GISdevelopment.net --> Proceedings --> GISDECO

Politics and performance: The implications of emerging governance arrangements for urban management approaches and information systems

Carole Rakodi
International Development Department,
School of Public Policy,
University of Birmingham, UK

1. Introduction
Use of the term governance implies that urban management is not the exclusive responsibility of public sector agencies, but involves relationships between government and non-governmental actors. First, the recent evolution of thinking and practice with respect to the distribution of responsibilities for urban management is reviewed, focusing on political arrangements and processes, decision making, accountability, relations between government agencies and resource allocation. This is followed by a critical analysis of attempts to define criteria by which the quality of urban governance arrangements and performance might be judged. Finally, some implications of recent trends in urban governance for the development and use of geographical information systems will be discussed.

2. From government to governance
To govern is to steer, guide, direct, control, regulate, influence or determine. In particular, it is to rule with or exercise authority and to administer the affairs of a state. Government, therefore, refers “.. to the formal institutional structure and location of authoritative decision making” (Stoker 1998)(p.34), embracing the legislative and executive branches of the state apparatus and those who control them (Hewison et al. 1993). Governance, in contrast, refers to “.. the action, manner or system of governing in which the boundary between organisations and public and private sectors has become permeable… The essence of governance is the interactive relationship between and within government and non-governmental forces” (Stoker 1998, p.38). It implies, Stoker asserts, joint action and thus a common purpose, a shared framework of values and rules, continuous interaction, and the desire to achieve a collective benefit which cannot be achieved by either acting separately. It is about relationships between the state and ‘civil society’, rulers and the ruled, government and the governed. It implies interdependence but does not prejudge the locus or character of real decisional authority, instead being concerned to disentangle the relationships and practices involved in governing (McCarney et al. 1995). Governance is about the way the power structures of the day and ‘civil society’ inter-relate to produce a civic public realm (Swilling 1997). In many discussions of governance there is, however, considerable ambivalence over how ‘civil society’ is constituted, in particular where the private sector, powerful individuals or informal associational life fit in a conceptualisation of governance as the relationships between the state and civil society.

Until the 1980s, the predominant assumption was that governments had the authority and capacity to govern: to formulate and implement policy, and to realise development goals. Translated into an urban context, the state-led approach to development implied that public sector organisations could plan and manage urban development and the debate focused on an appropriate allocation of roles and responsibilities between central and local government and between the administrative departments of government and semi-autonomous public sector agencies. The tasks urban government is expected to perform are similar everywhere (Davey 1996)(p.47):

• provide infrastructure for the efficient operation of cities
• provide services which develop human resources, improve productivity and raise the standard of living of residents
• regulate private activities that affect community welfare and the health and safety of the urban population
• provide services and facilities that support productive activities and allow private enterprise to operate efficiently.

However, the organisational structure of urban government, the precise allocation of power and responsibility between organisations and the capacity of public sector agencies to perform these functions effectively vary.

Frequently, the effectiveness of urban government, judged in terms of its technical competence, efficiency in the use of resources, financial viability, responsiveness to the needs of urban growth, sensitivity to the needs of the urban poor and concern for environmental protection was very limited. By the middle of the 1980s, the inefficiency of traditional approaches to urban planning, under-performance by local government and failures of service provision had in many places given rise to advocacy of a managerial rather than blueprint planning or administrative approach to urban demands (Devas and Rakodi 1993). Although the deficiencies of earlier approaches were explained in terms of the structure of public agencies, the distribution of functions between them and problems in central-local relations, as well as deficiencies in the internal organisation and management approaches of government agencies and shortages of human and financial resources, the newer approaches were still state-centred.

However, two changes were increasingly challenging the state-led view of urban management. The first was the influence of neo-liberal economic thinking, expressed in the macro-economic policies associated with structural adjustment and a desire to reduce the role of the state to the minimum enabling functions necessary to support the operation of markets. These ideas have strongly influenced thinking about appropriate approaches to urban infrastructure and service provision.

The second was rooted in dissatisfaction with the ability of existing political systems to respond to the views and needs of all social groups, whether in well-established systems of representative democracy; formal democracies in which power is actually concentrated in the hands of a few; or authoritarian, bureaucratic and one-party states. The emergence of broad social movements, the proliferation of new forms of social organisation and demands for increased political participation led to the dramatic (re-)democratisation of many authoritarian regimes in the late 1980s and early 1990s, attempts to revitalise surviving democracies (such as India), and a re-focusing of attention from formal political structures and governments as the locus of decision making authority to the role of civil society in exercising democratic rights and functions (Stoker 1998). Increasingly, the need for forms of government that were less heavy-handed, inflexible and unsubtle than traditional forms were recognised, underpinned by shifts in political ideology both on the right (freedom and individual choice) and left (recognition of limitations to state capacity). Greater attention had to be paid to civil society organisations and social movements, both because of the role they had played in challenging undemocratic regimes and also because of the status many won for themselves as a result in the new constitutions drawn up to accompany the restoration of multi-party democracy, for example in Brazil, South Africa and the Philippines. Closely associated with formal democratisation at the national level were calls for decentralisation, based on the assumption that decentralised government is able to coordinate sectoral activities, more knowledgeable about local conditions, more accountable to local populations and so better able to match resources to local needs and priorities. In towns and cities, recognition of the important role of non-state actors and civil society organisations in the production and management of the urban built environment, filling gaps in state provision of services, and holding formal democratic structures to account was coupled with renewed attention to democratic decentralisation (Blair 2000; United Nations Centre for Human Settlements 2001).

These challenges to traditional public sector approaches to urban government have had an impact on various aspects of governance and management in the 1990s, including political arrangements, decision making processes, ways of ensuring accountability, the allocation of roles and responsibilities between public sector agencies and other potential services providers and between different levels of government, and the mobilisation and allocation of financial resources for urban development. These trends will be discussed in the remainder of this section.

2.1 Political arrangements for urban governance
In authoritarian political systems, local government typically had a number of common characteristics (Ruland 1992):

• Capital cities were brought directly under central government political or administrative control (and in fact central control of some cities, such as Buenos Aires and Mexico City, survived democratisation) (Nickson 1995)
• Local elections were generally abolished or suspended
• Local government (generally via appointed officials) was used to consolidate the power of the central government – it had only a support role in urban development and no decision making autonomy and the extent to which it delivered services was related to the national government’s need for support from the urban population. Examples have been documented from Thailand, Malaysia and the Philippines (Ruland 1992), Latin America (Nickson 1995) and Mozambique (Alexander 1997; Swilling 1997).
• Regime-sponsored neighbourhood organisations were created to increase control, disseminate the ideology of central government and suppress political dissent. For example, in the Philippines barangays were effectively arms of the security agencies (Ruland 1992), while in one-party states the neighbourhood organisations were branches of the political party (Swilling 1997; Clarke and Buys 1999).
• Social movements, regime-critical NGOs and self-help organisations were stigmatised and persecuted. However, while independent NGOs were limited by legislative means or repression, government-sponsored NGOs were often encouraged (for example in Indonesia and Thailand) (Korff 1986; Porio 1997).

Thus in authoritarian regimes, local government was used not merely to consolidate the power of central government but also to contribute to the de-politicisation of society by tight grassroots control (Ruland 1992). Change occurred as a result of internal collapse of the regime, changing social and economic conditions, economic liberalisation and/or the emergence of social movements. The new political arrangements established at both national and local urban levels following democratisation were, with few exceptions, formal representative democracies based on the theoretical assumptions of liberal democratic theory (Held 1996).

In such political systems
“.. liberalism determines the nature of the state (formal, abstract), its structure (separate from the autonomous civil society, a clear separation between public and private), its rationale (protection of the basic rights of its citizens) and its basic units (individuals rather than groups or communities). Democracy specifies who constitutes the legitimate government and wields the authority inherent in the state (the elected representatives), how they acquire authority (free elections, choice between parties) and how they are to exercise it (in broad harmony with public opinion)” (Parekh 1993)(p.165).

However, in addition to the widely recognised problems with such systems, the difficulties experienced in consolidating formal representative democracy in previously authoritarian or oligarchical regimes have provoked doubts about the applicability of political models based on these assumptions, rules and institutions to countries in the South. Even in well-established democracies there is a danger that voting becomes merely a means of selecting between competing elites for leadership, leads to tyranny of the majority and disenfranchisement of minorities and can be manipulated to sustain oligarchic rule.

