to be interested primarily in the exploitation of primary resources (Seddon, 1995). There seems little prospect of investment in the more labour-intensive stages of the manufacturing process in profitable high-technology sectors flowing into Africa as it has into East and South-East Asia, leading Rogerson (chap. 10) to term Africa's current state "stalled globalization." Nevertheless, in some interesting quarters, including Japanese trading companies, Afro-American investors, and South African manufacturers, there is a spark of interest in further investment, perhaps because high profits may be made where the potential is unexploited and the risk high. However, poor transport and communications, inadequate infrastructure, labyrinthine and ineffective bureacracies, and uneducated labour forces continue to deter large-scale foreign investment. Recent recognition of this by national governments, city administrations, and international agencies is already starting to bring about changes, and these are likely to accelerate, at least in some countries, in the next decade.

In addition, the increased cultural hegemony of global consumerism seems unstoppable, as transnational producers penetrate global markets with the aid of transnational-controlled media - media that have, with satellite technology, become more accessible to remote and low-income populations, as well as to the urban te, in every African country. The demand for "global" consumer goods, from televisions to Coca Cola (see Simon in chap. 3), combined with IMF insistence on liberalization of import controls, has biased the composition of imports away from those needed for productive purposes towards consumer goods. Although import controls were undoubtedly in need of reform and streamlining, given the limited capacity of African countries to earn foreign exchange, such a wholesale abandonment of policy objectives in favour of free market economics has been extremely damaging to countries' fledgling manufacturing bases and other sectors such as transport. Even if liberalization is slowed and some protection sustained, as recommended, for example, by Riddell (1993), penetration by global consumer goods industries into African markets is likely to continue. Because of the coincidence of satellite-based media technology and political liberalization, it seems likely that the use of the media to foster consumerism, disseminate global culture, and even spread religious views will outweigh the potential advantages of more accessible media for educational and developmental purposes. In Ghana, for example, the government-controlled press, which might choose to use some of its capacity for civic education, is under attack in the name of political liberalization in favour of an ostensibly independent but in practice party-political press. However, increased access to global channels of communication also creates opportunities for easier interaction between academics and professionals, and especially between non-governmental organizations (NGOs), with unpredictable results.

The political game in Africa has changed, with the end of the Cold War, the collapse of communist regimes in Eastern Europe, and the association made by international agency economists and others between economic liberalization and political liberalization on a European-North American model. Although interference by the colonial powers, especially France, persists, there are signs of an increasing reluctance to prop up patrimonial rulers merely because of their Western sympathies, first because it is clear that development has not resulted from often protracted periods of such rule, and secondly because such personal authoritarian rule does not fit easily with the West's advocacy of greater political competition elsewhere in the continent. Although much of the explicit or implicit political policy conditionality is crass and superficial, and the type of political liberalization advocated may have economically and politically destabilizing effects, it is opening up political spaces that were constricted during the most authoritarian periods of one-party and military rule. Changes in international political forces are, therefore, provoking a widespread reconfiguration of domestic politics, with particular implications for urban areas that will be explored further below.

The pervasive but contradictory influences of the international agencies in Africa have been referred to again and again in this volume. The institutions themselves have, as a result of the far from uniformly successful and sometimes devastating effects of their policies, been under attack mainly, but not only, from recipient countries. Although criticism has led to some shifts in the policies advocated and ways of operating (Green, 1993; Ravenhill, 1993; Green and Faber, 1994), the basic rationale of the system and its neoclassical economic theory underpinnings remain unchanged. Until decision-making power in the institutions is more evenly distributed between backers and recipients, the hegemony of a theoretical world-view that is based on faith rather than empirical evidence is challenged (George and Sabelli, 1994), improved economic understanding is accompanied by more nuanced political and social analysis, and their mode of operation is less centralized and autocratic, the relationship between African countries and these institutions seems unlikely to change, despite attempts to negotiate more appropriate reforms (Aboyade, 1994).

However, it is wrong to be wholly pessimistic about the impact of global forces on the African continent: Egypt has benefited from its roles in Cold War politics and as a labour source for Middle Eastern countries, Libya from high oil prices, and other North African countries from their proximity to Europe and ties with the Arab world. Although their economies still have structural shortcomings, poverty is widespread, and urban problems unsolved, these benefits are manifest in economies that are more prosperous and diversified than in much of sub-Saharan Africa (SSA). In southern Africa, the achievement of political freedom in the continent's industrial giant has dramatically changed domestic and regional prospects for economic growth and development with, as yet, unpredictable consequences. Some smaller countries, notably Botswana, have achieved significant economic and social progress, and others, such as Uganda and Mozambique, have achieved peace after periods of civil war. Although Africans have in the recent past had little success in advancing their own alternatives to the stabilization and structural adjustment policies of the IMF and World Bank, convincing policy statements are starting to emerge, notably from the UN Economic Commission for Africa (Brown, 1995). Cooperation between African countries has been deterred by, for example, the concern of politicians in the immediate post-independence period to consolidate and sustain the nation-states based on artificial boundaries in which they took power, the varying colonial heritage, ongoing ties with the colonial powers, late struggles for independence in parts of the continent, and the effect of international Cold War politics. However, there are signs that some of the more recent groupings, such as the Southern Africa Development Conference (SADC), will be more enduring, thus potentially providing a basis for stronger and more diversified regional economies (Adedeji, 1993; Brown, 1995).

Urbanization

Urban population growth

The terms on which African countries interact with the world economy and the way in which these have changed over time are a major, but not the only, factor determining patterns of urbanization. The inherited colonial pattern of urbanization, designed to facilitate the exploitation of resources and administrative and political control, was developed from scratch in parts of the continent and overlain on older patterns of urban development in others, as sketched in chapter 2. The asymmetrical pattern, centred on ports, large-scale European commercial farming areas, mines, and the transport routes connecting these, has persisted. Indeed, the continued location of most capitals in the former colonial capital, accompanied by the state-centred model of development and centralized administrative and political systems that characterized countries in the decades after independence, and the continued dependence on exports of primary products and imports of most manufactured and all capital goods, reinforced the inherited pattern. Similarly, the constraints on development of secondary cities and small towns arising from colonial neglect of peasant agriculture and the exclusion of indigenous populations from many business activities were reinforced in the years after independence by continued neglect of peasant agriculture, state domination of agricultural input supply and marketing, and the obstacles placed in the way of budding entrepreneurs by the general shortage of credit, by poor infrastructure, and by heavy-handed government regulation. Making up in some cases for colonial/settler restrictions on rural-urban migration, and fuelled by the urban bias of investment and state activity, the largest cities grew extremely rapidly in the years immediately after independence. Since then, rates of urban growth have slowed down but, insofar as it is possible to tell in the absence of reliable and regular censuses, cities continued to grow at rates above already fast rates of natural increase. As discussed by Simon in chapter 3 and Potts in chapter 13, although there is some evidence of even slower rates of growth in the worst crisis years, typically in the 1980s, and in the urban settlements whose economic base has been hardest hit by recession and liberalization policies, it seems that such slow-downs were temporary. Even modest rates of growth in the largest cities, of course, represent large absolute population increases, and may reflect continued net in-migration, because there is evidence from some countries of lower fertility among urban populations.

The continued growth of large cities is partly due to the natural increase of their own populations and this is likely to decrease only slowly, despite the as yet unpredictable effects of AIDS. The continued attraction of the cities for rural migrants, in a context in which recession and liberalization policies have had an adverse effect on public and large-scale private sector employment and wages, while simultaneously favouring rural production, is more puzzling. In addition, urban management systems are generally unable to keep pace with infrastructure needs and, in some cases, have all but collapsed. Continued rural-urban migration in such apparently discouraging circumstances throws doubt on the validity of earlier models of migration and the urban economy. The experience of cities such as Lagos and Kinshasa, documented in this book, shows that it is unrealistic to anticipate any significant or permanent slow-down in rates of urban population growth. And should economies begin to recover, as a result of increased world demand for Africa's products or structural adjustment policies, cities are likely to grow even more rapidly. Although the attention to rural development and the agricultural sector that characterizes current policies in some countries, such as Ghana, may give an impetus to the growth of second- and third-tier settlements, the continued desire for industrialization and the political importance of urban populations will fuel large-city growth and will (and should) not permit urban neglect.

Urban economies

What is clear from the trends of urban development in the 1980s and 1990s is that our understanding of the economic dynamics of African cities is partial. In the 1970s, when the existence of an informal sector providing small-scale economic opportunities for the working poor was officially acknowledged, it seemed clear that demand generated by those working in the formal sector was what sustained informal sector activities. Indeed, Beavon's account of Johannesburg (chap. 5) and Dubresson's of Abidjan (chap. 8) in this volume still portray a dominant formal sector and an informal sector largely dependent on demand generated by formal economic activity and lacking an independent dynamic of its own. Rogerson (chap. 10), however, documents the increasing predominance of informal employment in many cities and draws on recent work, especially by the International Labour Office Jobs and Skills Programme for Africa (JASPA), to increase our understanding of the sector.

