Resilience,transition or transformation? A comparative analysis of changing water governance systems in four southern cities

Abstract

Delegating state responsibilities for the management of water resources to regional bodies and the provision of drinking water and sanitation to local governments has led to new configurations in urban water governance. Drawing on case studies from four cities in the global South (Guarulhos, Arequipa, Lima and Durban), this article analyzes recent changes in these configurations, with particular attention to the role and power of the municipality in this process. This paper explores to what extent these new configurations reveal a move towards resilience, transition or even transformation. It concludes that there are clear indications of transition in all cases, and in Guarulhos and Durban even some signs of transformation. Given that transformative changes in the legal and institutional framework, and even in values and attitudes, have not yet affected the existing power structures, the question is to what extent these signs of transformation will reach their full potential.

I. Introduction

Globally, water governance practices are shifting from a hierarchical, state-led and sectoral approach to a more integrated and participatory approach. The rescaling of responsibilities has included increased participation by non-state actors in decision-making and a shift away from public ownership of water supply infrastructure, or strict regulation of its provision.⁽¹⁾ The provision of water and sanitation often became a municipal responsibility, guided more by the general impetus for decentralization than by specific challenges in the water sector.⁽²⁾ The general trend towards privatization of urban services has also meant varying degrees of ownership or control over different aspects of urban water provision and sanitation. Neither privatization nor increased municipal control has, as yet, resulted in access in all cities to water and sanitation services through a universalized network.⁽³⁾ In virtually all cities in the South, part of the population still relies on alternative water provision, either through water vendors, private wells, boreholes or by buying packaged water.

Water is an important input for agriculture, non-agrarian production processes, hydropower generation, ecosystem services and recreation, as well as household consumption. The Integrated Water Resource Management (IWRM) principles, which have significant global influence, stipulate that river basins or catchments are a more appropriate unit for managing water than the administrative boundaries of political systems, and can strike a balance between these competing uses. River basin management is a well-established concept,⁽⁴⁾ however, it focused previously on human use only. Its current conception has broadened the scope to include environmental concerns and the importance of non-consumptive uses of water. It is argued that the river basin or catchment presents water’s own natural boundaries, which is a better spatial unit for water management than administrative boundaries designed by sociopolitical systems. The growing popularity of “participation” and consequently the need to involve a wide array of actors in river basin management has also been part of the revival of the concept. There have been attempts to set up river basin organizations in many countries and the International Network of Basin Organizations (INBO) has regionalized branches. With the diversity of actors in river basin organizations, the river basin has become a wider political arena where contrasting interests and approaches to water can be debated and reconciled, potentially challenging conventional decision-making and expertise that are insufficient to tackle increasingly complex water-related problems.⁽⁵⁾

Drawing on four case studies⁽⁶⁾ – Guarulhos (Brazil), Arequipa and Lima (Peru) and Durban⁽⁷⁾ (South Africa) − this paper uses an analytical framework developed by Pelling⁽⁸⁾ to answer the following questions:

• What are the main consequences of recent changes in the institutional framework and rescaling of responsibilities for urban water governance, with specific attention to the mandates and roles of municipalities?

• To what extent do the new institutional structures and responsibilities foster social learning and knowledge generation in water governance?

• To what extent do the new arrangements and the social learning processes foster resilience, transition or even transformation?

II. Resilience – Transition – Transformation

Pelling’s classification of adaptation practices⁽⁹⁾ identifies three levels of adaptation: resilience, transition and transformation, distinguished primarily by the extent to which they challenge the status quo. This framework is useful for assessing changes in water governance, given that water is a primary medium through which climate change will affect daily existence and ecosystems.⁽¹⁰⁾ Water governance systems will have to adapt to a range of possible climate impacts, including increased flooding, heavy rainfall events or droughts.

The defining quality of resilience, in Pelling’s classification, is the desire to maintain functional integrity. “Adaptation as resilience” focuses on improving existing practices without questioning underlying assumptions,⁽¹¹⁾ and can hence allow unsustainable or socially unjust practices to continue. Short-term and partial remedies dominate because they best serve established value priorities.⁽¹²⁾

Transition involves incremental changes in governance systems, but with overarching norms, principles and sociopolitical regime remaining unchanged. Rules and decision-making are altered but norms and principles are not. Transition requires reflection on how problems are framed⁽¹³⁾ but can also result from renegotiation of policy priorities.⁽¹⁴⁾

Transformation requires a regime change, in which underlying values are questioned from the level of individual behaviour to the mechanisms and structures of the global political economy. Transformation addresses the root causes of vulnerability and deprivation, requiring an understanding of the environmental crisis in the context of our relation with the earth. This echoes Freire’s statement that “… we are not only ‘in’ the world but also ‘with’ the world, that is, we are essentially related to it.”⁽¹⁵⁾ Understanding the environmental crisis as an internal crisis and an expression of our distorted relation with the world requires a profound shift in our world view, in how we understand our relations with nature and the social world and our role in the political processes that shape our social-ecological relations. Transformation requires strong sustainable development goals, which can only be achieved through radical changes in the value system underpinning the dominant capitalist production system, including the public−private dichotomy that has favoured private actors and marginalized the role of the state.⁽¹⁶⁾

The main characteristics of the three modes of adaptation are summarized in Table 1.