Moreover, it is difficult to ensure accountability of a complex and entrenched bureaucracy, citizenship is seen as essentially passive (except during periodic elections) and equality of political power is undermined by socio-economic inequality (Miller 1993; Held 1996). An additional problem in new democracies is the liberal definition of individuals as essentially self-contained persons, protection of whose rights is the primary purpose of the political system, in societies where individuals are defined in communal terms. In many societies, social groups believe that their members’ individual rights can and should be legitimately restricted; freedom, equality, loyalty, power and authority are conceptualised differently; and representation tends therefore to be communal, and to oblige an individual to distribute the state resources to which political office gives access to members of his or her social group (Parekh 1993; Preston 1998; Chabal and Daloz 1999). Further, the winner/loser system is thought to sit badly with the culture of consensus and competitive multi-party elections to be divisive, especially where parties are not formed around policy platforms and electoral discourse is consequently forced back on personal characteristics and local (often ethnic) identity, one of the problems which led to the replacement of earlier multi-party democracies in the first place (Allen 1995; Hodder-Williams 1999). The tendency of a democratic system to focus on leadership choices based on elite competition or membership of ascriptive groups is exacerbated by inhibitions on political participation by the poor, the weakness or lack of autonomy of civil society (Attahi 1997) and the inadequacy of (or lack of control over) state resources to implement domestically determined political agendas (Abrahamsen 2000). Finally, the relationship of democratic institutions to other authority structures and their influence on state-society relations is often problematic, especially in Africa (Mamdani 1996; Sklar 1999; Hyden 2000). As a result, in Latin America, for example, opinion polls in the late 1990s showed that only 27% of the population had confidence in existing democratic institutions (UNCHS 2001, p.66).

These problems are replayed in cities and towns when democratic government is (re-) constituted at the urban level. However, their form, the ways in which some of the problems have been tackled and their implications for urban government vary between countries. To understand the political system and processes at the urban level, it is necessary to identify the key political actors, ascertain how they perceive their interests and seek explanations of how they act. In ten cities studied in a recently completed research programme, the actors playing an important role were found to include (Devas et al. 2001):

Politicians and political parties. Mayors may be symbolic (in systems modelled originally on the UK) or executive. If the latter, these politicians are the most significant actors at the city level, but elected councillors and national/state politicians may also be important. Elected executive mayors wield considerable power and may be responsive to the needs of poor residents if their votes count. Successive mayors in Cebu City in the Philippines, for example, have depended on electoral support from poor residents but the picture with respect to delivery on their campaign promises is mixed (Etemadi 2000). Appointed chief executives are not locally accountable – in Kumasi, Ghana, the appointed mayor is able to override elected Assembly members (Devas and Korboe 2000). Political parties rarely have a coherent policy platform; instead they are treated by aspirants to political office as a means of organising and delivering the necessary votes (see also (Pasteur 1996; Mengers 1997).

Public sector agencies, including both local and higher levels of government. Bureaucracies and their staff often promote their own agendas (self interest, professional standards), resisting or subverting local political decision making; and co-ordinating action is often hindered by competition between agencies. The division of responsibilities, power and resources between central and local government is an important determinant of the latter’s ability to respond to local priorities and needs, as is the extent to which semi-autonomous bodies are responsible for the delivery of essential services. Where local government is overshadowed by parastatals accountable to higher level political and administrative levels, municipal capacity is likely to be undermined. These issues are discussed further below. NGOs. These may undertake their own developmental initiatives, generally on a small scale and with limited impact on wider decision making; provide services for the government, potentially sacrificing their independence to access funds; act as intermediaries between residents and public sector agencies; support local associations of residents or micro-entrepreneurs; or campaign for civil rights and political participation (Douglass 1998; Desai and Preston 2000). Studies in the ten cities referred to above and elsewhere have found that NGOs tend to be small, to adopt competitive rather than collaborative ways of operating, and to have weak links with each other, poor residents and the political structures. As a result, they are, in practice, rarely influential political actors (Acharya and Thomas 1999; Sahley and Danziger 1999; Devas et al. 2001).

Trade unions often played an important role in struggles against authoritarian rule, but they are primarily concerned to protect the interests of their members, who are not the poorest workers. Changes, for example to arrangements for municipal service delivery, may be resisted by the unions of municipal workers (as in South Africa and India) (Franceys and Sansom 1999; Dutta 2000; Beall et al. 2001). Overall, the pressures of globalisation and liberalisation have weakened trade unions in most countries.

Enterprises and their associations range along a continuum from large scale enterprises, the significance of which to the local economy is such that they may wield systemic power without needing to enter into explicit political alliances, to micro-enterprises, which can rarely influence decision making to their advantage without organisation. Whether organisations such as Kumasi’s market traders (King et al. 2001) or Cebu’s street vendors and trisikad drivers can influence political decision making in their own favour depends on whether the changes for which they are pressing compete with the interests of other groups, especially large scale enterprises (Etemadi 2001). The latter have, for example, backed clearances of street vendors from city centre streets at various times in Kumasi, Mombasa, Cebu and Colombo (Refs).

Associations range from informal groups based on identity or voluntary membership to formal associations. They may or may not have a political agenda and influence (see also Tripp on Dar es Salaam)(Tripp 1997).

Residents can exert power and influence over decision making only through their votes or the culture of personal contacts within the political or bureaucratic systems, unless they are organised into associations around their area of residence or interests, which raises a number of issues about leadership. This can be formal or loose and informal, unified or fragmented and competitive, and motivated by altruism or self interest. Ward and Chant distinguish four main categories of community leadership with different degrees of legitimacy, different propensities to mobilise and act and different relations with the local state. They also note that typically not all interests are represented by community leaders – in particular the interests of women are often under-represented (Ward and Chant 1987) (see also Sahley and Danziger1999 on Lima). Rodriguez and Winchster, reviewing the Latin American experience, identify the conditions for participation of the ‘popular social classes’ in government as, firstly, the existence of popular organisations with some presence at the local level, and secondly, the occupation of positions in the municipality by parties or individuals (especially mayors) favourable to participation (Rodriguez and Winchester 1996). There is some evidence that pro-poor policies are more likely to be politically viable if universalistic initiatives that benefit all groups are pursued (e.g. primary education for all, malaria reduction) rather than targeted poverty reduction initiatives (Blair 2000).

Power and influence is exercised through political strategies and practices which may relate to the formal political structure or be played out through informal relationships. The key aspects are representation and accountability. The latter will be discussed further below. With respect to the former, the question is how and whether the representative mechanisms represent the interests of different social groups.

Conceptions of the purpose of ‘representation’ influence how elected representatives behave once in office: whether they further the interests of their financial backers or known supporters, fulfil their electoral promises, advance their own political careers or line their own pockets. Where there were significant concentrations of poor people in the ten cities studied and ward-based elections, some indicated greater responsiveness by councillors to their constituents than in at-large electoral systems, especially those with a closed party list (Devas et al. 2001; see also Nickson 1995). In Bangalore, for example, ward councillors are accessible and accountable to poor voters, who bargain their votes for the benefits that councillors can deliver, reflected in much higher electoral turnouts in low than higher income wards (Benjamin 2000). However, systems of proportional representation, or mixed systems as in Johannesburg (Beall et al. 2001) or Venezuela (Nickson 1995) may give better representation to minorities. Quota systems in which a proportion of seats are reserved, as in India or the Philippines, attempt to ensure that groups such as women and minorities are adequately represented, but the accountability of such representation is not always certain (Blair 2000).

Political decisions are strongly influenced by the short time horizons of the political process. Short term, visible projects take precedence over the longer term, more mundane or less visible tasks required to improve service delivery and infrastructure maintenance, operationalise regulatory frameworks or improve municipal organisational capacity. Rules on terms of office (frequency of elections, length of term, eligibility for re-election etc) are, therefore, potentially important and can be changed to overcome some of the worst effects of inordinately short political time horizons, although the operational modes of public agencies must also recognise the legitimate political need for visible results.

Finally, city or metropolitan government is remote from residents, giving rise to a democratic deficit which may be overcome by instituting representative systems at a sub-city level, although councillors may resist if they see these as a threat (Jalal 2000). Cebu’s barangays and Recife’s micro-regions are examples of accessible elected bodies that influence decisions on expenditure at both city and local levels and can respond to local priorities, in the latter through participatory budgeting (Souza 2000; Devas et al. 2001; Etemadi 2001). Some of the ten municipalities in Abidjan introduced neighbourhood management councils or committees to ensure more participatory management of local services (Attahi 1999).