In both the global and domestic economies, it is clear that there are both formal and informal economic activities. Examples of informal activities in the global economy might include drug trafficking and smuggling. However, it is also clear, first, that there are continua within both sectors from illegal to legal and from profitable to marginal activities, which are not captured by the dualist conceptualization of the urban economy, and, secondly, that current trends include informalization and reduced dominance of the formal sector. There are, Rogerson notes, various layers of informalization, ranging from enterprises that simply operate without licences, pay no tax, or occupy a particular site without permission, to major corruption and crime. In addition, a distinction must be made between those enterprises that are dynamic, profitable, and have a potential for growth, and those that are involutionary - survivalist enterprises that help people cope with economic shocks but that are overcrowded and unprofitable. What is also clear from studies of recent changes in the organization of production and services is that increased use of subcontracting and processes of deskilling and casualization are creating forms of production that do not fit the common understanding of formal sector enterprises, shading as they do into small-scale, unprotected, and unenumerated sectors such as commission selling and home-work. Some of these changes, he suggests, create opportunities for clusters of small and medium enterprises that can then develop a growth dynamic of their own, although not yet in Africa.

During the 1980s open unemployment became more significant and visible, especially among young people. This illustrated not only the decline in formal wage employment but also, according to some authors, the growing inability of the informal sector to absorb more workers (Zeleza, 1994). However, with the decline in real public sector wages in the 1980s making them unimportant components of income for many households, not only is it less common than it ever was for households to rely solely on the formal sector for their livelihoods, but also, in many cities, it is apparent that the demand generated by either formal enterprises or formal sector wages is insufficient to explain the size, persistence, and diversification of the informal sector. Thus, not everyone agrees that the informal sector lacks potential. A study of five SSA countries (Botswana, Kenya, Malawi, Swaziland, and Zimbabwe) estimates that during the period from 1981 to 1990 over 40 per cent of the labour force increase was absorbed by small-scale non-primary enterprises, mostly through starts of new businesses but also by enterprise growth and graduation to the intermediate size category (50+) (Mead, 1994). Although less than a quarter of these were in the urban areas and the study was unable, because of the difficulty of collecting such information, to determine the level of profitability of the businesses surveyed, it implies a significant and dynamic small-scale sector, in which high failure rates amongst new businesses are more than compensated for by the number of start-ups. The jury is still out on the role and dynamics of the small-scale informal sector and its relationship to a declining or stagnant and partly informalized large-scale formal sector. What is certain is that available studies and published statistics are an inadequate base for analysis and understanding in either economic or spatial terms.

Economic activities occupy different areas of economic space, but they also have a physical manifestation. Whereas the organized employment centres of industrial areas and central business districts (CBDs) are well established in most African cities, and informal activities have been ubiquitous in some, the concentration of clusters of informal enterprises in particular urban spaces, as well as their proliferation throughout the city regardless of zoning, are more recent phenomena in many, as liberalization of both economic policies and restrictive regulations has proceeded. The significance of site and location to large-scale and formal enterprises is reasonably well understood, but the role of site, location, and proximity to companion or competing enterprises and customers for small-scale enterprises has been little studied.

Property markets

The mechanism through which economic and other activities occupy space and become manifest physically is the property market. My review of evidence on the real, as opposed to the theoretical, functioning of urban property markets in African cities in chapter 11 illustrates the prevalence of informal but not ill-organized transactions, although to a differing extent in different cities. Cities thus range from Harare, with universal private tenure and a formal land market, at one extreme; through the cities where the state struggles to maintain control over subdivision, issue of title, and construction in the face of increasingly prevalent illegal subdivision of privately or communally held land; to Kinshasa at the other extreme, where agents of the state have "privatized" land supply. The squatting that was prevalent in many cities in the 1960s and 1970s was less anarchic than it appeared to many analysts, drawing as it did on understandings of property rights embodied in communal tenure systems. In addition, access to urban land through the legitimate exercise of communal tenure rights was important, especially in cities that predated colonial rule. Both these ways of obtaining land have changed as urbanization has become more established, so that the predominant means of getting access to land in many cities today is through the illegal subdivision of legally owned communal or private land. Increasingly, those with claims to land have taken the law and procedures into their own hands, as the capacity of the state to facilitate and regulate land transactions has declined and land markets have become increasingly commercialized. Clearly, just as with most small-scale activity, some accommodation with the prevalent mechanisms for land and property development must be made by the state. This point will be returned to in the discussion of urban management below.

Social structures and processes in cities

As most clearly recognized by francophone analysts, land transactions and their outcomes, as well as the concepts of property on which they are based, reflect not merely economic but also social and political relations. Thus the claims of members of kinship networks to usufruct rights in land held under communal tenure, or of lineage members to accommodation rights in family houses, as in Ghana, are rooted in pre-existing social structures. The adaptation of social processes and patterns to the demands of urban development concerns both Aina and Potts, as well as a number of the authors of the city case-studies in this volume. While Aina in chapter 12 is more concerned with social networks and their intersection with political organization, Potts (chap. 13) focuses on the economic and social basis for urban livelihoods and thus shares an area of concern with Rogerson. Drawing on evidence from many parts of the continent, she demonstrates the increasing poverty and inequality that have accompanied recession and structural adjustment (see also O'Connor, 1991). Although the deliberate anti-urban bias of structural adjustment policies was in some respects justified, their economic rationale and crude design failed to recognize both that urban populations are differentiated and that reduced investment in infrastructure and education would in the longer term adversely affect both existing economic activities and the longer-term potential for recovery and diversification. As a result, the already disadvantaged in urban areas have been affected worst, although many of the waged lower middle class have also been hard hit. Cairo is typical:

Cairo's overcrowdedness, deteriorating physical infrastructure and public services are compounded for the majority of its population by glaring inequalities of power and wealth. As there is struggle for Cairo's soul, there is an even more intense struggle for the limited resources and privileges. Cairo's elite (top 5 percent) in recent years has been oblivious to the fate and conditions of the majority of the rest of the city's poorer quarters.... It is the middle classes, especially the lower rungs, which feel the squeeze, and the youngsters are steaming with frustration and anger. Much of Cairo's future, and hence of Egypt's, may very well lie in their hands. (Eddie Ibrahim, 1987, pp. 225-226)

In these conditions, some analysts suggest, crime and prostitution increase (see Obudho on Nairobi in chap. 9), while Tourt al. (1992), in a study that emphasizes increased inequality and poverty, discriminatory investment in infrastructure, and the precariousness of life for the majority in Abidjan, also draw attention to a rise in xenophobia in that cosmopolitan city, which they attribute to the need to find scapegoats for economic hardship.

The reactions of urban households have been diverse and ingenious, drawing on migration strategies, rural-urban links, entrepreneurial skills, female labour, and access to unused urban land. Cultivation, of both rural land, to which many urban residents continue to have rights, and undeveloped urban land, has been an important component of the livelihood strategies for many urban residents, as stressed by both Rogerson (chap. 10) and Potts (chap. 13). However, Rogerson acknowledges that much of the evidence on the extent of urban agriculture is impressionistic; that the dividing line between urban cultivation, which is primarily for household consumption, and pert-urban agriculture, which may be subsistence or commercial, is unclear; and that information on its environmental implications is lacking. It is clear from the city case-studies in this volume that the extent of urban agriculture varies: unused land that can be cultivated is less available in higher-density cities (such as Cairo) or parts of cities (such as the central areas of Lagos), in some of the cities with more commercialized land markets (although not Nairobi), and in cities where residents are predominantly renters rather than house-owners. Although most contributors acknowledge that urban cultivation and sometimes animal rearing may be a significant element in the livelihood strategies of many (not always poor) urban households, the contribution of such food production to the total city food supply is unknown and suspected by some (for example Piermay in the case of Kinshasa - chap. 7) to be limited. Analysis of the potential of and constraints on urban agriculture is lacking. The use of land for cultivation, especially on a permanent basis, has opportunity costs, while the effectiveness of support to urban agriculture compared with support to pert-urban agriculture or other types of small-scale enterprise is not known. In addition, there has been little investigation of the environmental impacts of urban agriculture, both cultivation and livestock keeping, or the effects of air and water pollution on the products of urban gardens.

The last major aspect of urban development to be dealt with in a thematic chapter and to varying extents in the city case-studies is the social and political organization and processes that have come increasingly to be grouped under the term "civil society." Civil society is a general term that refers to social phenomena beyond formal state structures but not necessarily disconnected from the state (Woods, 1992). Subject to longstanding debate in the academic literature, there seems to be no consensus on its precise meaning, boundaries, or relationship with the state (Aina in chap. 12; see also Woods, 1992; Mamdani, 1992). In African societies, certain components of society and politics other than the state have been important focuses of attention, in particular ethnicity and, more recently, class, as well as specific groups such as trade unions. These have not, on the whole, been theorized in terms of an over-arching concept of civil society that can be set against that of the state. Indeed, more recent analysts seem to disagree on which types of association should be regarded as part of civil society. Woods (1992), for example, appears to exclude pre-colonial ethnic and kinship groups, asserting that ethnically based associations in urban areas, which were formed in the social space created by colonial urbanization, mark the emergence of civil society. The forms of political power and leadership that have predominated in post-independence Africa and the state-centred model of development that was adopted have resulted in most attention in the political science literature being given to the nature of the state, the characteristics and shortcomings of patrimonial rule, the rise of authoritarianism, and the prevalence of military regimes (Healey and Robinson, 1992). Relatively little attention has been given to either civil society or the particular characteristics of urban politics.

In part this can be explained by the nature of post-independence regimes and the political and developmental models they adopted. Authoritarian rule and state-centred development were accompanied by restrictions on the formation of NGOs and often repression of groups critical of the prevailing political and economic order. In addition, urban populations benefited from the jobs created in an expanding public sector and lacked avenues for accumulation that did not depend on access to state resources. Not until the 1980s, as the economic crisis threatened the accumulation strategies of the urban te and middle class, did critical voices start seriously to challenge the political structures and economic policies (Woods, 1992; Chazan and Rothchild, 1993). This is as true in the cities of north Africa as elsewhere in the continent. Findlay (1994, p. 189) finds it unsurprising that cities in the Arab world have become the focus of unrest and protest:

[It is in cities that] inequalities within society are most starkly displayed... and the contested images of Arab society are most powerfully portrayed in terms of the symbols of the built environment. And it is in the urban riot or demonstration that those without power can most forcefully challenge political decision makers about the future directions of state development policies.