Table 1

Modes of responding to external threats or disturbances

	Resilience	Transition	Transformation
Mode	Refining existing practice	Incremental change	Radical change, regime shift
Goal	Functional persistence in a changing environment	Realize full system potential through the exercising of rights in established regime structures	Reconfigure the structures of development
Scope	Change in technology, management practice or organization	Change in governance practices as rights are exercised; incremental change at individual policy level or in specific geographical area	Reform in overarching political economy, cultural norms or scientific paradigm; local accountability of global capital
Policy focus	“Disaster-resilient” building practice	Implementation of legal responsibilities by public and private actors; exercising of rights by citizens; change in rules, not norms and principles	New political discourses
Social learning	Directed at improving existing practices and technologies	Questioning of existing structures; shifting balance between stability and reform	Connecting individual to social learning, which leads to changes in norms, principles and value systems

Source: Adapted from Pelling, M (2010), Adaptation to Climate Change: From Resilience to Transformation, Routledge, London, 224 pages and Pelling, M (2012), “Resilience and transformation”, in M Pelling, D Manuel-Navarrete and M Redclift, Climate Change and the Crisis of Capitalism: A Chance to Reclaim Self, Society and Nature, Routledge, pages 51−65.

a. Social learning: an essential component of adaptation

Pelling considers social learning as essential to all three forms of adaptation, but with a different focus and outcome in each case.⁽¹⁷⁾ In resilience, it is directed at improving existing practices, for instance through technological innovation or changes in institutional practice, whereas in transition and transformation it fosters changes in organizational frameworks and institutions. The necessary conditions for change are trust, a willingness to take risks and experiment, and an active engagement with civil society. Where strong levels of organization cause institutional inertia, change is impeded. Where emergent social structures foster experimentation and innovation, transition or even transformation can be achieved. Pelling argues that critical reasoning, a cornerstone of deliberative democracy, is essential for transformation.⁽¹⁸⁾ Critical reasoning inspired by Freire’s political pedagogy⁽¹⁹⁾ supports dialogue between the marginalized and those in power. Transparency is also needed to challenge embedded values.

There are considerable knowledge deficits around the complexity of social-ecological systems and the non-linear interactions between processes at multiple scales. The need to face this complexity and uncertainty with incomplete knowledge⁽²⁰⁾ gave rise to the adaptive management paradigm, an approach in which learning is expected to take place among scientists, resource managers and policy makers through deliberate experimentation, constant monitoring, joint action and reflective practice, presumably leading to improved decision-making.⁽²¹⁾

However, social-ecological systems theory has been criticized for being “power blind” and a-political,⁽²²⁾ neglecting power differences between actors taking part in the decision-making process. Originally, the learning was mainly restricted to the professional community, but with the growing importance of environmental movements and indigenous communities in resource management, it has started to incorporate “local knowledge” or “lay knowledge”. It is important to distinguish between practices that are restricted to professionals and those where a wider variety of actors are incorporated, opening up social learning to a multiplicity of world views and enabling values to be questioned. The growing influence of a rights-based approach to development and the call for greater participation in resource management led to the development of collaborative management.⁽²³⁾ The inclusion of a wider variety of actors in resource management implied an imperative for learning; to participate meaningfully in decision-making, participants need to be knowledgeable. Social learning within this context expands on the learning within adaptive management, depending on deliberation and sustained interaction between individuals, sharing knowledge, perspectives and values in a trusting environment.⁽²⁴⁾ However, it is important to recognize that power and knowledge asymmetries, as well as practical concerns such as time constraints and variability in attendance at meetings, may seriously affect the outcomes of such processes.⁽²⁵⁾ Adaptive and collaborative management are also distinguished by the relationship between learning and action. In adaptive management, joint action leads to reflection and learning, which improves the next joint action. In collaborative management, learning through deliberation implies that actors are exposed to differing values, have to analyze and debate alternatives and create a common ground for collective action.⁽²⁶⁾

III. Water Governance: Different Approaches and their Associated Discourses

Water governance, or the range of systems for developing and managing water resources and services, is much broader than just water management.⁽²⁷⁾ Actors at multiple levels – including government, business, political parties, civil society, international agencies – are involved in confrontations and debates around how river basins and water services should be governed.⁽²⁸⁾ We use the term “water governance” to refer to the multi-actor processes through which objectives are set (hence the political and often conflictual processes), and the term “water management” for the administration, implementation and functioning of water infrastructure.