2.2 Decision making
In theory, in a democratic political system, decisions are made by the elected leaders or representatives in general accordance with public opinion. The role of government bureaucracies is to provide decision makers with sufficient information to enable them to make informed choices, while ensuring that the rule of law and government procedures designed to ensure fiscal balance, accountability and efficient performance are adhered to.

The operation and maintenance of services, exercise of regulatory functions, management of revenue collection and expenditure, and management of personnel are, in practice, especially in large cities, both complex and routine. The issue of delegation is, therefore, important: what are the strategic policy choices and management decisions that should be taken by elected politicians and what can more sensibly be delegated to bureaucrats; what should be the structure and role of the executive and the legislature; and is decision making by a limited number of elected representatives sufficient to reflect the views of diverse urban populations?

In addition, given the (increased) dispersal of functions to organisations outside local government, how can decision making that involves a number of different central and local government and non-government bureaucracies best be handled?

The arrangements for executive control are important: an elected executive mayor may deliver more decisive city government and clearer accountability than an executive council system, but the checks and balances may be fewer and the risks greater than in a system where councillors have both legislative and executive powers. Whether that is the case depends on the scrutiny powers available to the legislature and its ability to exercise them, as well as on the management arrangements, especially the tradition of confidence appointments (Nickson 1995; Pasteur 1996; Batley 1997; Devas et al. 2001). The political and administrative relationships between central and local government are also important.

The limitations of both state organisations and formal structures of representative democracy have increasingly been recognised. Although the design of these structures can be adapted to improve decision making, as briefly discussed above, introduction of the mechanisms of direct, deliberative or participatory democracy is increasingly advocated (Miller 1993; Held 1996; Smith and Blanc 1997). These are based on more active notions of citizenship and complement representative democracy rather than replacing it. In addition, decentralisation and the division of responsibilities between increasing numbers of agencies has led to increased recognition of the need for negotiated agreements between actors, perhaps along the lines of French contrats de ville (UNCHS 2001). One form these attempts at more inclusive negotiated decision making has taken is collaborative planning (Healey 1997; Rakodi 2000)

An example is the City Development Strategy process promoted by UNCHS, UNDP and the World Bank through the Urban Management Programme, which also broadens the traditional view of an urban plan as a land use plan to a multi-sectoral strategic development plan with links to implementation. The CDS process is intended to arrive at a shared view of priorities, produce more appropriate policies, widen ownership of decisions and result in more action than traditional approaches to policy formulation and planning. Ideally, the political process should be capable of doing all these things and where democracy is consolidated, the CDS process can be integrated into regular decision making practices, increasing the likelihood of implementation. Where it operates parallel to and slightly outside the regular government structure, as in Dar es Salaam, its acceptability and effectiveness are more problematic (Kombe 2001). However, it should not be assumed that agreement on priorities and consensus on policies and actions is possible or that all have equal influence – even in attempts at inclusive decision making, it is likely that some interests will be excluded or marginalised (Rakodi 2002).

In none of the ten cities referred to above had a CDS process been instituted at the time of study, but a number of other extensions to the formal representative system were observed. These included integration of organised interests into political structures, as in Recife, Cebu and Johannesburg; formal consultation of interests (Pasteur 1996); involvement of users in service management; and participatory democracy at the neighbourhood level, as in Colombo (under the former Million Houses Programme) and in participatory budgeting in Recife or Porto Alegre (Russell and Vidler 2000; Souza 2000; Abers 2001). Such plural structures and mechanisms of representation can ensure that the views of the poor are heard in policy making, programme design and implementation, especially at a project or area level. The Ley de Participatión Popular in Bolivia introduced a plural representative system which includes democratic decentralisation at municipal level and recognises CBOs. The latter send representatives to a Vigilance Committee which has powers to hold municipalities to account. However, Blair notes that men’s rather than women’s groups are recognised where both exist and occupational groups are not represented (Blair 2001). Similarly, a case study of the Community Development Forum in Diepsloot, Johannesburg showed that some poor groups are excluded (Beall et al. 2001).

The danger with forms of direct democracy, such as referenda, is that they over-simplify issues and exacerbate the tyranny of the majority that is already a danger of many formal representative democratic systems. Deliberative and participatory democracy may also be problematic. On advisory bodies government nominees tend to have more information and influence than residents, while citizen representatives may fail to report back or become incorporated into the agency’s perspective (Smith 2000). Participation may be irregular and selective, not inclusive; may function merely to legitimate decisions taken by the authorities; and may create tensions with councillors, professionals and other bureaucrats (Smith and Blanc 1997). ‘Partnerships’ are task or area specific, time-limited and often conceal unequal power relationships. Although they have positive aspects, therefore, they are not a substitute for the ongoing city-wide activities of government (UNCHS 2001).

2.3 Accountability
Accountability mechanisms in the public sector are essential to hold to account those who hold the public trust and to ensure quality control and probity (Olowu 1999). However, absolute control is unnecessary, infeasible and inefficient, so there is a difficult boundary to be drawn at the limits of discretion. In theory, elected municipal government is more easily held to account than national government because it is physically relative accessible, local politicians are relatively active and numerous and there are likely to be local media (Davey 1996). Accountability may be achieved by internal and external mechanisms, as well as political voice (Olowu 1999). These are underpinned by transparency with respect to procedures and information.

Internal mechanisms include, for example, merit-based recruitment and reward systems, codes/regulations, and internal reviews and audits. External mechanisms include administrative responsibility to the political executive (although collusion can instead encourage corruption and reduce efficacy) (Mengers 1997), judicial channels and legislatures. In practice, legislatures are often not good at holding the executive to account, because they may also have executive functions (especially in the British model), party systems are fragmented and unstable (Blair 2000), or members of the legislature broker patron-clientelistic relations between the mayor or bureaucrats and their constituents (Nickson 1995). More recent initiatives include ombudsman-like institutions and independent scrutiny bodies. Finally, central government regulation is important but is generally either too heavy-handed and controlling, or ineffective because of limited capacity or political alliances between central and local politicians and bureaucrats (Haque 1997; Blair 2000).

Voice exercised through the electoral system is an effective way of ensuring accountability if voter registration and turnout is universal (or at least widespread) and the ratio of elected representatives to the population is adequate (Nickson 1995). However, elections may only legitimise the power of the local elite (Haque 1997), they are occasional and if competition is based on alternative policies at all, these are very broad and general. Voice may be supplemented by public meetings, the media (although local media often lack the resources for investigative journalism), the activities of civil society organisations and opinion surveys/referenda (Campbell 1997; Paul and Sekhar 1997). Systematic public feedback makes it more difficult for agencies to dismiss complaints as anecdotal and provides a weapon for citizens and watchdog organisations.

A variety of mechanisms of accountability were observed in the cities studied by Devas et al, ranging from elections, through voice (public meetings, a free media), to the exercise of controls internally by the executive and/or legislature and externally by government. The mechanisms were not studied in detail but clearly varied in their effectiveness and accessibility to the poor (Devas et al. 2001). Blair asserts that those accountability mechanisms with the most links (civil society organisations and political parties) are more important, although he fails to unpack and specify his broad definition of civil society. He concludes that renewed local democracy initially provides opportunities for elite challenge and only in some cases leads to increased scope for middle groups, minorities and the poor to gain influence and exercise scrutiny (Blair 2000). The chance of this happening is greatest when there are multiple structural channels and mechanisms for ensuring accountability, as in Bolivia (Blair 2001).

2.4 Relations between government agencies
Urban governance functions are divided between central and local levels of government and also between line departments, other public sector bodies and, increasingly, non-governmental operators. The roles and responsibilities of the various levels of government may be based on the principle of subsidiarity and enshrined in a constitution or may be defined and allocated by central government. Even in the former, local government may perform a combination of delegated and autonomous tasks and may be served or complemented by central government technical or operational agencies. Similarly, even where local government is an agent of central government, it may have wide powers and considerable autonomy or merely a residual role (Davey 1996). The extent of democratic decentralisation depends primarily on central government motives and objectives but also on political history and the salience of local politicians. Central government’s motives in theory include political power sharing, to enhance local responsiveness and accountability, and economic efficiency, based on the hope that local decision making will be more efficient because it will more accurately reflect consumer preferences, and more cost effective, especially when financial resources are locally generated, and there is local accountability i.e. devolution and commercialisation of service delivery will together improve allocative and productive efficiency, equity and effectiveness (Nickson 1997).