The voluntary associations that have proliferated to articulate various interests, including professional associations, religious groups, and women's associations, interact with ascriptive groups including ethnic and kinship associations. Their interaction with each other and the state determines the characteristics and boundaries of civil society (Chazan and Rothchild, 1993). Although social structures based on ethnicity are associated predominantly with rural areas, and the emerging pluralistic associational movement is associated broadly with the urban middle class, overlaps are more significant and a crude rural-urban distinction is inappropriate. Traditional social structures are said to be underlain by values of solidarity and communitarian responsibility and to have advantages over imported social, economic, and political systems based on individualism and competition. Said to have been undermined by urbanization, it is clear that, although rapid social change has occurred, ethnic allegiances and support systems remain significant for urban as well as rural people and are important influences on both urban politics and daily life. In addition, the growth of informal sector economic activity during the 1980s was accompanied by the development of collective organizations to cope with the scarcity of credit, establish alternatives to government monopoly trading networks, and provide services (Chazan and Rothchild, 1993; Stren et al., 1994). These are confined neither to the middle class nor to urban areas.

The increasing numbers of NGOs are, it is hypothesized, both an outcome of political and economic change and likely to contribute to political change and make it difficult to reverse. Fowler (1991) attempts to assess whether NGOs have, as expected, been a significant force in eastern and southern Africa's political liberalization. He concludes that, with the exception of South Africa, they have not, because fragmentation of the sector inhibits the formation of strong representative bodies to defend their interests and argue for them on development policy issues; there are still only a few strong mass membership organizations; there is little indication of a coalescence of community-level bodies into politically significant networks and broader independent associations; many are top-down, undemocratic, tist organizations or consultants by another name; external funding is a mixed blessing because of the additional controls it implies and the ambivalent reaction of governments to its being channelled to NGOs; and governments disregard or discriminate against professional organizations such as lawyers' and journalists' associations. Thus, although the NGO sector has expanded, there is little evidence that this has encouraged political pluralism beyond whatever changes the state itself finds expedient and increasing evidence that "existing elites are both consolidating themselves within the voluntary sector and limiting its political impact by external controls" (Fowler, 1991, p. 74). In addition, while recognizing the continued significance of traditional social structures and their potential to provide alternative models of social and political organization, and the importance of social relationships based on trust and reciprocity in groups such as home town associations or rotating credit associations, care must be taken not to romanticize either ascriptive or voluntary groups in civil society. They are no more immune to patriarchy, clientelism, and venality than state bureaucratic or political interests. Nevertheless, the renewed attention that is being paid to urban social and political forces because of their significance in national political change is important because, it is suggested,

the emergence of powerful forces at the local urban level within civil society, combined with both national and international support for the institutionalisation of decentralization and democratization, have led to new forms of local governance and local arrangements for solving urban problems. (Stren, 1995, p. 12)

To what extent do the arrangements for urban administration and the policies and practices of urban planning and management revealed in the city case-studies in this volume bear out Stren's assertion? Can the current arrangements for urban management cope with the challenge?

Coping with the challenge

The inherited legal, institutional, and financial arrangements for managing urban development, although different in many details in different African countries, have clearly been unable to cope with the challenge for decades. In part this has been due to the inappropriate laws, institutional and decision-making structures, and revenue-generating instruments on which they are based and the failure of post-independence governments to make significant and appropriate changes. In part it has been due to the stress imposed on any institutional structure by rapid and unpredictable growth and change. And in part it has been due to the contradictions inherent in a state-centred model of economic and political development.

The inability of many central states to raise sufficient resources, develop sufficient administrative capacity, and operate in a sufficiently flexible way to deliver their development promises at the local level was already becoming clear in the late 1960s. Decentralization has periodically formed part of the political rhetoric of central governments since then, and administrative reforms have been attempted in a number of countries. However, political rhetoric has not been accompanied by devolution of decision-making power and the decentralization of administrative responsibilities has not been matched by the development of administrative and financial capacity to fulfil them. Although the political importance of urban populations has been recognized in national policies, such as the cheap food policy pursued by many countries, the damaging results of such policies have outstripped any progress in generating resources within urban areas to finance and manage their own developmental needs. The result has been unaccountable and poorly coordinated urban administration; failures in service delivery, development promotion, and regulation; and the decreasing ability of the formal system of management to control more than a minor part of the urban sphere. The cities have shared in the general crisis of public administration (Mutahaba et al., 1993), as is clear from the accounts in this book. Practical failures of service delivery and daily administration have given rise to fiscal crises, crises of political legitimacy, and opportunities for mismanagement and corruption. More recently, cities have suffered considerably from the effects of economic policy reforms, illustrating that there are limits on the political influence of urban groups (Healey and Robinson, 1992).

The inability of central government and city administrations to cope with the challenge of growth has had adverse implications for the development of national and urban economies, for the stability of political systems, and for the lives of urban residents, as is clear from the accounts presented here. The solution is said to be improved urban management. In assessing the likelihood that the management systems of African cities can rise to the challenge in the next decade, both the political and institutional arrangements for governance and the functions and tasks to be performed by administrative agencies need to be considered. These will be discussed in turn.

Urban governance

The "good governance" debate in which the international agencies have engaged African governments is related to the economic and political liberalization agendas discussed above. By the end of the 1970s the state was diagnosed as being overburdened, and this was said to account for its failure to cope with demands, its reduced efficiency, and the bloated size of the public sector. Structural Adjustment Programmes were, therefore, associated with demands that the size and role of the public sector should be reduced and its efficiency increased via budgetary reform and the adoption of private sector management methods. Expenditure cut-backs, reductions in employment, and some privatization occurred, but there was little evidence of enhanced capacity in the administrative system because of a failure adequately to consider the new institutional capacities demanded by economic and political reform, defects in the process of policy management, and organizational dysfunctions, including a proliferation of agencies with poorly defined roles and inadequate coordination mechanisms, the stultification of local governance, and the neglect of NGOs (Mutahaba et al., 1993).

The extent to which the cities described in this volume have arrangements for local government - defined as "a broadly accountable system of governance with respect to a wide range of functions" (Stern, 1991, p. 15) - varies. In the francophone countries of Africa, urban government was established on the French model. The largest cities had elected councils and elected mayors, and were responsible for a significant range of local services. However, immediately on independence, the national governments of countries such as Senegal and Cd'Ivoire took over control of the largest cities, explicitly because of their financial insolvency and administrative incompetence but implicitly in order to establish control over potential political opposition. In anglophone Africa likewise, elected local councils had been put in place by the end of the colonial period, but their performance fell far short of their responsibilities and growing demands, and their political autonomy and fiscal base were progressively eroded during the 1960s and 1970s, with few exceptions (Stern, 1991). Although the failure of city councils to deal with urban problems was bemoaned, central governments failed to give them adequate funds or revenue-raising powers or to ensure that they had sufficient decision-making powers and trained staff to improve the situation.

By the 1980s, however, a new recognition of the importance of effective municipal government seemed to be emerging. In Abidjan in 1980, for example, the City of Abidjan was established for the 10 communes, and a similar arrangement was made in Dakar in 1983. Local government was re-established in Dar es Salaam, Tanzania, and Kampala, Uganda. Increasingly, governments and international agencies seem to be recognizing that improved urban management, decentralization, and local democracy are interlinked.

Urban governance is concerned with the political arrangements to ensure that urban residents have a say in resource allocation, that decision-making is transparent, and that public agencies are accountable. It is, therefore, concerned with the roles and representation of a variety of actors and with informal as well as formal politics. It is also concerned with the relations between central and local government, which are typically contradictory and problematic. The variety of possible institutional arrangements for the management of urban growth and development is clear both from the city case-studies in this volume and from other research (Davey, 1993a,b). No single institutional arrangement is appropriate everywhere, and it is to be expected that political and institutional changes currently under way in many African countries will produce different institutional configurations and results.

Earlier attempts to decentralize had been ignored by most of the aid agencies and counteracted by the centralizing tendencies of one-party and military regimes. The practical results of more recent attempts to decentralize urban government, where these have occurred, have been limited. Kasfir (1993) considers that the new structures for participation introduced in the 1980s in Ghana, Burkina Faso, and Uganda "were given neither sufficient financial support nor political autonomy by national leaders" (Kasfir, 1993, p. 43). He expects new decentralized arrangements to continue to fail because of the contradictions inherent in central-local relations and deepening poverty. However, Stren (1995) suggests that various features of the current situation are more promising than hitherto, including the fact that reforms have been backed by the international agencies, are occurring in the context of a process of democratization' and are beginning to recognize the role of the associations of civil society (see also chaps. 14 and 15).

This is not to imply that forging new divisions of responsibility between central and local government, NGOs, community-based organizations (CBOs), and the private sector is unproblematic. In Nairobi, for example, the restitution of elected local government has not resolved the political rivalry between central government and Nairobi City Council that led to the suspension of the latter, and mechanisms have not evolved to coordinate the activities of the increasing numbers of NGOs. In Cairo, the lack of coordination between central government agencies and the three governorates into which the city is divided continues to hinder strategic planning and service delivery, while the role of Islamic organizations in providing welfare in a situation of increasing inequality adds to political/religious tensions. In Johannesburg, radical administrative and political reform is proceeding apace, but the large, diversified, and vibrant non-governmental sector is experiencing difficulties in making the adjustment from protest politics to partnership in urban administration.