Differences in power or conflicts of interest can underlie struggles over water, but Castro argues that it is also about how actors value it.⁽²⁹⁾ The most dominant discourses in water governance debates are around “water as an economic good/commodity” and “water as a common good and human right”. Other approaches, namely “water as a social-ecological good” and “water as a technical sector” are also important (Table 2).

Table 2

Four approaches to water

	Water as a sector	Water as an economic good/commodity	Water as a socio-ecological good	Water as a human right/social good
Development discourse	Sectoral (mainly pro-large water infrastructure)	Pro-growth	Pro-green/pro-life	Pro-poor
Main concern	Optimal, efficient (economic and social) use of the resource	The market	The environment; human and other beings	Human beings
Management principles	Predict, command and control	Market principles	Adaptive management	Collaborative management
Main actors	Water functionaries (public and private water boards)	Private sector; national and sub-national governments; water boards	Civil society organization–environmental networks; peasant and indigenous communities; local and sub-national governments	Local and sub-national governments; civil society organization networks (development practitioners); communities
Territorial scale	Dictated by mandate for city, region or wider area	Cities; companies; households	Regional, (multiple) river basins; aquatic ecosystem	Cities and neighbourhoods
Main democracy approach	Representative	Representative or authoritarian	Participatory and/or deliberative democracy	Deliberative democracy

Source: Based on Miranda Sara, L, M A Hordijk and R K Torres Molina (2011), “Water governance key approaches − an analytical framework”, Chance2Sustain Working Paper No 4, EADI, Bonn, page 13; also Munnik, V and J Burt (2013), “Overview of need and potential for social research in WRC and the SA Water Sector. Towards developing a social science research agenda for the South African water sector”, K8/1024/1, Water Research Commission, South Africa, page 13.

Since the 1980s, “water as an economic good” has become the dominant approach in water governance and management, treating water as a “resource” with an economic value. According to this approach, water should be priced, water delivery should be open to the market, and for hard-liners water sources should be privately owned.⁽³⁰⁾ Neo-liberal principles of the capitalist global economy dominate. Cost-recovery is a central concern in urban water provision and the ability to pay has to be factored into services.

This market-based approach stands in stark contrast to the “water as a human right” approach. In 2010, after decades of debate, clean drinking water and sanitation were declared essential human rights,⁽³¹⁾ and universal access to sufficient water for basic needs is considered a non-negotiable priority, with governments and other actors accountable for ensuring its availability and affordability. The state is required to generate policies that guarantee access via subsidies or other instruments and puts provision under democratic control and state regulation. Proponents of this approach, often from a “pro-poor”, “development” or “brown agenda” background, support a preference for collaborative management principles, in which processes of deliberative decision-making in governance networks ensure that the voices of a wide variety of actors, including the marginalized, are heard.

The “water as a social-ecological good” approach supports the position that water is essential to all living beings, and it extends the right to water to nature as a whole, with an emphasis on the finite nature of water and a concern for future generations and ecosystem health. Key proponents are associated with the green (or “pro-life”) agenda and are sometimes called “anti-development”. The related discourse has gradually become embedded in public policy in the South, particularly in Latin America,⁽³²⁾ and adheres to the principles of adaptive management, in which experimentation, monitoring and reflection play a major role.

The last approach is the traditional “water as a technical sector”, and includes a strong pragmatic emphasis on water management, infrastructure and provision, as well as concerns for the productive uses of water based on “predict, command and control”, where solutions are sought in infrastructure developments. The financing of infrastructure investments and the efficient management of water and sanitary services is thus a central concern. This approach is confined to water management, with decisions taken by managers and engineers, often based on policy goals set by other actors.

Social learning and knowledge are perceived differently within these approaches.⁽³³⁾ Both the technical and economic good approaches premise a level of certainty and predictability, implying that “command and control interventions” based on sound science can engineer solutions to water problems. Knowledge comes from a technocratic/scientific rationale, based on quantitative indicators of either physical/natural and technical conditions (water engineers) or financial/economic efficiencies.⁽³⁴⁾ Learning, conceived mainly as an individual process attained through formal education or inputs from science, is limited to professionals and directed at “how to engineer the environment out there”. As Baud et al. have noted, this is very much in line with Tovey and Bruckmeier’s “elitist model” of knowledge-building.⁽³⁵⁾

Within the “water as a human right” and “water as a social-ecological good” approaches, social learning is more participatory and deliberative, with multiple views and values contested and debated to achieve some level of social and ecological justice. Actors supporting these approaches construct their knowledge from a wide range of philosophical, theoretical and moral perspectives, resulting in multiple “truths” and a reflexive process. The questioning of underlying values, which Pelling argues is necessary for transformation, occurs through these more deliberative processes. Within sectoral or economic good approaches, existing practices are merely refined, or incremental changes made without altering underlying structures.