In practice, a central government’s motives for power sharing may include desires to enhance political stability and its own hold on power and reduce the extent to which residents blame it for poor performance. However, local political organisation has more often been seen as a threat than a benefit, demonstrated by the frequency with which central governments suspend local elected bodies, even in democracies (Nickson 1995; Chaplin 1999) and the extent to which they vest powers and resources in unelected bodies. For example, in Bangalore, much infrastructure investment and large scale land development is undertaken by State parastatals (Benjamin and Bengani 1998).

In practice also, central government’s motives may be more to do with reducing demands on national revenue than increasing economic efficiency, although the desire to encourage local revenue generation and expenditure management struggles with an equally strong desire to maintain control, while evidence on the greater efficiency of resource allocation by local government is scarce and inconclusive. The ability of local government to fulfil devolved roles and responsibilities depends crucially on its capacity to do so, which is related to the financial, physical, legal and human resources available to it as well as the political arrangements. Often, that capacity is limited: the powers, political will and administrative systems to raise and spend substantial local revenue are lacking, while central-local fiscal transfers are irregular, unpredictable and inadequate to compensate for revenue shortfalls and inequalities between resource rich and resource poor cities and towns. Moreover, inadequate finance, central government controls and the relative unattractiveness of working for local government restricts the expertise available to local authorities (Bird 2001; Ebel and Vaillancourt 2001).

The chronic weakness of local (as well as central) government and failures in urban service delivery, coupled with economic crisis, led to radical reconsideration, at least in theory, of the arrangements for service provision. Especially where the World Bank and other external agencies have been influential, this has resulted in reforms to arrangements for the provision of urban water supply, solid waste management, electricity, public transport and so on.

The key underlying principles for the new approach are, first, that these services should be treated as economic goods and, second, that direct and indirect provider roles should be reallocated where appropriate (Kessides 1993; Nickson 1997; Bartone 2001; Batley 2001; Bird 2001; Gulyani 2001). Only those components of service delivery systems with public or merit good characteristics or where natural monopoly or other factors lead to market failure should, it is asserted, by provided by public sector agencies. However, the public sector must continue to exercise indirect provider functions such as planning, co-ordination and regulation and may continue to provide capital for investment and subsidies. To ensure financial sustainability and generate funds for extension of services, all users should pay and charges should, in principle, cover marginal costs. To increase responsiveness to consumers and encourage efficiency, a more performance oriented approach is required: replacing a supply-led approach to urban water supply and sanitation with a demand orientation; increasing competition by encouraging new entrants, especially large but also small informal operators; unbundling sectors by task or area; and subcontracting installation, operation, maintenance or particular tasks under a variety of arrangements.

Although public ownership and provision of urban services is still the most common, and resistance to change is widespread, reforms in this sector have included

• Arrangements for increased and more predictable financing for a) public sector programmes, often through central government administered municipal loan funds for which the eligibility criteria include improved financial management and b) community level improvements, sometimes through central or local government development funds to which applications can be made by CBOs or NGOs for funding for local projects (UNCHS 2001).
• Commercialisation and increased managerial autonomy for service providing departments and agencies by, for example, corporatisation (the establishment of publicly owned companies).

The number of examples of PSP has increased dramatically in recent years, with large private operators bidding (often less rather than more competitively and transparently) for water and sewerage concessions and contracts, BOT schemes, solid waste collection contracts etc.

Increasingly, the important existing role played by informal independent operators (micro-enterprises, CBOs, NGOs etc) is being recognised and, in some cases, attention is being paid to the possibilities of incorporating them in the planning and operation of service delivery systems (Cotton et al. 1998; Bennett et al. 1999).

These changes to service delivery arrangements have implications for urban governance. In theory, government remains in control of decision making with respect to priorities, standards and targets, and can ensure accountability of autonomous agencies through the political process and private operators through competitive market mechanisms or tying the release of public funds to adherence to contractual conditions. While in some cities, PSP in service provision has resulted in improved performance and the extension of services to low income residents, in others outcomes have been mixed or unsatisfactory. Corruption in the award of contracts, by no means restricted to developing countries, has prevented the problems being addressed in some cities. In others, lack of expertise and experience in the public (and even the domestic private) sector accounts for poor contract design and enforcement. Only in a few has government, often pushed by public unrest, rescinded contracts, not least because of the threat of legal action. The dominance, especially of urban water and sewerage, by a few (mainly French based) multinationals means that bargaining power and expertise is unevenly distributed between contractors and their clients, exacerbating the governance problems.

Furthermore, the proliferation of arms length providers may strain weakly developed democratic systems. One of the functions of urban government which is closely linked to infrastructure provision and service delivery, as well as to private property development, is development planning. Decision making aspects of this have been discussed above, but the relationships between government land use planning and development regulation functions and the new arrangements for service delivery are ignored in recent literature on the latter.

While PSP can, if competitive and well designed, increase efficiency and coverage, it is clear that governance places as many demands on government as direct public sector provision (Nickson and Franceys 2001). Thus whether ownership and operation is in the public or private sector is less important in ensuring improved service delivery than political stability, clear objectives backed by political will and administrative capacity in the relevant public sector bodies, especially local government.

3. Good governance?
At both the national and urban levels, as noted above, the acceptance that government cannot achieve developmental objectives alone has led to a broadening of involvement in policy formulation and implementation to encompass other societal actors, captured in the term ‘governance’. In assessing whether or not the relevant development objectives are being achieved, therefore, earlier concerns with appropriate performance indicators and monitoring systems for public sector agencies have given way to attempts to specify what is meant by ‘good governance’ and how progress towards this might be monitored and assessed. The normative identification of the characteristics of good governance rests on (often implicit) assumptions about the nature of political and administrative cultures and systems and is, in the literature, rarely separated from analytical accounts of how governance responsibilities are in practice fulfilled. Here, recent attempts to define good urban governance will be briefly presented and critiqued.

Criteria for good urban governance have been advanced by a number of development organisations and commentators (see Table 1). UNDP distinguishes economic, political, administrative and systemic governance. Economic governance refers to decision making which (in)directly affects economic activity; poltical governance to decision and policy making by the legitimate and authoritative state; administrative governmance to the implementation of policy through an efficient, autonomous and accountable public sector; and systemic governance to the societal structures and processes that guide socio-economic and political relationships, protect cultural and religious values and produce an environment in which health, freedom, security and quality of life can be secured. The criteria of good governance derived from the above, as expressed by TUGI (The Urban Governance Initiative), are given in Table 1. Strengthened indigenous capacity is sometimes added by UNDP.

Table 1 Criteria of good urban governance

For the World Bank, good governance includes competent, efficient administration and a legitimate and democratically elected government. The criteria it uses are summarised by Adamolekun (Adamolekun 1999):

• The rule of law, underpinned by an independent court system, which forms a basis for resolving conflicts, helping to ensure respect for property rights and contracts, and preventing capricious government action
• Electoral legitimacy, derived from periodic open, competitive and free elections that provide an elected political executive with a mandate to govern
• Freedom of expression and association, to nurture a democratic political culture and increase accountability, and to provide partners for the delivery of services and micro-credit
• Accountability and transparency – accountability implies responsive to the demands of the governed, enforced by respect for the rule of law and an independent judiciary, together with elected bodies exercising oversight, widespread access to information and a free media. Weak accountability and lack of transparency is demonstrated by corruption “the misuse of public office for private gain” (Adamolekun 1999, p.9)
• Development-oriented leadership i.e. leadership which is committed to the development and transformation of the whole society.
• Competent, efficient administration.

In the most recent World Bank document, these are expressed as goals and objectives for sustainable cities, to be realised through a mixture of components and preconditions, rather than criteria for good governance (World Bank 2000). They comprise

• accountability, transparency and integration of local government
• local government institutions sensitive to the needs of the poor
• cost effective fulfilment of local government service objectives, through sound financial management and access to market-based credit
• a social compact with users for fees-for-services

These are to be achieved by broad participation of all groups in governance, incentives to encourage probity and performance, review mechanisms, building service delivery capacity, transparency, and efficient revenue generation and financial management. They are also to be achieved by trust in local government by both the public and higher levels of government, although this seems to be an outcome rather than either a component or a precondition. Most recently, norms or principles of good urban governance have been developed as part of the UNCHS Global Campaign on Good Urban Governance. Seven norms are further spelt out as objectives and operational principles, with the latter subject to ongoing discussion and refinement. To the familiar criteria of participation, subsidiarity, equity (of access to decision making and the basic necessities of urban life), transparency, accountability, civic engagement and citizenship through participation, efficiency in the delivery of public services and in promoting local economic development are added sustainability in all dimensions of urban development (social, economic and environmental) and security of individuals and their living environment.