Whereas some authors view the emergence or revitalization of local voluntary organizations capable of guaranteeing security, adequate incomes, and services in the face of government inability to do so as positive for both society and the state, averting even more of a legitimacy crisis than that to which it was already subject (Tripp, 1992), others fear that such associations, especially if based on ethnic or religious affiliations, will exacerbate conflict and polarization. Certainly, if the resources of all the actors on the urban scene are to be drawn upon in rising to the challenge posed by continued urban growth, explicit and directed attempts to define appropriate roles and forge appropriate relationships between the various actors in urban societies will be necessary (Halfani, 1995).

These relationships have both political components, defined in law and the constitution, expressed through electoral systems, and safeguarded through the courts and a lively civil society, and institutional components. City government, it is suggested, can provide effective leadership if it covers the whole built-up and developing area, has wide responsibilities for services related to the living environment, has a buoyant revenue base and well-qualified professional staff, and is politically accountable to the local population (Davey, 1993a,b). However, many systems of urban government are geographically fragmented (for example, Cairo and Johannesburg), functionally fragmented (Cairo, Lagos), or lack other of these attributes (Nairobi, Lagos, Kinshasa). Internal management, Davey concludes (1993a,b), seems to work best either where elected decision makers operating via a council and/or committees are backed by a professional administrator, whose department has corporate policy-making responsibility and capacity, or where political executive leadership encourages integrated management where there is functional fragmentation. In either case, forms of community representation and user participation are needed for local development, service provision, and management. There is, however, no consensus on the most appropriate administrative structure for large cities. Intra-city organization for area management and coordination of services, participation by community representatives in the direction of local services, representation of community interests to public agencies, or contribution to service provision may occur through administrative decentralization, separate municipal governments, publicly sponsored community representation structures, or independent community organizations. Delegated management of individual services to cost centres is increasingly being encouraged in order to improve efficiency. Davey advocates two-tier government for only the largest cities because of the resource requirements and risks of political conflict between the tiers, but does not recognize the problems that tend to arise if the city administration is too distant and centralized. He also comments apparently approvingly on the increasing prevalence and use of community-based structures (as do El-Shakhs, chap. 14, and Wekwete, chap. 15) without fully appreciating the potential difficulties created by the presence of a fragmented and competing set of community organizations that do not cover the whole metropolitan area.

Jaglin's evaluation (1994) of the government of Burkina Faso's attempt to mobilize residents of the peripheral areas of Ouagadougou to build or improve services is more perceptive of the drawbacks of reliance on CBOs as well as of the achievements. Between 1983 and 1990 in each neighbourhood, elected "revolutionary committees" organized land regularization and helped to mobilize and support improvements in local public facilities. Community labour and local financial levies helped to compensate for the lack of public sector resources, but attention was focused on the construction of water and social facilities, while operation and maintenance of these facilities and the organization and management of larger-scale initiatives and services, such as drainage and sanitation, were neglected. Other limitations of the system included the failure of the revolutionary committees to deal with large-scale infrastructure and long-term needs because it was possible to mobilize active community support only for short-term initiatives; and their inability to counteract inequalities within and between residential areas without public sector support.

Evolving political structures and processes that are accountable and stable at national level in African societies is proving difficult, so it is unsurprising that urban political systems are also unsatisfactory and volatile. The implications of current political trends for both urban governance and the tasks of urban management will need careful examination as they evolve.

Urban management

Traditional approaches to the management of urban development were fragmented, dividing responsibility for the operation and management of individual services (water, sanitation, transportation, health, etc.) sectorally and associating "planning" with physical planning. The results are well known: poor coordination between service providers, disjunctures between installation and operation and maintenance, and static blueprint land-use plans that were not implemented because of their reliance on (ineffective) controls over private sector development. Although exacerbated by the weakness of local government, tricky central-local relations, inadequate revenue bases, inherited land-use patterns, and widening inequalities between urban groups, approaches to the urban development process were conceptually and practically deficient (Stren and White, 1989; Devas and Rakodi, 1993a). Attempts to devise more adequate approaches have redefined "planning" and adopted it in tandem with "management" to embrace the full range of governmental interventions in the development and day-to-day operations of the city. Whereas urban planning is concerned primarily with anticipating and preparing for the future, and particularly with the spatial and land-use dimensions of urban development, urban management is concerned more with the operation of a range of public services and with a variety of interventions that affect urban conditions as a whole (Devas and Rakodi, 1993b, p. 44). Thus arguments are advanced for an integrated approach, in which responsibilities are clearly defined and allocation and coordination mechanisms established, as well as a dynamic approach, in which flexibility to respond to changing circumstances is built into planning, policy-making, and implementation by the institutionalization of monitoring and evaluation (see, for example, the approach advocated by McGill, 1994, based on experience in Malawi).

The approaches to urban management established in colonial days and promoted subsequently by donors, education and training, and research outputs have been subject to widespread criticism (Kironde, 1992; Okpala, 1987). In addition, problems with the design, implementation, and continued operation of externally financed urban projects in the 1970s (see chap. 2) led to some reconsideration of urban lending (World Bank, 1991). Although it is unanimously agreed that changes were and are needed, the adoption and promotion of "urban management" by the international agencies have been treated with considerable scepticism. Stren (1993) criticizes the World Bank and its partners in the Management Programme for failing to define the term. He explains this failure in terms of the need for an all-purpose concept that would, first, enable the agencies to move away from shelter projects to more institutional approaches; secondly, signal a shift from a public sector statist perspective of development to a neo-liberal perspective concerned with efficiency and the coordination of public and private sector activities; and, thirdly, maintain organizational flexibility to vary programme portents. Jones and Ward (1994) take the critique further. Although applauding the shift from large-scale urban projects, in which government expects to be the provider, towards an enabling role for a more efficient and transparent public sector, they doubt whether the policy prescriptions advocated by the World Bank are likely to be effective. The approach is also criticized for being technocratic and still highly sectoral (Stern, 1991).

Effective urban management requires competence in running individual services, integrated approaches to redevelopment and development of new areas, and responses to overall challenges such as economic decline or environmental deterioration. Consideration must thus be given to the specific functions of an urban management system and also to mechanisms for decision-making and coordination. Administrative arrangements are needed to generate resources, define responsibilities, and provide staff, as well as to ensure the tasks of policy formulation, resource allocation, implementation, operation, maintenance, regulation, and promotion are carried out in an efficient and coordinated manner. Actions are needed with respect to urban economic development, physical and social infrastructure and services, land, shelter, and environmental problems.

The financial basis for urban administration

Typically the most buoyant taxes are retained by central government, and local authorities are left with less buoyant and politically more tricky revenue sources, such as property tax. This limited tax base, exacerbated by central government restrictions on rates, limited and erratic central-local transfers, poor performance with respect to collection and enforcement, out-of-date valuations, and the increasingly widespread occurrence of unenumerated property and economic activities, has meant that, invariably, the revenue available to local government has been insufficient for it to perform its functions (Stern, 1989a). As a result, it lacks legitimacy in the eyes of urban people and businesses, further curtailing its ability to generate revenue. Escape from this vicious circle requires political and legislative change, an injection of funds, and a more explicit political and practical strategy to deliver improved services, in order to increase legitimacy and thus increase the potential for further revenue generation and implementation of regulatory functions.

Both capital and revenue funds are needed for urban development. In the absence of well-developed capital markets, government action is essential to provide capital funds, although the transfer of responsibilities to local government without the powers and resources to fulfil these responsibilities, as Wekwete points out, is more typical. Backed by an injection of external funds and technical assistance to weaker local authorities, revolving municipal development funds administered by central government are a promising initiative in a number of countries including Nigeria and Zimbabwe, although Davey points out some of the pitfalls (Davey, 1993a,b). The promise of access to further borrowing in future provides a lever to improve local government financial management, although it can also be used to secure political compliance from elected local authorities. Revenue funding may be partly provided by central government transfers, especially if central government monopolizes most tax revenue, but the unpredictability of such transfers can hinder the planning and operation of local councils and so increased local resource mobilization is essential. Although there is insufficient space here to go into detail, research and guidance are emerging from the urban management programme (Bahl and Linn, 1992; Davey, 1993a,b). Both taxes on property and taxes on local economic activities are advocated.

Urban economic development

Economic activity located in urban areas is the driving force of urban growth and the jobs created the main attraction for rural migrants. Buoyancy in urban economies was taken for granted, especially in the years after independence, when expansion of the public sector and the pursuit of import substitution industrialization policies gave a boost to urban employment opportunities. Indeed, the economic preeminence of the largest cities was typically seen as a problem and measures (however ineffective) taken to divert investment elsewhere, either to rural development or to secondary cities and urban centres in rural areas.

Not until the drastic decline in urban infrastructural standards that followed recession and structural adjustment in the 1980s was the economic importance of adequate infrastructure widely recognized. This recognition was assisted by the continued failure of African cities to compete for international investment in manufacturing and services (especially compared with Pacific Rim Asian cities); by research that demonstrated the costs to manufacturers of poor infrastructure and the important contribution made by economic activities located in urban areas to national economic growth; and by pressure from a growing business lobby that, especially where entrepreneurs had emerged from the informal sector, was less dependent on the state for its access to the means of production and accumulation than before.