IV. Shifts in Water Governance in the Case Study Cities

a. The city contexts

The four case study cities differ considerably in population size, economic base and geographical characteristics (Table 3), but they share some important characteristics in terms of water governance, which allow for comparisons in relation to Pelling’s three categories of adaptation. Although all but Durban are in water-abundant countries, they all face water scarcity. Water resources are highly unevenly distributed within Peru and Brazil, and the management and distribution of water resources in South Africa is problematic due to a lack of bulk infrastructure. Water governance in Guarulhos is constrained by a lack of sufficient water in the São Paulo Metropolitan Region;⁽³⁶⁾ Lima is situated in a desert area with less than nine millimetres of rainfall per annum; and the Andean highlands around Arequipa are also dry. Durban is located in the wettest region of a water-scarce country, but South Africa’s apartheid legacy and the lack of planning have hampered the development of the bulk infrastructure required in a fast-growing city.

Table 3

Main characteristics of the case study cities

	Guarulhos	Metropolitan Lima	Arequipa (metropolitan area)	Durban
Population size	1.22 million	8.5 million	850,000	3.6 million
Characteristics	City within Metropolitan São Paulo; São Paulo’s main airport; second largest GDP in the state of São Paulo	Capital; produces 45 % of Peruvian GDP; mainly services	Secondary city; produces 4% of GDP; services, trade, agriculture, mining and regional centre; strategic position in the south of Peru	Third largest city in South Africa; 9% of total economic output; largest port in southern Africa; manufacturing, trade, transport and services
Geography and climate	Sub-tropical; humid; situated on a high plateau; annual rainfall 1,489 mm	Desert; second driest city in the world; annual rainfall 9 mm	Andean city (2,330 metres above sea level); surrounded by volcanoes; dry climate; annual rainfall 93 mm	Coastal city; dry winters; humid, wet summers; annual rainfall 1,009 mm
Percentage of population with safe water supply	95% (connected to official network)	93% (connected to official network)	90% (connected to official network)	91% (within 200 metres of the house)
Percentage of population connected to sewerage	88%	84%	95%	77%
Main water-related climate change threats	Increasing intensity and frequency of intense weather events, especially rainfall, causing flooding and mudslides; critical flooding in 2011; heat islands	High uncertainty − might be more intense rainfall and tropicalization, prolonged droughts or a combination of both; rapid glacier melt threatening low-income neighbourhoods, highways, bridges, the port and the airport; heat islands	High uncertainty − might be more intense rainfall and tropicalization, prolonged droughts or a combination of both; critical pluvial flooding in 2012 and 2013; affected by very rapid glacier melt	Increasing intensity and frequency of intense weather events, both heavy rains and prolonged droughts; wetter winters and wetter summers; sea level rise
New water law adopted	1997	2009	2009	1997 and 1998
Annual per capita budget	US$ 1,127 (2011)	US$ 95 (2010); Lima generated 49% of its own budget	US$ 285 (2010); Arequipa generated 4% of its own budget	US$ 1,019 (2011); Durban generated 91% of its budget

SOURCE: Hordijk, M, L Miranda Sara, C Sutherland, J Sydenstricker-Neto, A J Noles and A Gomes Rodrigues Batata (2013), “Water governance in times of uncertainty, complexity, fragmentation, innovation”, page 7, Chance2Sustain fieldwork report series available at www.chance2sustain.eu.

Internationally driven discourses and reforms, a shift towards democracy and pressure on scarce water resources have shaped the development of new water governance structures in all four cities. All three countries have experienced a significant change in water governance structures either after a return to democracy (Brazil and Peru) or the establishment of democracy (South Africa), which resulted in a decentralization of mandates and responsibilities in water governance and management and an increased emphasis on participation. These changes in water governance, and the extent to which they reflect resilience, transition and transformation, are explored in all four cities below (Table 4).