Friedmann, TUGI and UNCHS developed their criteria specifically for urban governance. TUGI’s ‘strategic vision’ is, in Friedmann’s conceptualisation, translated into ‘inspired leadership’, capable of articulating a common vision, obtaining consensus around it and mobilising the resources needed for implementation. In the UNCHS norms, the strategic vision refers specifically to sustainable human development, balancing the social, economic and environmental needs of present and future generations and including a commitment to poverty reduction. Accountability is to be ensured via channels for participation and monitoring, information on performance, or transparency; civic engagement via the right to participate, and both representative and direct democracy; and responsiveness via decentralisation, the right of citizens to claim rights, express grievances and obtain redress.

TUGI’s rather vague ‘consensus orientation’ is expressed more positively as institutionalisation of ways to resolve state-citizen conflict in Friedmann’s formulation. Reference is made in the UNCHS norms to the need for leaders to have the ability to reconcile divergent interests for the common good. The tension between a management/delivery model for service provision, aimed at increasing efficiency, and an access/demand model, which locks residents into arbitrary and dependent relationships with decision makers or gatekeepers can potentially be resolved through governance structures which build links between civil society and the state.

The first three lists of criteria do not mention gender, presumably subsuming it within criteria such as ‘equity’ (UNDP), responsiveness (UNDP, TUGI, Friedmann) and inclusiveness (Friedmann). However, Beall considers that gender sensitivity should be made an explicit criterion, including both women’s access to all the available means of civic engagement and government responses, via fostering gender awareness among both women and men in the political arena, policy and practice (Beall 1996); the UNCHS norms state that cities should be equitable and inclusive, specifically referring to equal representation of women and men in decision making, equal attention to their needs and equality of access to resources; and the World Bank notes the need for sensitivity by local government institutions to gender differences in service requirements.

The criteria of good governance are mostly based (implicitly) on political theories of liberal democracy and the organsational theories of the new public management, especially those advocated by the World Bank (Abrahamsen 2000). They tend to assume that the organisational arrangements of multi-party representative democracy are appropriate in all circumstances and will produce the desired outcomes of responsiveness and accountability.

However, the UNCHS norms are slightly more circumspect, following adverse reactions to the draft norms from countries such as China. None recognise the effects that inequalities of power and economic resources have on political outcomes, although the potential for conflict, difficulty of reaching a compromise and the danger that a ‘consensus orientation’ will fail to recognise conflicting, especially minority, interests is recognised by Friedmann. Although often the importance of an appropriate legal framework and economic liberalism are stressed, competent public sector management is also acknowledged as critical in achieving development objectives.

The criteria in Table 1 include a mixture of ends and means: goals, performance criteria, specific organisational arrangements and institutions. Contradictions between the criteria are not recognised. For example, equity is a criterion but, as discussed in the previous section, neither political actors nor civil society groups are necessarily interested in equity or equitable and inclusive in their practices. Furthermore, devolution to local government potentially makes it more important to local elites, who may then be unwilling to develop symbiotic relationships with civil society groups or to mobilise and redistribute resources lest they lost votes and power (Beall 1996). Operationalising and assessing good urban governance in an unequal world is complex and problematic and the desire to promote high standards should be informed by an understanding of political realities and urban experience, as analysed in previous sections.

4. Implications of changing governance arrangements for GIS creation and use
Urban management may be defined as incorporating a corporate policy and resource allocation strategy dealing with health, education, transport, local economic development, poverty reduction, infrastructure provision and land use, translated into programmes and projects for implementation. Operations involve the management of assets and services, involving planning and allocating responsibility for service delivery, maintenance, regulation and revenue generation. In an urban management system, a number of potential roles for GIS may be envisaged, both as a technical tool and as a communication medium. Some of these are listed below, in rough order of complexity and ambition.

• The storage, manipulation, retrieval and display of spatially referenced data, including land use, linear services and infrastructure, population, house types etc. For example remote sensing is being used to map land use change, land suitability and environmental sensitivity in Indian cities to a scale suitable for city or regional strategic planning. If linked to other databases, which are typically zonal (administrative or census areas) thematic maps can be produced (Kasturirangan and Rao 1997). The potential has been demonstrated in an analysis of Saharanpur City (population 378,000 in 1991) where satellite images, aerial photographs and field checks have been used to measure the loss of agricultural land to urban expansion between 1988 and 1998. The analysis shows the location of agricultural land losses, the quality of land lost and the nature of new temporary or permanent uses (Fazal 2000). Parcel-based systems for monitoring land supply and capacity also have potential for tenure registration and revenue generation. Traditionally, such GIS projects have been technocratic, top down and large scale, but more recently their potential for using them to make data available to non-governmental actors, including communities, has been identified (Hill and McConnachie 2001). Increasing the availability of data can potentially improve the political efficacy of residents, councillors and other organisations, although it is also noted that the selection of data and audiences to which they can be made available is not neutral, and can be manipulated for political and other ends.
• Automated data handling (e.g. payroll administration; valuation; billing and collection; storage, tracking and retrieval of development applications) is often the first attempt at computerisation made by local governments. Although not necessarily spatially referenced, there is potential for linking the data available from records to a GIS.
• Improving the availability and flow of information in and amongst organisations responsible for urban management, with a view to improved monitoring of change, anticipation of issues and monitoring of implementation.
• Land suitability analysis and hazard mapping, by linking population and environmental health data to identify areas at risk from environmental pollution; and by linking the degree of risk to population distribution and infrastructure in order to identify priority areas for preventive activities (Paulsson 1992).
• Predictive and prescriptive (or optimising) models of urban spatial interaction can be used to forecast the impact of changing land use and transportation infrastructure on the distribution of population and other activities and vice versa (Batty and Densham 1996). The former include, for example, shopping models and transportation models, and the latter models for locating facilities. Recent developments are increasing the speed of such models and visualisation, increasing the potential for using them as a basis for discussion in decision making and for any stakeholder to simulate alternatives. These models originally assumed rational/technocratic decision making, but as visualisation improves, it is suggested that they have potential for informing public discussion and political decision making (Batty 1997).
• Increased access to information via increased access to computer and the development of the web means that potentially a wider range of actors in urban areas can be informed about planning and management issues. As the potential for using web-based information resources including GIS (and models) interactively develops, there is also potential for using them in planning and decision making to increase opportunities for consultation during decision making and planning, through the development of on-line discussion forums. As the systems are further developed, there might also be potential for people to redesign what is viewed on screen remotely, either individually or in discussion with others. However, in addition to technological issues, such extensions to current uses raise a number of issues related to access, the objectives of such exercises and protocols for behaviour (Batty 1997).

The potential uses of GIS in urban management, however, must be considered in the light of political and administrative realities. First, the introduction of GIS involves not only technological problems but is also a complex process of managing change in environments characterised by both uncertainty, on the one hand, and entrenched institutional procedures and staff motivations on the other. There may be resistance from the less technically literate, uncertain of the use and potential of GIS and feeling threatened by it (Paulsson 1992; Lee Lik Meng and Tan TS 2002). Departmental rivalry is common in the public sector, and can be expected to increase with the increased involvement of other governance actors, particularly private sector operators, where commercial interests are often said to require confidentiality (Hill and McConnachie 2001; Paulsson 1992). In Penang, for example, the Federal department funding local plan preparation refused to use the municipal computerised cadastral base map, resulting in poor quality work for those local plans not prepared in house, and involving remedial work to these plans by the municipality (Lee Lik Meng and Tan TS 2002).

Second, the collection, processing and dissemination of data reflect social relations and are not politically neutral (Hill and McConachie 2001). The selectivity of data generated and inequitable access to it may cause problems in democratic contexts. This is illustrated by the last minute cancellation of the Pune environmental infrastructure project in India in 1998, after two and a half years’ preparation, because of the loss of political support. Concerns had been raised about costs, profit levels, viability and the process of awarding tenders. An analysis of the reasons for the about-turn stressed that the information made available was incomplete and not available to all, so that the consultation process had failed to convince all the stakeholders (Zérah 2000). Because resource availability differs, there is a danger that the information and expertise gap between rich and poor local authority areas and residents will increase (Hill and McConachie 2001).