The health of city economies has not, except more recently in South Africa, been within either the legal remit or the area of concern of local government, or indeed of central agencies responsible for local services, which have been preoccupied with routine administration, service provision, and political survival (Harris, 1992). Analyses of city that chart trends in demand for the package of goods and services produced in the city, identify activities with current or future comparative advantage, pinpoint constraints on the productivity of economic enterprises and their labour force, and analyse the spatial distribution of economic activities in the city and the links between them, as a basis for formulating economic development policies, are rare. Rogerson in chapter 10 advocates such studies and a more proactive role for public agencies in local economic development. The "urban managers" in this context, Harris (1992) suggests, comprise the local authority, the chamber of commerce, and relevant industrial, commercial, financial, and professional associations (and, I would add, labour unions).

Some initiatives were mentioned in the case-studies, for example Nairobi's City Convention, the alliance between local government, big business, and inner-city residents to revitalize Johannesburg's CBD, and collaboration between local authorities, employers, and infrastructure companies to improve security and services in the Ikeja and Apapa areas of Lagos. The number and scope of such alliances in contemporary African cities compared with nineteenth-century and contemporary cities in, for example, the United States, Japan, or Europe are, however, limited. This reflects the absence of large-scale indigenous enterprises; governments' assumption that the management of the economy, installation and operation of infrastructure and services, and provision of housing were their responsibility; the lack of encouragement given to alternative associations during the authoritarian and statist post-independence decades; and the inheritance of good-quality infrastructure in the formally planned parts of cities, which, at least initially, was in working order. The forging of such alliances between local politicians and entrepreneurs (large and small scale), backed by officials more prepared to be proactive than traditional local government bureaucrats, can be expected to become more common in future, although the formation of politically strong local alliances in the largest cities is likely to provoke ambivalent reactions by central governments, even if they are committed to democratization and decentralization.

Infrastructure and services

One of the most visible and disturbing characteristics of the poorer cities... is the decline of their infrastructural base. As urban populations grow, and as available resources decline, public infrastructure is being degraded to a point where cities are seriously losing their capacity to operate as productive entities... In many African cities... refuse is uncollected and piles of decaying waste are allowed to rot in the streets; schools are... overcrowded; some urban roads deteriorate into quagmires in the rainy season, and are pitted with dangerous potholes during the dry season;... private telephones... are an impossible dream; public transport systems are becoming seriously overloaded; and more and more people are obliged to live in unserviced plots... Not only is little new infrastructure constructed, but existing infrastructure is poorly maintained. (Stern, 1991, p. 7)

Stren's analysis refers to African cities in general. The situation he describes is characteristic of large parts of some of the cities described in this volume, especially Nairobi, Lagos, and Kinshasa, and is becoming increasingly widespread in lower-income areas of the others (see also Ngom, 1989, on Dakar; El Sammani et al., 1989, on Khartoum). The situation is even worse in many smaller cities and poorer countries (Onibokun, 1989).

Stren and White's research in the 1980s showed that most water and electricity supply agencies in African cities were either central government parastatals or attached to central government ministries. In most cases, supply has failed to keep pace with urban growth. Where local government continues to be responsible, experience varies, from Zimbabwe on the one hand, where the supply of water has generally kept pace with urban population growth, to Kenyan local authorities on the other. However, problems should be attributed not only to local government failings, but also to central government policies and inaction. In some francophone countries, the water supply function is contracted out to private companies. The Cd'Ivoire's experience is that such an arrangement can be efficient if the company concerned is given adequate autonomy with respect to operation and pricing, although there can be difficulties in ensuring supply to low-income areas (Stern, 1989b). An intermediate arrangement, of a national (Ghana) or municipal (some Zambian local authorities) public sector company, is becoming more wide spread, as countries attempt to improve efficiency without outright privatization.

Waste management arrangements in most cities are also far from satisfactory. Whereas anglophone countries rely on a public service, elsewhere solid waste collection may be subcontracted, to a single company as in Abidjan, or to local entrepreneurs (Stern, 1989b). Whatever the arrangement, effectiveness is inhibited by inappropriate collection arrangements and shortages of foreign exchange to import equipment, and equity is reduced by the lack of finance to subsidize services in low-income areas where necessary.

The failure of land subdivision and servicing programmes to keep pace with urban growth, which has led to widespread illegal and informal development, and not only of low-income areas (see below), has hindered the extension not only of water, electricity, and solid waste collection services but also of adequate sanitation arrangements and road networks to large areas of Africa's cities. In the absence of collective sanitation arrangements, households are forced to devise individual solutions suited to their incomes and physical circumstances. Reliance on pit latrine sanitation can give rise to groundwater pollution and problems occur as density rises, while the use of septic tanks needs to be backed up by private or public sludge removal services and suitably located disposal sites. Where waterborne sewage disposal to conventional treatment works is available, it is costly, limited in coverage, and poorly operated, because revenue and foreign exchange shortages have made it increasingly difficult to obtain spare parts and maintain systems. As densities increase, unacceptable levels of sharing occur, or residents cannot get access to any toilet facility and are forced to "use the bush" ("free range" in Accra). Both have adverse health and environmental implications.

In the early years after independence, public transport was typically a public monopoly, whether a parastatal bus company or a local government function. Invariably it was inadequate, owing to shortages of foreign exchange for parts and new buses and poor management. Increasingly, the public monopoly has been supplemented or replaced (legally or illegally) by private operators, a trend that increased its momentum with economic liberalization. Although the availability of public transport has generally improved as a result, the failure of the public sector to provide an adequate road network results in patchy coverage and high operating costs, while its inability to regulate the private sector has resulted in poor safety standards and sometimes violent competition to ply particular routes.

A review of the privatization of urban services in Asia suggests that the advantages, in theory, include competition, leading to lower production and delivery costs, increased efficiency, better access to current technology, greater capacity to obtain and maintain capital equipment, more flexibility and speed in decision-making, and reduced pressure on government. However, opposition is likely from political leaders who control patronage and from public sector unions, as well as from some consumer groups and low-income residents' organizations, who fear higher costs. In addition, private and NGO management skills may be poor and, where financial markets are undeveloped, private contractors may not be able to get access to capital. Finally, services that are collective goods, those that require lumpy investment, and those for which it is difficult to charge users directly are unsuitable for privatization. Privatization may occur through a variety of mechanisms, including sale of public assets, deregulation and liberalization, contracting out, public-private or public-NGO partnerships, transfer to NGOs, and government support for private providers (Rondinelli and Kasarda, 1993). Lessons from the Asian experience for African cities demonstrate the need to take care in designing the mechanisms for private/NGO involvement, to devise a workable regulatory framework, and to select only those services for which privatization is appropriate (for example, transport, housing, refuse collection, and perhaps aspects of health care) (see also Davey, 1993a,b). Institutional reforms related to service provision are under way in African cities. Wekwete, in chapter 15, describes a shift from a public sector-based approach to urban management to a public private sector model. However, he stresses the uncertain future of the new approach, owing to its reliance on external funding, its dependence on cooperative relations between central and local government, and the lack of formal structures at city level through which NGOs and CBOs can be systematically involved in decision-making. In addition, he notes the lack of evaluation of recent attempts at privatization and related changes, to update the review carried out in the 1980s (Stren and White, 1989).

Housing, land, and planning

Public sector housing policies focused initially on the provision of complete house units, typically for public sector employees as well as low-income groups. To some extent this approach has been superseded by sites and services, but neither has succeeded in matching supply with demand, let alone need, for a variety of reasons, including inadequate land administration systems, lags in infrastructure provision programmes, lack of capital, shortages of building materials, and limited capacity in the construction sector. As a result, most additions to the housing stock have been made by the private sector, with quantitatively by far the largest contribution coming from individual households and small landlords. Almost invariably, at least one aspect of this housing is illegal - lack of formal land tenure failure to obtain development permission, failure to satisfy building regulations, or disregard of rent control. It is financed, built, and exchanged outside the formal systems for mortgage lending, construction, and sale.

Where pressures on land are greatest and infrastructure most deficient, the quality of accommodation can be appalling, but much of the housing produced is adequate to satisfy basic needs for shelter. Thus infrastructure improvements and provision of education and health services are much more important than improvements to the dwelling in improving the quality of the residential environment. Whereas regularization and upgrading programmes were accepted reluctantly by politicians and officials in the 1970s, there are signs of a new realism as the financial and administrative weakness of public agencies has become increasingly clear, and the potential of more participatory community-based approaches to infrastructure installation and operation is demonstrated in residential areas, often with donor and NGO assistance. The need to regularize informal tenure arrangements sufficiently for them to be used to generate revenue may assist in this process (Durand-Lasserve, 1993). However, it should be recognized that regularization of unplanned areas is no substitute for a rapid land subdivision and strategic infrastructure investment programme linked to a land-use planning process.

This is not to advocate a return to the land-use master plans of earlier decades, with their inappropriate levels of detail, inflexibility, dependence on foreign consultants for preparation, and lack of connection to resource allocation decisions. The assumptions on methods of implementation, administrative capacity, and planning standards built into that approach to planning are demonstrably inappropriate (Stern, 1989b; Devas and Rakodi, 1993b; see also Findlay and Paddison, 1986, on Tunis and Rabat; Dubresson, chap. 8 in this volume, and Attahi, 1991, on Abidjan). Strategic planning in which land-use and other policy proposals are linked to capital investment programmes is needed, together with more detailed guidance for development control in some areas. such as the CBD, and more action-oriented approaches to development or improvement in others. Increasingly, adaptive responses are becoming more common than coercive removal of offending structures or people, but there is still political and bureaucratic resistance to "lowering" of standards, while the flexibility of strategic planning frameworks compared with zoning maps also increases the opportunities for corruption in decision-making on development applications.