Table 4

Water governance reforms in Guarulhos, Lima, Arequipa and Durban

	Guarulhos	Lima	Arequipa	Durban
Approach to water – legal framework	Water is a public good with economic value and a limited natural resource (Water Law 1997)	Water is a public good with socio-cultural, economic and environmental value; drinking water provision has priority, even in times of scarcity (Law on Water Resources 2009)		Water is a human right (Constitution 1996); water as an economic good; water as a socio-ecological good (National Water Act 1998)
Management approach according to legal framework	Participatory management; IWRM principles	Participatory management; IWRM principles; participation of users associations but not of civil society organizations		Participatory management; IWRM principles
Management principles in practice	Participatory, deliberative but with dominance of state and sectoral actors	Inter-river basin council being set up; still centralized management	Concertative, but with dominance of private actors	Adaptive management, with managerial approach to participation
Rescaling of responsibilities	Drinking water and sanitation to municipal level; water resource management to river basin level	Drinking water and sanitation under responsibility of central government agencies; decision on water allocation final responsibility of national government; inter-river basin council set up but not yet operational	Drinking water and sanitation to provincial municipalities; decision on water allocation final responsibility of national government; river basin council set up but not yet fully operational	Water resource management at the national level; integrated catchment management at the provincial level because CMAs have not yet been formed; water and sanitation provision at the municipal level
Municipal autonomy in water governance	De jure medium, de facto low; under pressure from basin council; limited resources to finance infrastructure	Low; no influence on public water company; no control over major investments; fragmented institutional landscape	De jure medium, de facto low; influence on public water company but dependent on mining company for financing infrastructure	High; strong leadership in water provision and climate change adaptation

b. Shifts in water governance arrangements: decentralization and diversification of actors

i. Guarulhos, Brazil

Decentralization, participation and use of the water basin as a unit for water governance are most strongly evident in Brazil. The provision of water and sanitation was decentralized as early as 1967, and municipalities could opt for public, private or para-statal provision. Guarulhos opted to establish its own municipal water and sanitation company (SAEE); however, they buy bulk water from the state level water and sanitation company SABESP, which provides 56 per cent of São Paolo State municipalities with water, and has strong control over prices for raw water. Both SAEE and SABESP approach water mainly as an economic good, and to a lesser extent as a sector. The new Brazilian Water Law of 1997 however characterizes “… water as a public good with an economic value and a limited natural resource”,⁽³⁷⁾ which should be managed in a participatory manner. This initiated a participatory system of water basin management, implemented from the national to the sub-basin level. The National Council on Water Resources⁽³⁸⁾ sets the parameters for 180 basin committees. These committees have representatives from municipalities, civil society organizations and the state. The presidency of the basin committee is always held by one of the mayors, whereas the executive secretary is always a state organization representative. The basin committees form a platform to address issues of common interest, informed by members’ positions, backgrounds and approaches to water. Environmental NGOs that view water as a socio-environmental good can find themselves at odds with the Department for Water and Hydropower (DAEE), which has a sectoral approach. However, the formation of alliances is also possible. Until 2010, none of the sewage in Guarulhos was treated, leading to serious pollution of the Tiête River and problems for downstream municipalities. Combined pressure from the Alto Tiête Basin Committee and DAEE forced the Municipality of Guarulhos to invest in sewage treatment, and by the end of 2012 it treated 35 per cent of its sewage. The Alto Tietê Basin Committee also worked on the formulation of a new state headwaters protection law, which enables committees and sub-committees to create specific laws for areas where springs are located. The basin committees have also developed a charging system for water use, to compensate for ecosystem services provided by municipalities in protected areas and to create revenue for the committee. In the past, these kinds of decisions were taken by state or national actors alone. However, major investment decisions are still taken outside the committees. State government, for instance, constructed a megaproject − the Tiete Linear Park, with significant implications for the riverbanks in Guarulhos − without municipal actors being consulted or even informed. Guarulhos was not consulted regarding the dyke built on the São Paulo side of the Tiete riverbanks, which increases flood risks for inhabitants in the Guarulhos flood plains.⁽³⁹⁾

ii. Lima and Arequipa − Peru

The decentralization of water governance in Peru is similar to the Brazilian model, but with some significant differences. The new water law was only adopted in 2009, 12 years later than in Brazil. The Peruvian law states that water is a public good, the property of the state⁽⁴⁰⁾ and should be managed in an integrated and participatory manner. The National Water Authority (ANA), which now falls directly under the Ministry of Agriculture,⁽⁴¹⁾ also has responsibility for ensuring that climate change adaptation plans are developed at local, regional and national level.⁽⁴²⁾ Although local water authorities (Autoridades Locales de Agua, ALAs) are responsible for advising on the administration and regulation of water quantity and quality at basin level, final decisions are taken by the ANA. The new water law has also tightened control of previously more autonomous user associations.⁽⁴³⁾ In parallel to the ALAs, river basin councils have been formed. But, whereas in Brazil all organized civil society can participate, in Peru participation is limited to water users. Peasant and indigenous communities can partake as user associations but other community-based organizations cannot. According to the law, regional authorities, local authorities and water users only participate and monitor in coordination with the ANA, and they have substantially less influence than in Brazil. As in Brazil, major infrastructure decisions are taken outside the councils, often without any consultation with local stakeholders.