Third, spatial planning (including planning for transport and other infrastructure and land use) is relatively ineffective and slow to change (Rakodi 2002). As a result it is marginal to the main decision making processes and may also be seen as a constraint on urban economic development and the activities of the private sector by local stakeholders and external agencies alike. In recent attempts to improve urban management, the financial capacity of urban agencies, environmental management, the health status of urban populations and the efficiency and sustainability of utility provision, spatial aspects (including land and tenure) have been neglected or dealt with separately from the traditional land use planning system (Burgess et al. 1997). Proposals to improve policy making and the plan preparation process include improving understanding and analysis of different components of the urban development process, legislative change, improved co-ordination, moving from inflexible blueprint plans to a combination of strategic and detailed plans and programmes, greater involvement of the public, better data and analytical techniques, and improvements to development regulation processes (Rakodi 2002). However, even if approaches and tools are improved this “.. will not necessarily increase the extent to which policies and plan proposals guide decision making and resource allocation, nor will improvements to the regulatory system necessarily increase the proportion of development which complies with plan proposals, regulations and standards” (Rakodi 2002, p.216), unless those involved understand and recognise in their practice its relationship to organisational dynamics, political processes and the wider governance environment.

Fourth, the development of GIS for planning purposes in the urban areas of developing countries is complicated by the fact that much development is informal, rights in property unregistered, registration procedures cumbersome and cadastres outdated or incomplete. As a result informal areas escape the property tax net and the development regulation system, as well as being under-provided with infrastructure and services. Parcel-based GIS systems are, therefore, difficult to develop except in formally planned parts of cities, even though a partial cadastre involving registration of occupation can be key to improving revenue generation and extending service coverage (Farvacque and McAuslan 1992). The technical ability for computerised tracking of change in the status of parcels exceeds the capacity of institutions to use parcel-based systems, even in the US, where in the late 1990s relatively few local authorities were using such systems (although 25%+ were building them), despite the availability of a cadastre and financial resources the exercise of effective control over all development. Problems included difficulties in defining developable land, categorising land in terms of its suitability for development and factoring in land costs (Moudon and Hubner 2000).

Last, GIS are only tools – their value depends on high level managerial and political support and institutional capacity to define information needs, set priorities and then make appropriate use of the information generated. However, corporate policy formulation and decision making capacity may be weakly developed, especially in local authorities organised on traditional sectoral lines, with an administrative rather than developmental view of their role) and operating to short political time horizons. Involvement of users in the design of a GIS system, and training and education for all those involved, are important, but may be insufficient, as different users are likely to have different requirements and practices which may not be compatible (e.g. in Bangkok). Long term commitment is needed, which is often negated by political urgency and conflicting priorities (Bishop et al. 2000), while the unpredictability of central-local fiscal transfers and financial crisis inhibit longer term projects – strategic development planning in general and GIS in particular.

In Penang, computerising the zoning map on a parcel base has taken five years to date, in part because the municipality had to create its own base map, since the Survey Department could not supply one. The base map took two years to develop. To it were then added layers, including topography, land use, buildings, heritage buildings, approved and committed development, roads and zoning. Development of this GIS for planning purposes has been slow and its uses so far confined to data capture and mapping. Nevertheless, Penang’s experience demonstrates the potential to develop a basic GIS at relatively low cost (Lee Lik Meng and Tan TS 2002). Despite its potential advantages, the development of a common multi-purpose digital map base and GIS for a city is costly, administratively complicated, slow and difficult. Because there are few practical short term pay-offs, it is, as the Bangkok experience shows, difficult to maintain support (Bishop et al. 2000). The value of GIS, Bishop et al argue, is in small strategic applications, for example spatial mapping of selected data relevant to service planning and delivery – the location of facilities, health status, or the use of integrated remote sensing/GIS to improve land suitability analysis (see also (Thomson and Hardin 2000).

5. Conclusion
“Today, governance involves multiple stakeholders, interdependent resources and actions, shared purposes and blurred boundaries between the public and private, formal and informal, state and civil society sectors, and greater need for co-ordination, negotiation and building consensus” (United Nations Centre for Human Settlements 2001, p.57).

The key tasks for government in this context are co-ordinating an increasingly complex and fragmented governance landscape, steering inter-dependent activities through new institutional frameworks and bargaining processes, and integrating and managing diverse networks. The question when considering these issues is under what conditions do different governance strategies and arrangements work effectively and for which purposes.

The repertoire of strategies comprises markets, hierarchies and networks, associated with private, public and civil society spheres respectively (UNCHS 2001). If markets are relied upon, there is a risk that approaches will be short term and fragmented, and will result in increased inequality – the need for steering and integration is not addressed. Reliance on hierarchy implies an emphasis on rule setting and organisational design to ensure co-operation and compliance. Typically this involves public bureaucracies, which allocate roles and responsibilities, devise and enforce rules, and design administrative practices that, in principle, balance efficient and equity. However, such bureaucracies lack flexibility and, because of their poor performance and corruption, legitimacy. Networks, on the other hand, use informal coalitions, trust and negotiation to produce co-operation. The key role of government in a wider governance system is steering, co-ordination and enabling, although the latter means more than merely enabling the market. These tasks are as, if not more, difficult than the traditional command, control and direct investment roles, raising questions about the expertise required, the financial and legal base which is needed, and the compatibility between existing and new institutional and political cultures. It has implications for political and governance arrangements and the way tools are selected, designed and used.

Political culture, formal structures and procedures, and the organisation, representation and practices of key political actors interact, giving rise to political relations of different types with different outcomes. Generally, democracy at the national level has been accompanied by democratic decentralisation. Even where, as in India, political democracy has been sustained at national level, democratic systems at city level were often suspended for long periods in the 1980s and only in the 1990s was the local democratic space widened and safeguarded. On the face of it, therefore, conditions for urban governance systems which incorporate a wider range of actors than governments and provide for means of ensuring representative involvement in decision making and accountability over and above periodic elections have been auspicious since the early 1990s. In practice, the picture has been mixed, as illustrated by the comparative research on ten cities on which this paper has drawn (Devas et al 2001).

At one extreme are authoritarian regimes, in which coercive or command power is exercised by an authoritarian leader (or oligarchy) with or without the trappings of democracy. Decision making is firmly in the hands of the leader or small elite, to whom the bureaucracy is subordinate. The interests of groups not critical to maintenance of control, especially the poor, tend to be addressed periodically and unsystematically, if at all. However, even to authoritarian regimes, the appearance of legitimacy is important, and windfall benefits to excluded groups may be used to secure their compliance, if not their active support, as in Santiago and Cebu before the return to democracy (Dockemdorff et al. 2000; Etemadi 2000).

Many ostensibly democratic regimes are in practice highly clientelistic. Decisions are made informally and the allocation of resources depends on the balance of power between patrons and clients. Politicians, bureaucrats and residents are all locked into personalised relationships as both patrons and clients. Thus in a system in which municipal capacity is lacking, resources inadequate and bureaucratic processes lack efficacy, it is important for residents and businesses to cultivate not only political but also bureaucratic relationship. Characterised in Bangalore as a ‘porous bureaucracy’, the informal exchange relations between clients at all levels may provide some of the poor with channels for obtaining access to services as personal or group favours (Benjamin 2000). The negotiations involved require persistence and may involve the payment of ‘tips’ (Kumasi) or bribes. As democratisation proceeds, the scope for group lobbying increases, and groups may be able to extract benefits in return for their explicit policy support, as in Bangalore or Cebu.

In inclusive consolidated electoral politics, the bureaucracy is directed by and accountable to political decision makers, channels exist for deliberative and collaborative decision making incorporating a range of actors including poor groups, and access to services is by entitlement. Cebu, Recife and Johannesburg show some of these characteristics (Devas et al. 2001).

Where multiple actors operate, a clear framework is needed for the allocation of powers, functions and resources and for the distribution of legislative and executive responsibilities. In practice, these are often unclear, contested and unpredictable. In addition, the supposed clear boundary between elected representatives undertaking political functions and bureaucrats providing them with technically well informed but neutral advice that is implied by traditional views of the desirable separation between public, political and private spheres and interests does not reflect reality, whether in authoritarian or democratic political systems.

The shift from government to governance, pleas that bureaucrats should be more politically aware and engaged to increase the responsiveness and efficacy of administrative systems, and the involvement of a wide range of actors in decision making and implementation raises difficult issues about which interests and practices are legitimate. However, these issues are barely recognised in current debates on urban governance and attempts to clarify them are in their infancy.