Environmental problems

Related to deficiencies in infrastructure provision and waste management, environmental problems also arise from the inability of public sector authorities to enforce regulations governing land development, industrial emissions, etc. The urban poor are most exposed to environmental hazards because they are most likely to depend on untreated water, to live in risky areas, to be unprovided with sanitation and solid waste collection services, to live in overcrowded conditions and work in unregulated enterprises, and to be exposed to high levels of indoor pollution from cooking fuels. Their vulnerability to ill health arising from environmental problems is increased by poor nutrition and inadequate access to health care. Action on environmental problems is more likely to occur, however, when the better-off are affected, as for example by air pollution, contaminated food, or bad effects on economic activities such as tourism. It may also follow disaster. Thus many unsustainable practices and environmental problems are not tackled until disaster occurs or until organized groups increase the pressure for change. Action is needed both to tackle negative externalities such as pollution directly and to monitor the upstream resource-using and downstream disposal or recycling of wastes in the city and its region.

Guiding and controlling urban growth

Should African countries try to restrain the growth of their largest cities? The largest cities probably are more expensive per head of population to service and may have absorbed a disproportionate share of public investment. However, they generally also contribute a major share of GDP. In addition, there is no evidence that, if given scope to raise revenue locally and use appropriate standards, they need be a drain on government resources or that disamenities are necessarily greater in very large cities (Richardson, 1993). Appropriate levels of investment and development programmes in rural areas are needed. There is also scope for increasing the attractiveness of secondary cities by supporting the development of economic activities linked to their hinterlands, investment in infrastructure, and development of greater institutional capacity. However, even if such programmes are successful, they are unlikely to have more than marginal effects on the rate of growth and size of the largest cities. More important for the latter, therefore, are improved management, some of the components of which have been discussed above, and spatial restructuring. The latter is one of the main concerns of El-Shakhs, who in chapter 14 advocates the acceptance and promotion of urbanized corridors, such as the one that is developing between Cairo and Alexandria, and decentralization of urban growth within city regions in preference to continuous expansion of the built-up area.

This volume has been concerned with both the dynamics of urban growth and development and its management. The summary of key trends and issues has, more than once, referred to incomplete knowledge and unanswered questions. Neither improved understanding nor more adequate responses to the challenge is possible without better data on and analysis of urban characteristics. The final section of this chapter will, therefore, outline some of the outstanding research needs that have been revealed.

Research priorities

As part of a large project funded by the Ford Foundation to review recent research on urban areas in the developing world, five regional reviews were carried out in Africa (Stern, 1994). While mainly concerned to identify and summarize research undertaken during the past 30 years, the project also aimed to outline a research agenda for the 1990s. The research priorities identified below arise from the experience of producing this volume for the United Nations University, but parallels with the views of contributors to that project will be noted.

The biggest problem in identifying the relationship between global forces and urbanization in Africa, trends in urbanization, the characteristics of cities, and the outcomes of policy and management interventions is the lack of both basic data and original research. This volume has been written in the absence of recent reliable census data for the majority of countries and cities considered. Ascertaining the dimensions of and trends in urbanization in the absence of such data is virtually impossible, hence the accounts are full of estimates, general statements, and alternative figures, where greater precision is desirable. Although urban trends and characteristics are better documented in some countries and cities than others, owing not only to the absence of census and other published data but also to the incomplete topic and disciplinary coverage of urban issues, nowhere is the level of knowledge satisfactory. The papers in this volume, especially but not only the thematic papers, were designed as reviews of existing knowledge and data. They are, indeed, valuable reviews and many draw on sources that are inaccessible to an international readership within and outside Africa. However, what can be learnt from existing work, which declined noticeably in the 1980s, especially in eastern Africa, has, in my view, largely been exhausted. The research priorities identified below cannot be addressed without substantive new research, drawing both on national data and on primary data collection designed to elucidate the specifically urban issues outlined. At both the national and city levels, better use needs to be made of available data, especially census data. However, formally collected data are not always suitable for addressing the critical issues identified below, and they need to be supplemented by both nation- and city-wide studies and smaller-scale quantitative and qualitative data-collection exercises. As donor and country interest in urban development increases, the need for improved knowledge of urban dynamics and the outcomes of policies and management practices poses a challenge for research funders and researchers alike.

The urban economy

A better understanding is needed of the dynamics of urban economies because, without this, it is impossible to judge the efficacy of any policies aimed at urban economic development and employment creation. Sector studies that embrace enterprises of all sizes are needed. They should explore the potential of and constraints on enterprises of different sizes, as well as links between them and with other sectors in terms of subcontracting, training, transfer, and labour movement. Obvious candidates are leather, food processing, textiles and clothing, wood processing and manufacturing, construction and building materials production, business services, and wholesale and retail trade. However, there is also some evidence to show that the potential of and constraints on enterprises of different sizes have some similarities. Thus there is also a need for intersectoral studies of large, intermediate, small, and micro-enterprises, in order to ascertain what the outcome of structural adjustment policies and past support programmes has been and what type of support programmes might be appropriate in future. Not only should business enterprises be studied as discrete units, but the associations between them, in addition to processes and procedures for operating contracts, accessing credit, and regulating malpractice, should also be examined. It is important to consider both formal and informal associations and procedures. First, associations of small-scale enterprises may come to be as significant as the traditional associations of large enterprises, such as employers' federations or chambers of commerce, in the less state-centred model of policy development likely to emerge in future years. Secondly, the procedures and operating practices evolved by small-scale businesses may be more appropriate and workable for African countries today than those that typify the large-scale sector, for example with respect to accessing credit (Tripp, 1992; Halfani, 1994, 1995; Attahi, 1994; Kharoufi, 1994; Onibokun, 1994).

As competition between the largest cities to benefit from trends in the global economy increases and becomes more explicit, studies of the comparative advantage of cities, the way in which stakeholders within them articulate their economic interests, and the scope for cooperation, if not at a continental then at a regional level, are needed. In addition, evaluations of the prospects for and components and outcomes of local economic development strategies are needed (Swilling, 1994).

Urban society, poverty, and equity

To underpin both economic development and urban politics and governance (see below), a deeper understanding is needed of the social changes taking place in cities, especially given the strains generated by the rapid socio-economic changes currently taking place and the importance of household structures and social networks in enabling urban residents to cope with economic deterioration and declining investment in social services. The themes to be investigated include the family and social networks, culture, community, and crime and insecurity (Halfani, 1994; Swilling, 1994). The social and cultural consequences of urbanization, in particular social disorder and urban violence, which appear to be on the increase, are also emphasized by Onibokun (1994), as is the need for study of social movements by Kharoufi (1994).

Attempts to improve understanding of urban poverty are under way (Rakodi, 1995a; Vanderschueren et al., 1996) and need to be taken further by assessments of whether the holistic conceptualization of urban poverty currently being advocated is an advance on earlier concentration on poverty lines, both analytically and in terms of policy formulation (Attahi, 1994; Onibokun, 1994). The implications of recession in general, structural adjustment policies in particular, and political changes for the extent and nature of urban inequality and poverty need more systematic consideration, drawing on the national poverty assessments that are increasingly being funded by the World Bank, UNICEF and others. The response to increasingly widespread evidence of increased hardship has typically been the design, limited funding, and partial implementation of programmes to ameliorate the adverse effects of Structural Adjustment Programmes rather than reconsideration and redesign of structural adjustment policies. Although the scope for such programmes to do more than tinker at the margins of poverty appears to be limited, their effects need to be evaluated.

Land and property markets

The importance of the land market processes by which agricultural land on the urban periphery is converted for urban use was noted in chapter 11. The extent to which this is a legal process varies from city to city, but, except in one or two countries, such as Zimbabwe and, to a lesser extent, South Africa, most of the land development processes under way operate wholly or partly outside the formal legal process. The scope for universalizing formal freehold or leasehold tenure systems and effectively controlling subdivision and development seems limited in most African countries. If ways of guiding development to avoid the worst impacts of unplanned development and unregistered land rights are to be found, it will be necessary to devise systems that can work alongside prevailing land markets and development processes rather than futilely attempt to control them. A start has been made in devising such systems, for example in countries such as Uganda and by some agencies, such as the World Bank/UNCHS/UNDP Urban Management Programme for Africa (Durand-Lasserve, 1993), but, to make progress, a greater depth of understanding of existing informal and semi-formal processes is needed (see also Halfani, 1994; Onibokun, 1994). Such research should investigate the factors that determine processes and patterns of urbanization, including the tenure preferences of urban residents, property prices in relation to incomes, tenure systems in the rural areas surrounding cities, and the institutional framework for land management in the city and its surrounding region. Institutions in this context include official land management agencies and planning authorities, arrangements for coordination between those administering the city and those responsible for the surrounding rural areas, and social relations governing the rights of indigenous individuals and groups to use and dispose of land.

The extent and nature of the impact of cities on their hinterlands are relevant to this topic as to others (Halfani, 1994; Onibokun, 1994) - changes in economic activities in response to urban opportunities may either intensify rural land uses, for example intensive vegetable production for urban markets, or disrupt them as labour is withdrawn from agricultural production and holdings are subdivided. In addition to land tenure, urban employment opportunities and transport links are important influences on the extent to which urban sprawl occurs and the ways in which economic and spatial patterns and relationships in the city region are transformed.