The new reforms work out differently in Lima and Arequipa. Lima’s territory covers three different basins; a process is underway to set up a council covering all three but there has been little success so far. Metropolitan Lima consists of two provincial municipalities, Lima⁽⁴⁴⁾ and Callao, with 43 and nine district municipalities, respectively, each with its own urban planning, environmental issues and disaster management departments. Coordination is weak. The public water company SEDAPAL provides water and sanitation to both Lima and Callao,⁽⁴⁵⁾ and is the only water company in the country operating under the Ministry of Housing and Construction and the National Fund for Financing Economic Activities of the State, FONAFE, rather than the provincial municipality. The ministry has refused to have representatives of the municipalities of Lima and Callao on its board, and responds more to the demands of large real estate investors and small-scale land invaders than to the needs and demands of local governments. Metropolitan Lima has very little influence, then, on how water governance is shaped in its territory.

Arequipa has a relatively well-functioning, effective, informal, participatory water governance system that has operated for three decades, with “participation” limited to water users and the main water infrastructure managers. In 1983, after a serious drought, the principal water users (the public electricity company EGASA, the mining company Cerro Verde and the users association of the large-scale farmers) created a multi-sectoral committee that included a number of public actors: the water and sanitation company SEDAPAR;⁽⁴⁶⁾ the authority overseeing the large-scale irrigation project Majes−Siguas; AUTODEMA (which managed dams);⁽⁴⁷⁾ and others. Despite this public influence, the committee remained dominated by the large-scale users. Major decisions on infrastructure investments were taken by this informal body, with the agreement that the principle partners (the mining company, the electricity company and the large-scale farmers associations) would finance a share relative to their economic power. In the first decades of its existence, they mainly decided to construct dams. But since 2004, the mining company in particular has financed infrastructure for urban water provision, extending the network to poorer more peripheral areas and constructing a new water plant. Over the last decade, the mining company has invested more in new city water infrastructure than the water company has. The river basin council created as a result of the new law consists mainly of the same actors, although the water company SEDAPAR and the electricity company EGASA have chosen to be represented by the mining company, which is the major player in economic terms. As a result, the dominant approach frames water as an economic good and considers it to be an infinite resource. No consequences are anticipated from climate change on the premise that new infrastructure can always channel water from elsewhere. The Municipality of Arequipa has a stronger influence on its public water company than Metropolitan Lima, but has made itself dependent on the mining company for infrastructure provision.⁽⁴⁸⁾

iii. Durban – South Africa

Significant post-apartheid legislative and policy reforms provided opportunities for change in water governance arrangements in South Africa. The right to sufficient water was enshrined in the new Constitution and the ANC government rapidly developed policies to address racial inequalities and backlogs in service delivery, simultaneously decentralizing service delivery to local government level. The Water Services Act (1997) and the National Water Act (1998), which reflect Integrated Water Resource Management (IWRM) principles,⁽⁴⁹⁾ explicitly highlight both sustainability and equity as overarching principles; also, a National Water Policy (2001) stipulates that all poor South Africans are entitled to a free basic water supply, a right that would be fulfilled by the ability of each local government to supply free basic water.⁽⁵⁰⁾ Water in South Africa is therefore recognized as a social good, supporting the country’s pro-poor agenda. Cost-recovery and efficient provision and management, however, are also principles of the National Water Act, raising concerns as to the tension between cost-recovery and social and environmental justice.⁽⁵¹⁾

Participation is a cornerstone of the National Water Act, which promotes the establishment of catchment management agencies that are designed to be representative and include communities in decision-making.⁽⁵²⁾ However, to date, few of these agencies have been established and none in KwaZulu-Natal (where Durban is located). Here, the provincial branch of the national Department of Water Affairs oversees responsibilities. In 2013, the uMgeni Ecological Infrastructure Partnership (UEIP) was formed, with a range of actors led by eThekwini Municipality’s Water and Sanitation Unit (EWS) and Environmental Planning and Climate Protection Department (EPCPD). It aims for an integrated approach to climate and water governance, and management at the regional scale, focused on ecological infrastructure enhancement and regional climate change adaptation; hence, using the approach of “water as a social-ecological good”.