The multi-sectoral nature of urban management implies that a number of action programmes are needed, each likely to involve more than one agency. However, the ability of any one agency, even local government, to devise a coherent and binding strategic development framework is limited. There is, therefore, a need to simultaneously develop capacity within agencies and to improve collaboration between them. Neither is easy: modes of operation are entrenched in institutional culture and resistance to change widespread; rivalry, conflicting aims and priorities, and political competition may prevent conflicts being resolved and consensus reached. GIS are a potential tool for improving the information available to all governance actors and thus the information base for decision making. As they develop, they might provide additional scope for both more informed decision making (through predictive or prescriptive modelling and visualisation) and more inclusive and interactive participation in policy formulation, planning and project design, but plans for their use must be financially, institutionally and politically realistic.

References

• Abers, R. (2001). Learning democratic practice: distributing government resources through popular participation in Porto Alegre, Brazil. The Challenge of Urban Government: Policies and Practices. Stren, R. Washington DC, World Bank Institute: 129-44.
• Abrahamsen, R. (2000). Disciplining Democracy: Development Discourse and Good Governance in Africa. London, Zed.
• Acharya, S. and L. Thomas (1999). Finding a Pathway: Understanding the World and Performance of NGOs in Ahmedabad, India. Oxford, INTRACT Occ Paper 22.
• Adamolekun, L. (1999). Governance context and reorientation of government. Public Adminstration in Africa. Adamolekun, L. Boulder CO, Westview: 3-16.
• Alexander, J. (1997). "The local state in post-war Mozambique: political practice and ideas about authority." Africa 67(1): 1-26.
• Allen, C. (1995). "Understanding African politics." Review of African Political Economy 65: 301-20.
• Attahi, K. (1997). Decentralisation and urban participatory governance in Francophone Africa. Governing Africa's Cities. Swilling, M. Johannesburg, Witwatersrand University Press: 161-209.
• Attahi, K. (1999). Metropolitan governance in Abidjan. Seminar on Development and Urban Africa, Centre d'Estudis Africans, Barcelona.
• Bartone, C. (2001). The role of the private sector in municipal solid waste service delivery in developing countries: keys to success. The Challenge of Urban Government: Policies and Practices. Stren, R. Washington DC, World Bank Institute: 215-23.
• Batley, R. (1997). "Social agency versus global determinism in Latin American urban development." Third World Planning Review 19(4): 333-46.
• Batley, R. (2001). Public-private partnerships for urban services. The Challenge of Urban Government: Policies and Practices. Stren, R. Washington DC, World Bank Institute: 199-214.
• Batty, M. (1997). Digital planning: preparing for a fully wired world. Computers in Urban Planning and Urban Management, Bombay, Narosa.
• Batty, M. and P. J. Densham (1996). Decision support, GIS, and urban planning. London, Centre for Advanced Spatial Analysis, University College London.
• Beall, J. (1996). Urban Governance: Why Gender Matters. New York, UNDP, Gender in Development Monog Series 1.
• Beall, J., O. Crankshaw and S. Parnell (2001). Towards Inclusive Urban Governance in Johannesburg. Birmingham, International Development Department, School of Public Policy, University of Birmingham, Urban Poverty and Governance WP 24.
• Benjamin, S. (2000). "Governance, economic settings and poverty in Bangalore." Environment and Urbanization 12(1): 35-56.
• Benjamin, S. and N. M. Bengani (1998). The civic politics of industrial districts in India. Decentralized Production in India: Industrial Districts, Flexible Specialization and Employment. Holmstrom, M. New Delhi, Sage: 376-92.
• Bennett, E., P. Grohmann and B. Gentry (1999). Public-Private Partnerships for the Urban Environment: Options and Issues. New York, UNDP Public-Private Partnerships for the Urban Environment and Yale University PPUE WP Series Vol. 1.
• Bird, R. M. (2001). Setting the stage: municipal and intergovernmental finance. The Challenge of Urban Government: Policies and Practices. Stren, R. Washington DC, World Bank Institute: 113-28.
• Bird, R. M. (2001). User charges in local government finance. The Challenge of Urban Government: Policies and Practices. Stren, R. Washington DC, World Bank Institute: 171-82.
• Bishop, I. D., F. J. Escobar, S. Karuppannan, I. P. Williamson, P. M. Yates, K. Suwarnat and H. W. Taqub (2000). "Spatial data infrastructures for cities in developing countries. Lessons from the Bangkok experiences." Cities 17(2): 85-96.
• Blair, H. (2000). "Participation and accountability at the periphery: democratic local governance in six countries." World Development 28(1): 21-39.
• Burgess, R., M. Carmona and T. Kolstee (1997). Contemporary spatial strategies and urban policies in developing countries: a critical review. The Challenge of Sustainable Cities. Kolstee, T. London, Zed Books: 111-24.
• Campbell, T. (1997). Innovations and Risk Taking: The Engine of Reform in Local Government in Latin America and the Caribbean. Washington DC, World Bank DP 357.
• Chabal, P. and J.-P. Daloz (1999). Africa Works: Disorder as Political Instrument. London and Bloomington, James Currey and Indiana University Press.
• Chaplin, S. E. (1999). "Cities, sewers and poverty: India's politics and sanitation." Environment and Urbanization 11(1): 145-58.
• Clarke, P. and R. Buys (1999). Local Governance Study: Nicaragua. London, One World Action.
• Cotton, A., M. Sohail and W. K. Tayler (1998). Community Initiatives in Urban Infrastructure. Loughborough, Loughborough University, Water Engineering and Development Centre.
• Davey, K. (1996). The structure and functions of urban government. Urban Management: The Challenge of Growth. Pasteur, D. Aldershot, Avebury: 47-102.
• Desai, V. and I. Preston (2000). "Urban grassroots nongovernmental organizations in Bombay: a suggested typology." Government and Policy 18(4): 453-68.
• Devas, N., P. Amis, J. Beall, U. Grant, D. Mitlin, C. Rakodi and D. Satterthwaite (2001). Urban Governance and Poverty: Lessons from a Study of Ten Cities in the South. Birmingham, International Development Department, School of Public Policy, University of Birmingham.
• Devas, N. and D. Korboe (2000). "City governance and poverty: the case of Kumasi." Environment and Urbanization 11(2): 123-36.
• Devas, N. and C. Rakodi (1993). Planning and managing urban development. Managing Fast Growing Cities. Rakodi, C. Harlow, Longman: 41-62.
• Dockemdorff, E., A. Rodriguez and L. Winchester (2000). "Santiago de Chile: metropolitanization, globalization and inequity." Environment and Urbanization 12(1): 171-84.
• Douglass, M. (1998). World city formation on the Asia Pacific Rim: poverty, 'everyday' forms of civil society and environmental management. Cities for Citizens: Planning and the Rise of Civil Society in a Global Age. Friedmann, J. Chichester, Wiley: 107-38.
• Dutta, S. (2000). "Partnerships in urban development: a review of Ahmedabad's experience." Environment and Urbanization 12(1): 13-26.
• Ebel, R., D. and F. Vaillancourt (2001). Fiscal decentralization and financing local governments: framing the problem. The Challenge of Urban Government: Policies and Practices. Stren, R. Washington DC, World Bank Institute: 155-70.
• Etemadi, F. U. (2000). "Civil society participation in city governance in Cebu City." Environment and Urbanization 12(1): 57-72.
• Etemadi, F. U. (2001). Towards Inclusive Urban Governance in Cebu. Birmingham, International Development Department, School of Public Policy, The University of Birmingham, Urban Governance, Partnership and Poverty WP 25.
• Farvacque, C. and P. McAuslan (1992). Reforming Urban Land Policies and Institutions in Developing Countries. Washington DC, World Bank, UNCHS and UNDP, Urban Management Program Policy Paper 5.
• Fazal, S. (2000). "Urban expansion and loss of agricultural land - a GIS based study of Saharanpur City, India." Environment and Urbanization 12(2): 133-50.
• Franceys, R. and K. Sansom (1999). India: Urban Water Supply. Birmingham, University of Birmingham, School of Public Policy, International Development Department, The Role of Government in Adjusting Economies Paper 35.