Although the conversion of agricultural land for urban use will continue to be very important, as cities mature, markets in undeveloped land and property within the broadly built-up area become increasingly important (see Kharoufi, 1994, on the need to study the dynamics of resident mobility and property markets in old city centres in north Africa). Identifying and explaining trends in these property markets is essential to an improved understanding of urban dynamics and to devising appropriate policies with respect to, for example, bringing forward undeveloped land for development, redevelopment, or densification. A range of interventions in urban land markets are currently being implemented or considered in one or more cities in Africa. At present, our understanding of their likely impacts is limited and evaluations of attempts to implement them are needed. Relevant policies include:

- the use of capital investment in major infrastructure to guide new development;

- the imposition of high rates of property tax to encourage the sale or development of undeveloped land;

- the use of financial incentives and regulatory changes to encourage densification;

- decontrol of rents;

- community land trusts as an alternative to private individual tenure. especially for regularization of informal settlement areas.

Urban politics and governance

The failure of public administrative systems to cope with the challenge of urban growth in Africa has been relatively well documented. The exceptional cases where there is considerable administrative capacity at the local urban level have also been analysed (see, for example, Rakodi, 1995b, on Harare, and Pasteur, 1992, on Bulawayo in Zimbabwe; Smith, 1992; Swilling et al., 1991; and Tomlinson, 1994, on South Africa). Elsewhere, the extent of change to urban politics and governmental arrangements varies from very little to quite extensive. The differences are a result partly of the "variation in the degree of institutionalisation and richness of civil society in different countries, reflecting both historical and cultural factors and the coercive power of the state" (Healey and Robinson, 1992). The latter in turn governs the strategies open to people to challenge the prevailing political order, which vary from open criticism and opposition to the expression of discontent through disengagement from the state or non-compliance with state policies. Until the 1980s, most institutional effort and research had been directed to the built environment and formal institutional arrangements for management and politics. The state was seen as the main agent directing urban change and the sphere outside the formalized sector as an unintended by-product or pathology of a formal process of urban development (Halfani, 1995). Anecdotal evidence reveals recent attempts to democratize local government (both non-party political, as in Ghana's district assemblies and Uganda's Resistance Councils, and party political, as in Lusaka or Nairobi); attempts to strengthen urban local government and its planning, revenue-raising, and management capacities; and attempts to develop partnerships with NGOs and CBOs (e.g. in Freetown or Lusaka) and with private sector service providers. Systematic documentation and evaluation of the structures, relationships, and processes involved and their outcomes is, however, limited (Halfani, 1994), and indeed the conceptual framework for doing this is undeveloped.

Hyden (1992) suggests a number of conditions that facilitate good governance and therefore effective problem-solving, which could form the basis for a set of evaluative criteria:

(a) citizen influence and oversight, determined by the scope for political participation, methods of preference aggregation, and the means of ensuring public accountability;

(b) responsive and responsible leadership, including the degree of respect evident for the public realm, adherence to the rule of law, and the openness of policy-making - Manor (1995) distinguishes in this respect between bureaucrats, who should be accountable to elected representatives, and elected representatives, who should be accountable to citizens;

(c) social reciprocity, including the degree of political equality, the extent of intergroup tolerance, and the degree of inclusiveness in associational membership.

In the 1960s and 1970s, as discussed in chapter 2, international lending was focused on urban projects and little attention was paid to developing institutional capacity. This has changed more recently, but evaluation of the role of the aid agencies in promoting and supporting political reform and institutional development at the urban level is lacking. The main conclusion of Healey and Robinson's (1992) review of the role of these agencies at national level is that they have been extremely vague with respect to their recommendations on appropriate reforms to institutional structures and processes. At this level there is no clear association between regime type and successful reform, or between political liberalization and improved economic performance. Early research into institutional reform, for example with respect to the water sector in Ghana, demonstrates that it has proceeded less rapidly than economic liberalization and less rapidly than hoped for by the lending agencies.

Healey and Robinson's (1992) research agenda related to democracy, governance, and economic policy at the national level has obvious urban counterparts. Comparative studies are needed of:

- the institutional and procedural context of policy-making in established democratic systems of local government, focusing on how interest groups are consulted;

- the outcomes of democratization for the operation of urban local governments and their relations with central government;

- the political implications of decentralization and its impact on policy-making, resource generation, and implementation;

- the nature of urban actors (including central and local political institutions and bureacracies, traditional authorities, NGOs, CBOs, informal associations of residents, and the large- and small-scale private sector) and of the evolving relationships between them, as illustrated by policy-making, implementation and service delivery, operation and maintenance, at community, district/neighbourhood, and city levels (see also Halfani, 1994; Swilling, 1994).

Thus evaluations are needed both of reforms of public sector arrangements for urban management and of mechanisms and forums designed to incorporate a wider range of actors into policy formulation and decision-making, such as the UNCHS-facilitated Sustainable Cities Programmes in Dar es Salaam and Accra, and the Local Forums and Commissions in South Africa. Without neglecting the politics of urban government, Halfani (1994), Attahi (1994), Onibokun (1994), and Swilling (1994) also stress the need for structural and functional analyses of the institutions responsible for urban management and the operating systems they employ. To address this more specific focus on evaluating the effectiveness of urban management systems and urban governmental institutions, Davey (1993a, pp. 1-2) suggests the use of six criteria:

(1) technical competence in the choice, design, and implementation of infrastructure investment, and its operation and maintenance;

(2) efficiency in the use of resources;

(3) financial viability based on local revenue generation and sound financial management;

(4) responsiveness to the needs arising from urban growth and the ability to plan the city and its services ahead of, or at least in pace with, demand;

(5) sensitivity to the needs of the urban poor and appropriate weighting of public interventions to promote their access to basic services, employment, and shelter;

(6) concern for environmental protection, through public service provision and regulation of the private sector.

To summarize, there is a need to assess effectiveness with respect to service delivery, regulation, overall economic and physical development of the city, poverty alleviation, and environmental protection (see also Halfani, 1994; Attahi, 1994). Healey and Robinson's (1992) call for studies of how particular policy changes were initiated, designed, and implemented is as relevant to the study of the processes involved in urban management as it is at the national level. Finally, cross cutting all the research themes identified above is the need to identify gender differentiation and relations with respect, first, to access to urban resources and decision-making, and, secondly, to the outcomes of policy and practice.

The project of which this volume is the outcome set out to describe and explain the evolution of some of the largest and most important cities in Africa; to investigate the linkages between these cities and global, continental, and national urban systems; to assess the policy implications of present and future patterns of large-city growth; and to assess the performance of urban management systems in coping with urban growth. The lack of adequate data and research studies on which to draw, and the enormous differences between the countries and cities of continental Africa, have hindered our achievement of these aims. However, the material drawn together in this volume will, we hope, make a significant contribution to contemporary analysis of urbanization in Africa.

References

Aboyade, O. 1994. Essentials for African economic transformation. IDS Bulletin 25(3): 73-78.

Adedeji, A. 1993. The case for remaking Africa. In: D. Rimmer, ea., Action in Africa: The Experience of People Involved in Government, Business and Aid. J. Currey, London, and Heinemann, Portsmouth, NH, pp. 43-57.

Attahi, K. 1991. Planning and management in large cities: A case study of Abidjan, Cd'Ivoire. In: UNCHS, Metropolitan Planning and Management in the Developing World: Abidjan and Quito. United Nations Centre for Human Settlements, Nairobi, pp. 31-82.

Attahi, K. 1994. Urban research in Francophone West Africa: Cd'Ivoire Senegal, Burkina Faso, Togo. In: R. Stren, ea., Urban Research in the Developing World, Volume 2: Africa. Centre for Urban and Community Studies, University of Toronto, Toronto, pp. 193-230.

Bahl, R. W. and Linn, J. F. 1992. Urban Public Finance in Developing Countries. Oxford University Press, New York.

Brown, M. B. 1995. Africa's Choices after Thirty Years of the World Bank. Penguin, Harmondsworth, Middx.

Chazan, N. and D. Rothchild. 1993. The political repercussions of economic malaise. In: T. M. Callaghy and J. Ravenhill, eds., Hemmed in: Responses to Africa's Economic Decline. Columbia University Press, New York, pp. 180-214.

Davey, K. 1993a. Managing Crowing Cities: Options for Urban Government. Development Administration Group, Institute of Local Government Studies, University of Birmingham, Birmingham.

Davey, K. 1993b. Elements of Urban Management. World Bank/UNCHS/UNDP, Urban Management Programme Discussion Paper 11, Washington D.C.

Devas, N. and C. Rakodi, eds. 1993a. Managing Fast Crowing Cities: New Approaches to Urban Planning and Management in the Developing World. Longman, London.

Devas, N. and C. Rakodi, eds. 1993b. Planning and managing fast growing cities. In: N. Devas and C. Rakodi, eds., Managing Fast Crowing Cities: New Approaches to Urban Planning and Management in the Developing World. Longman, London, pp. 41-62.

Durand-Lasserve, A. 1993. Conditions de mise en place des systs d'information fonci dans les villes d'Afrique sud-Saharienne francophone. World Bank, UMP 8, Washington D.C.

Eddin Ibrahim, S. 1987. A sociological profile. In: A. Y. Saqqaf, ea., The Middle East City: Ancient Traditions Confront a Modern World. Paragon House, New York, pp. 209-226.

El Sammani, M. O., M. El Hadi Abu Sin, M. Talha, B. M. El Hassan, and I. Haywood. 1989. Management problems of Greater Khartoum. In: R. E. Stren and R. R. White, eds., African Cities in Crisis: Managing Rapid Urban Growth. Westview, Boulder, Colo., pp. 246-275.

Findlay, A. M. 1994. The Arab World. Routledge, London.

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

Fowler, A. 1991. The role of NGOs in changing state-society relations: Perspectives from Eastern and Southern Africa. Development Policy Review 9(1): 53-84.