Following the British model, South African local governments, including eThekwini Municipality,⁽⁵³⁾ have a relatively high level of autonomy and revenue-generating competency.⁽⁵⁴⁾ As a result of strong leadership and capacity, and in the absence of a functioning catchment management agency, eThekwini’s EWS has played a leading role in shaping water governance both in Durban and in the country as a whole, taking full advantage of this autonomy. Durban was the first South African city to supply free basic water, which then prompted the national government to draft the national free basic water policy. EWS translated its national legal obligation to provide all households with a free basic supply – a recommended 6,000 litres per month per household − into 9,000 litres per indigent household per month.⁽⁵⁵⁾ This differentiated service provision approach has substantially reduced backlogs and almost universal coverage of basic water supply has been achieved. Sanitation provision has been more challenging, but EWS continues to adopt an innovative, research-driven and adaptive approach in dealing with these challenges. Waterborne sewerage and full-pressure water system connections to all households are both financially and environmentally unsustainable and have not been extended to peripheral areas,⁽⁵⁶⁾ provoking criticism about perpetuating existing inequalities, as most households outside the urban core are poor.⁽⁵⁷⁾ EWS has developed a wide range of technical participatory processes that enable the water and sanitation entity within local government to engage with citizens and other actors around these challenges and issues.

c. Participation, learning and knowledge-building in water governance in Guarulhos, Lima, Arequipa and Durban

Participation is a cornerstone of water governance in all three countries. The new legal institutional framework for Brazil and Peru stipulates the creation of participatory and concertative⁽⁵⁸⁾ bodies at river basin scale, and the catchment management agencies in South Africa are designed to offer channels for participation.

If conducted in a managerial or technical manner, participation tends to lead to resilience, maintaining the status quo, rather than transition or transformation. Transition can be fostered through “learning-by-doing”, with experimentation and constant monitoring as practiced in adaptive management. Most often, this does not address power relations nor put underlying value systems under scrutiny. When participatory processes support deliberation and social learning takes place, then possibilities for transformation begin to open up.

The Brazilian river basin councils offer an excellent opportunity to debate different approaches to water, to learn, and to integrate knowledge from different sources. The diversity of actors in these councils did indeed encourage deliberative processes, leading to a diversification of approaches. Actors explicitly acknowledged that water is more valued than in the past, its finite nature gradually being recognized. This change was attributed to environmental education in general, not to the council. Most respondents placed high value on participatory processes themselves, a considerable break from the authoritarian rule of the past. However, participatory knowledge production turned out to be difficult in practice, since technical knowledge still dominates discussions. Management and policy documents were based on codified knowledge provided by experts, and civil society organization representatives lacked the necessary capacities to engage meaningfully in this technical knowledge production. This still reflects an elitist model, implying that because of knowledge asymmetries, deliberation occurs among a limited set of actors. The councils’ functioning also suffers from the practical concerns mentioned previously, such as time constraints and variability in attendance.

In Lima, the mandated basin council, still in the process of being established, is not yet achieving the co-production of knowledge. However, two positive processes have run in parallel in recent years. In the first, initiated by a German−Peruvian research programme,⁽⁵⁹⁾ a small group of actors (including representatives from the water company SEDAPAL, NGO representatives, a representative from the Cities for Life Forum and German specialists) gathered to run scenario-building processes for the water sector. Initially, this was a relatively closed process, focusing on scientific and technical knowledge. It gradually broadened in terms of actors and the inclusion of practice-based professional knowledge,⁽⁶⁰⁾ and the dominant approaches to water as a sector and as a commodity were being counterbalanced by the discourse of environmental NGOs.⁽⁶¹⁾ Some of the results of this process were taken up by the Municipality of Metropolitan Lima once a new, more human rights-oriented local government came to power. The discourse became more pro-poor and a pro-green climate change adaptation strategy slowly gained more influence. Different scenarios have now been debated in a variety of settings, both closed and fully open to the public, in technical gatherings and in political debates. A group of professionals have built a shared knowledge base as an outcome of this process and a sound platform for implementation has been prepared through numerous discussions.