• Friedmann, J. (1998). The new political economy of planning: the rise of civil society. Cities for Citizens: Planning and the Rise of Civil Society in a Global Age. Friedmann, J. Chichester, Wiley: 19-35.
• Gulyani, S. (2001). The demand-side approach to planning water supply. The Challenge of Urban Government: Policies and Practices. Stren, R. Washington DC, World Bank Institute: 183-92.
• Haque, M. S. (1997). "Local governance in developing nations: reexamining the question of accountability." Regional Development Dialogue 18(2): iii-xxiii.
• Healey, P. (1997). Collaborative Planning: Shaping Places in Fragmented Societies. Basingstoke, Macmillan.
• Held, D. (1996). Models of Democracy. Cambridge, Polity Press.
• Hewison, K., G. Rodan and R. Robison (1993). Introduction: changing forms of state power in Southeast Asia. Southeast Asia in the 1990s: Authoritarianism, Democracy and Capitalism. Rodan, G. St Leonards NSW, Allen and Unwin: 2-8.
• Hill, T. R. and D. McConnachie (2001). "The role of geographical information systems in developing planning in South Africa." Third World Planning Review 23(3): 289-300.
• Hodder-Williams, R. (1999). The bases of political systems which are both democratic and African. SCUSA Colloquium, University of East Anglia 5-8 Sept.
• Hyden, G. e. a. e. (2000). African Perspectives on Governance. Trenton, Africa World Press Inc.
• Jalal, J. (2000). Decentralization in urban governance in India: a tale of two cities. Sustainable Cities,
• Sustainable Development. Zetter, R. Oxford, Oxford Brookes University, Proceedings of the 17th Inter-Schools Conference 3-4 April: 181-95.
• Kasturirangan, K. and D. P. Rao (1997). Urban planning and management: space technology options. Computers in Urban Planning and Urban Management, Bombay, Narosa.
• Kessides, C. (1993). Institutional Options for the Provision of Infrastructure. Washington DC, World Bank, DP 212.
• King, R., D. Inkoom and K. M. Abrampah (2001). Urban Governance in Kumasi: Poverty and Exclusion. Birmingham, International Development Department, School of Public Policy, University of Birmingham, Urban Poverty and Governance WP 23.
• Kombe, W. J. (2001). "Institutionalising the concept of environmental planning and management (EPM): successes and challenges in Dar es Salaam." Development in Practice 11(2/3): 190-207.
• Korff, R. (1986). "Who has power in Bangkok? An approach towards the analysis of strategic group and class formation in an Asian primate city." International Journal of Urban and Regional Research 10(3): 330-56.
• Lee Lik Meng and Tan TS (2002). "GIS for plan-making in Penang Island: the road to online planning." Online Planning Journal.
• Mamdani, M. (1996). Citizen and Subject: Contemporary Africa and the Legacy of Late Colonialism. Princeton NY, Princeton University Press.
• McCarney, P. L., M. Halfani and A. Rodriguez (1995). Towards an understanding of governance. Perspectives on the City. Bell, J. K. Toronto, University of Toronto, Centre for Urban and Community Studies: 91-141.
• Mengers, H. (1997). Urban Development in the State Karnataka, India: Policies, Actors and Outcomes. Saarbrucken, Verlag fur Emntwicklungspolitik Saarbrucken GmbH, Nijmegen Studies in Development and Cultural Change Vol. 27b.
• Miller, D. (1993). Deliberative democracy and social choice. Prospects for Democracy: North, South, East, West. Held, D. Cambridge, Polity Press: 74-92.
• Moudon, A. V. and M. Hubner (2000). Introduction. Monitoring Land Supply with Geographic Information Systems. Hubner, M. New York, Wiley: 1-13.
• Nickson, A. (1995). Local Government in Latin America. Boulder, CO, Lynne Rienner. Nickson, A. (1997). "The public-private mix in urban water supply." International Review of Administrative Science 63(2): 165-86.
• Nickson, A. and R. Franceys (2001). "Tapping the market: can private enterprise supply water to the poor?" Insights 37.
• Olowu, D. (1999). Accountability and transparency. Public Administration in Africa. Adamolekun, L. Boulder, CO, Westview: 139-58.
• Parekh, B. (1993). The cultural particularity of liberal democracy. Prospects for Democracy: North, South, East, West. Held, D. Cambridge, Polity Press: 156-75.
• Pasteur, D. (1996). Internal organisation and management process. Urban Management: The Challenge of Urban Growth. Davey, K., with Batley, R., Devas, N., Norris,.M. and Pasteur, D. Avebury, Ashgate: 103-60.
• Paul, S. and S. Sekhar (1997). "A report card on public services: a comparative analysis of five cities in India." Regional Development Dialogue 18(2): 119-32.
• Paulsson, B. (1992). Urban Applications of Satellite Remote Sensing and GIS Analysis. Washington DC, World Bank/UNCHS/UNDP Urban Management Programme DP 9.
• Porio, E. (1997). State, civil society and urban governance in Southeast Asia. Urban Governance and Poverty Alleviation in Southeast Asia: Trends and Prospects. E. Porio et al. Quezon City, GURI in SE Asian and Ateneo de Manila University,: 1-39.
• Preston, P. W. (1998). Pacific Asia in the Global System: An Introduction. Oxford, Blackwell.
• Rakodi, C. (2000). Technical decisions or interactive processes? Towards collaborative planning and decision making in towns and cities. Urban Futures 2000, Witwatersrand University and Johannesburg Metro Council 10-14 July, Johannesburg.
• Rakodi, C. (2002). "Forget planning, put politics first? Priorities for urban management in developing countries." International Journal of Applied Earth Observation and Geoinformation 3: 209-23.
• Rodriguez, A. and L. Winchester (1996). The challenge for urban governance in Latin America: reinventing the government of cities. Cities and Governance: New Directions in Latin America, Asia and Africa. McCarney, P. L. Toronto, University of Toronto, Centre for Urban and Community Studies: 23-38.
• Ruland, J. (1992). Urban Development in SE Asia: Regional Cities and Local Government. Boulder. CO, Westview.
• Russell, S. and E. Vidler (2000). "The rise and fall of government-community partnerships for urban development: grassroots testimony from Colombo." Environment and Urbanization 12(1): 73-86.
• Sahley, C. and J. Danziger (1999). Rising to the Urban Challenge? The Roles, Strategies and Performance of NGOs in Lima, Peru. Oxford, INTRAC Occ Paper 25.
• Sklar, R. (1999). The significance of mixed government in Southern African studies. African Democracy in the Era of Globalisation. Hyslop, J. Johannesburg, Witwatersrand University Press: 115-21.
• Smith, B. (2000). "The concept of an 'enabling' local authority." Government and Policy 18(1): 78-94. Smith, D. M. and M. Blanc (1997). "Grass-roots democracy and participation: a new analytical and practical approach." Society and Space 15(3): 281-303.
• Souza, C. (2000). Participatory Budgeting in Brazilian Cities: Limits and Possibilities in Building Democratic Institutions. Birnmingham, International Development Department, School of Public Policy, The University of Birmingham, Urban Governance, Partnership and Poverty WP 28.
• Stoker, G. (1998). Public-private partnerships and urban governance. Partnerships in Urban Governance: European and American Experience. Pierre, J. Basingstoke, Macmillan: 34-51. Swilling, M. (1997). Introduction. Governing Africa's Cities. Swilling, M. Johannesburg, Witwatersrand University Press: 1-12.
• Thomson, C. N. and P. Hardin (2000). "Remote sensing/GIS integration to identify potential low-income housing sites." Cities 17(2): 97-110.
• Tripp, A. M. (1997). Changing the Rules: The Political Economy of Liberalization and the Urban Informal Economy in Tanzania. London and Los Angeles, University of California Press.
• United Nations Centre for Human Settlements (2000). Principles of Good Urban Governance, UNCHS Global Campaign on Urban Governance. 2002.
• United Nations Centre for Human Settlements (2001). Cities in a Globalizing World: Global Report on Human Settlements 2001. London, Earthscan.
• United Nations Development Programme (2000). The Urban Governance Initiative. Kuala Lumpur, UNDP. Ward, P. M. and S. Chant (1987). Community Leadership and Self-Help Housing. Oxford, Pergamon, Progress in Planning, 27, 2, 69-136.
• World Bank (2000). Cities in Transition: A Strategic View of Urban and Local Government Issues. Washington DC, World Bank, Infrastructure Group, Urban Development.
• Zérah, M. (2000). The Cancellation of the Pune Water Supply and Sewerage Project. New Delhi, Water and Sanitation Program - South Asia.