George, S. and F. Sabelli. 1994. Faith and Credit: The World Bank's Secular Empire. Penguin, Harmondsworth, Middx.

Green, R. H. 1993. The IMF and the World Bank in Africa: How much learning? In: T. M. Callaghy and J. Ravenhill, eds., Hemmed in: Responses to Africa's Economic Decline. Columbia University Press, New York, pp. 54-89.

Green, R. H. and M. Faber. 1994. The structural adjustment of structural adjustment: Sub-Saharan Africa 1980-1993. IDS Bulletin 25(3): 1-8.

Halfani, M. 1994. Urban research in eastern Africa: Tanzania, Kenya and Zambia towards an agenda for the 1990s. In: R. Stren, ea., Urban Research in the Developing World, Volume 2: Africa. Centre for Urban and Community Studies, University of Toronto, Toronto, pp. 113-192.

Halfani, M. 1995. The governance of urban development in East Africa: An examination of the institutional landscape and the poverty challenge. Paper to the Global Urban Research Initiative/African Research Network for Urban Governance Workshop, Nairobi, November, mimeo.

Harris, N. 1992. Productivity and poverty in the cities of the developing countries. In: N. Harris, ea., Cities in the 1990s: The Challenge for Developing Countries. UCL Press, London, pp. 173-195.

Healey, J. and M. Robinson, 1992 Democracy, Governance and Economic Policy: Sub-Saharan Africa in Comparative Perspective. Overseas Development Institute, London.

Hyden, G. 1992. Governance and the study of politics. In: G. Hyden and M. Bratton, eds., Governance and Politics in Africa. Lynne Rienner, Boulder, Colo., pp. 1-26.

Jaglin, S. 1994. Why mobilise town dwellers? Joint management in Ouagadougou (1983-1990). Environment and Urbanization 6(2): 111-132.

Jones, G. A. and P. M. Ward. 1994. The World Bank's "new" Urban Management programme: Paradigm shift or policy continuity? Habitat International 18(3): 33-51.

Kasfir, N. 1993. Designs and dilemmas of African decentralization. In: P. Mawhood, ea., Local Government in the Third World: Experience of Decentralization in Tropical Africa. Africa Institute, Pretoria, pp. 24-48.

Kharoufi, M. 1994. Reflections on a research field: Urban studies in Egypt, Morocco and the Sudan. In: R. Stren, ea., Urban Research in the Developing World, Volume 2: Africa. Centre for Urban and Community Studies, University of Toronto, Toronto, pp. 47-112.

Kironde, J. M. L. 1992. Received concepts and theories in African urbanisation and management strategies: The struggle continues. Urban Studies 29(8): 1277-1292.

McGill, R. 1994. Integrated urban management: An operational model for third world city managers. Cities 11(1): 35-47.

Mamdani, M. 1992. Conceptualising state and civil society relations: Towards a methodological critique of contemporary Africanism. In: C. Auroi, ea., The Role of the State in Development Processes. Frank Cass, London, pp. 15-23.

Manor, J. 1995. Democratic decentralization in Africa and Asia. IDS Bulletin 26(2): 81-96.

Mead, D. C. 1994. The contribution of small enterprises to employment growth in southern and eastern Africa. World Development 22(12): 1881-1894.

Mutahaba, G., R. Baguma, and M. Halfani. 1993. Vitalizing African Public Administration for Recovery and Development. Kumarian, West Hartford, Conn.

Ngom, T. 1989. Appropriate standards for infrastructure in Dakar. In: R. E. Stren and R. R. White, eds., African Cities in Crisis: Managing Rapid Urban Growth. Westview, Boulder, Colo., pp. 176-202.

O'Connor, A. 1991. Poverty in Africa: A Geographical Approach. Belhaven, London.

Okpala, D. C. I. 1987. Received concepts and theories in African urbanisation studies and urban management strategies: A critique. Urban Studies 24: 137-150.

Onibokun, A. G. 1989. Urban growth and urban management in Nigeria. In: R. E. Stren and R. R. White, eds., African Cities in Crisis: Managing Rapid Urban Growth. Westview, Boulder, Colo., pp. 68-111.

Onibokun, A. G. 1994. Urban research in Anglophone West Africa: Towards an agenda for the 1990s. In: R. Stren, ea., Urban Research in the Developing World, Volume 2: Africa. Centre for Urban and Community Studies, University of Toronto, Toronto, pp. 231-282.

Pasteur, D. 1992. Good Local Government in Zimbabwe: Bulawayo and Mutare. Development Administration Group, School of Public Policy, University of Birmingham, Birmingham.

Rakodi, C. 1995a. Poverty lines or household strategies? A review of conceptual issues in the study of urban poverty. Habitat International 19(4): 407-426.

Rakodi, C. 1995b. Harare. Inheriting a Settler-Colonial City: Continuity or Change? Wiley, London.

Ravenhill, J. 1993. A second decade of adjustment: Greater complexity, greater uncertainty. In: T. M. Callaghy and J. Ravenhill, eds., Hemmed in: Responses to Africa's Economic Decline. Columbia University Press, New York, pp. 18-53.

Richardson, H. W. 1993. Efficiency and welfare in less developed country mega-cities. In: J. D. Kasarda and A. M. Parnell, eds., Third World Cities: Problems, Policies, Prospects. Sage, Newbury Park, Calif., pp. 32-57.

Riddell, R. 1993. The future of the manufacturing sector in Sub-Saharan Africa. In: T. M. Callaghy and I. Ravenhill, eds. Hemmed in: Responses to Africa's Economic Decline. Columbia University Press, New York, pp. 215-247.

Rondinelli, D. A. and J. D. Kasarda. 1993. Privatization of urban services and infrastructure in developing countries: An assessment of experience. In: J. D. Kasarda and A. M. Parnell, eds., Third World Cities: Problems, Policies, Prospects. Sage, Newbury Park, Calif., pp. 134-160.

Seddon, D. 1995. Economic prospects for Africa. Review of African Political Economy no. 64: 275-278.

Sklair, L. 1991. Sociology of the Global System. Harvester/Wheatsheaf, New York.

Smith, D. M., ed. 1992. The Apartheid City and Beyond: Urbanization and Social Change in South Africa. Routledge, London, and Witwatersrand University Press, Johannesburg.

Stren, R. E. 1989a. Urban local government in Africa. In: R. E. Stren and R. R. White, eds., African Cities in Crisis: Managing Rapid Urban Growth. Westview, Boulder, Colo., pp. 20-36.

Stren, R. E. 1989b. The administration of urban services. In: R. E. Stren and R. R. White, eds., African Cities in Crisis: Managing Rapid Urban Growth. Westview, Boulder, Colo., pp. 37-68.

Stren, R. E. 1991. Large cities in the Third World. In: UNCHS, Metropolitan Planning and Management in the Developing World: Abidjan and Quito. United Nations Centre for Human Settlements, Nairobi, pp. 1-30.

Stren, R. E. 1993. "Urban management" in development assistance: An elusive concept. Cities 10(2): 125-139.

Stren, R. E. ed. 1994. Urban Research in the Developing World, Volume 2: Africa. Centre for Urban and Community Studies, University of Toronto, Toronto.

Stren, R. 1995. The GURI project: Urban researchers on three continents focus on governance. Paper to the Global Urban Research Initiative Workshop/African Research Network for Urban Management, Nairobi, November, mimeo.

Stren, R. E. and R. R. White, eds. 1989. African Cities in Crisis: Managing Rapid Urban Growth. Westview, Boulder, Colo.

Stren, R., M. Halfani, and J. Malombe. 1994. Urban politics in Kenya and Tanzania. The Urban Age 2(2): 1-5.

Swilling, M. 1994. Towards an urban research agenda for southern Africa in the 1990s. In: R. Stren, ea., Urban Research in the Developing World, Volume 2: Africa. Centre for Urban and Community Studies, University of Toronto, Toronto, pp. 283-374.

Swilling, M., R. Humphreys, and K. Shubane. 1991. Apartheid City in Transition. Oxford University Press, Cape Town.

Tomlinson, R. 1994. Urban Development Planning: Lessons for the Economic Reconstruction of South Africa's Cities. Witwatersrand University Press, Johannesburg, and Zed Books, London.

TourM., S. Ouattara, and E. Annan-Yao. 1992. Population dynamics and development strategies in the Ivory Coast. In: M. Tournd T. O. Fadayomi, eds., Migrations, Development and Urbanization Policies in Subsaharan Africa. CODESRIA, Dakar, pp. 7-50.

Tripp, A. M. 1992. Local organizations, participation and the state in urban Tanzania. In: G. Hyden and M. Bratton, eds., Governance and Politics in Africa. Lynne Rienner, Boulder, Colo., pp. 221-242.

Vanderschueren, F., E. Wegelin, and K. Wekwete. 1996. Policy Options for Urban Poverty Reduction: A Framework for Action at the Municipal Programme Level. UNDP/UNCHS/World Bank, Urban Management Programme Policy Paper 20. World Bank, Washington D.C.

Woods, D. 1992. Civil society in Europe and Africa: Limiting state power through a public sphere. African Studies Review 35(2): 77-100.

World Bank. 1991. Urban Policy and Economic Development: An Agenda for the 1990s. World Bank, Washington D.C.

Zeleza, T. 1994. The unemployment crisis in Africa in the 1970s and 1980s. In: E. Osaghae, ea., Between State and Civil Society in Africa: Perspectives on Development. Council for the Development of Social Science Research in Africa, Dakar, pp. 75-122.