In Arequipa, the situation is very different. The knowledge base is fragmented and less powerful actors acknowledge that they protect their information. Representatives from the water authority realize this, and they are working on a knowledge system that aims to process reliable up-to-date data from different organizations as a base upon which decisions can be taken, and with open access. However, this initiative is very recent and has as yet not yielded results. For the time being, “participation” is mainly limited to the old power holders in the comité multisectorial, who take the major infrastructure decisions.

eThekwini Municipality presents a strong example of integrating information and learning at the local government scale. EWS chooses not to operate under one overarching policy, so as to allow maximum flexibility in decision-making and to foster learning-by-doing. The staff within EWS and in the research institutions that support them are encouraged to be innovative by the head of the unit, and operate within a culture of experimental governance. The dynamic nature of the practices undertaken by EWS in service delivery reflects the continual adaptation within the unit in response to experience. For example, EWS has revised its free basic water policy twice in the past six years, in response both to concerns raised through participatory forums about the basic water needs of poor households⁽⁶²⁾ and to cost-recovery concerns raised within the municipality. EWS was an early adopter of GIS (in 2000/2001) and information in this system comes from and is shared among the municipality’s different departments. The information available is trusted and provides a strong basis for action but is highly technical and codified.⁽⁶³⁾ “Lay knowledge” is incorporated insofar as it concerns complaints about malfunctioning infrastructure and the provision of services at the community level.⁽⁶⁴⁾ Due to its managerial and technical nature it cannot be classified as “participatory” knowledge production, as it does not reflect the co-production of knowledge between different actors.

V. Conclusions

All the case studies discussed in this paper show substantive reforms in water governance, aligned with and reflecting the changing political regimes at the national level. Some reforms can be considered a sign of “transformation”,⁽⁶⁵⁾ based on the radical shifts that have taken place, such as the change from a central, hierarchical organization of water governance to a decentralized and more participatory system. The shift to the basin as a territorial unit for integrated water management is also a paradigmatic change. In Brazil, the radical change from an authoritarian to a democratic regime was marked by a strong emphasis on participation and deliberation in general, hence a change in development goals towards a more inclusive society. This is reflected in the importance given to participation and deliberation in water governance. In South Africa, both the transformation of post-apartheid society and the focus on human rights in the Constitution strongly influenced water governance. In Peru, decentralization and concertación were emphasized during the return to democracy, but are less apparent in the new water governance framework. The changes here in the mode and goal of governance are more characteristic of “transition”, implying changes in governance structures within the existing regime structures. In Lima, despite demands for more decentralized water management, centralized decision-making continues although more deliberative processes are influencing recent thinking. In Arequipa, the new structures formalize pre-existing informal structures.

A narrower knowledge base, reflected in the sectoral and economic approach to water, can be overcome by the diversification of actors participating in water governance. In three cases (Guarulhos, Lima and Durban), a diversification in approaches adopted for water governance was identified, as was an increase in the legitimacy of “water as a social-ecological good”. In Guarulhos, there was a growing awareness of water’s value and finite nature. These changes can be taken as important shifts as a result of social learning. Yet this is still limited to the “elite” in water governance, and sectoral actors still dominate decision-making. An essential element of deliberation in the Freirian sense is that it includes a true dialogue between the powerful and the marginalized. None of our cases showed this form of deliberation. We did not find any incorporation of “lay knowledge”, except in the participatory forums established by EWS in Durban, where incorporation of community knowledge occurs within a managerial frame.

The research reveals that the locus of power has not shifted towards the new participatory or concertative institutions. Although Guarulhos opted for an autonomous water company, its actual decision-making power is weak because it depends on outside bulk water supply and outside finance for infrastructure, and major infrastructure decisions are still taken outside the councils. Metropolitan Lima exerts very little influence over its water company and it cannot influence water infrastructure decisions. And although municipalities in Arequipa have a stronger hold on their municipal water company, they made themselves dependent on finance from the private sector for infrastructure provision. There are signs of transition, or even transformation, in water governance, but water management decisions are still based on classic sectoral and economic interests and knowledge, maintaining the status quo. No signs of local accountability for large infrastructure investment decisions were evident.

The case of eThekwini Municipality, where national reforms significantly strengthened municipal autonomy, offers an interesting counter-example. Before the reforms, local governments already enjoyed relatively high levels of autonomy and more financial independence than the other case study municipalities. EWS took full advantage of the new legal framework and translated transformative national policy goals into innovative policies in service provision. The other municipalities are in a dependent situation regarding bulk water supply; in Durban, this power relation is more balanced. Learning is also strong in eThekwini Municipality and is characterized as deliberate experimentation and learning-by-doing. The municipality has been criticized for being inward-looking and the development of the UEIP is potentially an important shift towards a regional perspective, with results remaining to be seen.

Although the radical changes in the political regime in Brazil and South Africa occurred almost two decades ago, their impact on the practice of water governance might need more time to reach its full potential. Water governance is a particularly inert system, since the hard infrastructure of pipes and dams determines to a large extent the room to manoeuvre for the soft infrastructure of the governing institutions. The crucial question for the transformation of water governance systems in all cases will be whether, in the long run, participation and deliberative decision-making are extended to decisions about hard infrastructure and the provision of local water and sanitation services, and whether local powers are indeed empowered to hold the approach of “water as an economic good” to account.