Water Policy sample issue

Water Policy 16 (2014) 1–18

Climate-readiness, competition and sustainability: an analysis of the legal and regulatory frameworks for providing water services in Sydney Joanne Chong Institute for Sustainable Futures, University of Technology, Sydney, PO Box 123, Broadway, Sydney, NSW 2007, Australia E-mail: joanne.chong@uts.edu.au

Abstract This paper examines whether key legislative and regulatory frameworks for the provision of water services in Sydney, Australia, successfully support the complex task of planning and managing urban water systems to balance water security, cost and sustainability considerations. The challenges of managing urban water systems under a changing and uncertain climate became starkly apparent during Australia’s ‘Millennium Drought’, a decade-long period of extremely dry conditions throughout the 2000s. As the drought progressed, several state and territory governments assumed control of planning and approvals processes in order to implement large water-supply infrastructure projects with great urgency. However, at the end of the decade La Niña rains saturated catchments, spilled over dam walls and devastated several communities with flooding. Analysis of the frameworks for third-party access, private-sector participation, planning, and water-conservation initiatives reveals that the rules, roles and responsibilities of the many actors are interlinked but not always effectively integrated. The introduction and expansion of competition in the urban water industry are an ongoing experiment with great influence on the governance of the sector and the ways in which water services are planned for and provided.

Keywords: Climate change adaptation; Competition; Economics; Planning; Public–private partnership; Regulation; Sustainability; Third-party access; Urban water; Water law

1. Introduction The challenges of managing urban water systems under a changing and uncertain climate became starkly apparent during the ‘Millennium Drought’, a decade-long period of extremely dry conditions across southern Australia throughout the 2000s. For most city dwellers, by the onset of the Millennium Drought more than a generation had passed since the prospect of water shortages had last had a direct doi: 10.2166/wp.2013.058 © IWA Publishing 2014

2

J. Chong / Water Policy 16 (2014) 1–18

impact on daily life (Keating, 1992; Sydney Water, 2009). This protected experience ended when restrictions on outdoor water uses were imposed in many Australian metropolitan areas, with the aim of slowing the decline of water storage levels (Chong et al., 2009). As the drought progressed, several state and territory governments assumed control of planning and approvals processes in order to implement large water-supply infrastructure projects, notably desalination plants, with great urgency.

However, in 2009 and 2010 across south-east Australia, the drought broke with the arrival of La Niña rains that saturated catchments, spilled over dam walls and afflicted many communities with devastating floods.

The water-use restrictions and major infrastructure developments implemented in response to the drought sparked heated debate about the effectiveness and appropriateness of the regulatory and institutional architecture of the urban water sector (National Water Commission, 2011). This stemmed from concerns about the alleged failure of governments to plan for and adequately invest in water supply systems; the financial, ecological and sustainability impacts of infrastructure; and the extent to which planning processes geared towards rapid deployment of infrastructure sacrificed processes that would have otherwise ensured transparency and community engagement (National Water Commission, 2011). The most recent cycle of rainfall extremes, and the controversies surrounding approaches to secure water supplies, have refocused the reform spotlight on the settings underpinning the supply and management of water for Australian cities. National government advisory bodies have found that significant further reforms are urgently required, despite several decades of reform built on national competition principles intended to promote innovation and efficiency (National Water Commission, 2011; Productivity Commission, 2011).

This paper critically examines whether key legislative and regulatory frameworks for the urban water sector in metropolitan Sydney, Australia, successfully support the complex task of planning and managing systems to balance water security, cost and sustainability considerations. The analysis and lessons presented have widespread relevance for those grappling with urban water reform and integrated planning internationally. Across many countries, regulators, participants and stakeholders are currently being challenged by questions of how to balance often competing policy objectives of access, affordability, equity, environmental protection and economic efficiency (François et al., 2010; Bayliss, 2011). Moreover, climate change, uncertainty and extreme droughts and floods, such as those typical of the Australian climate, are increasingly adding complexity to ensuring water security (Parry et al., 2007).

The paper focuses on the provision of water services but notes that wastewater services are integrated and there are significant overlaps in the case of recycling. The conceptualisation of ‘climate-readiness’ centres on the provision of services under a changing climate, amongst other climate-change issues relevant to the urban water sector (Water Services Association of Australia, 2012a). The questions it addresses are:

• Have key legislative and regulatory frameworks facilitated ‘climate-ready’ urban water systems: that is, do they ensure secure and reliable water services under more variable, less certain and potentially drier climate conditions, as expected for the Sydney region? (NSW Office of Water, 2010) • Do these frameworks balance the goal of water security with cost-effectiveness (value for money in terms of the options implemented to save or supply water), sustainability and other public-interest considerations?

• Do the competition reform elements embedded within the legislative and regulatory frameworks drive or inhibit the achievement of these multiple objectives?

J. Chong / Water Policy 16 (2014) 1–18

3

The paper is structured as follows. Section 2 is a brief description of the supply and demand for water and the institutional arrangements for the delivery of services in the Sydney metropolitan region. Section 3 outlines the urban water reform process in Australia. The legal and regulatory frameworks underpinning three key areas are then analysed: third-party access and private-sector participation in urban water service provision, particularly via recycling (Section 4); development controls, the planning regime and their application to water infrastructure (Section 5); and water conservation initiatives (Section 6). The risks to water supplies from coal-seam gas exploration and mining are beyond the scope of this paper but are, importantly, being addressed by others (National Water Commission, 2010; Cubby, 2012). Section 7 concludes. The paper outlines the many actors and their involvement in regulating and supplying water services in Sydney. As a guide, key examples of the statutory basis for these responsibilities are summarised in Figure 1, but this diagram necessarily contains a sample rather than a comprehensive map of obligations.

2. A brief description of metropolitan Sydney’s water services system 2.1. Supply and demand Most of the water supplied to greater metropolitan Sydney’s population of approximately 4.6 million is sourced from catchment-based run-off. Warragamba Dam is the main storage and, at over 2,000 gigalitres in size, is one of the largest in the world in terms of per-capita capacity (Department of Environment, 2010; Sydney Catchment Authority, 2012a). By 2015, alternative sources of desalination and recycling initiatives are expected to supply about a quarter of Sydney’s water needs, including via private-sector involvement in provision, as outlined in Section 3. Households consume about 73 per cent of all water used in greater Sydney. Water efficiency programmes for residences and businesses and increasing numbers of recycling initiatives have successfully reduced demand per person per day from 506 litres in 1990–1991 to 314 litres in 2009–2010 (Department of Environment, 2010).

Mandatory restrictions on outdoor watering were in place in Sydney from 2003 to 2005. These restrictions were nominally enforced through a system of fines, but compliance was achieved through public education and monitoring by members of the public. Sprinkler use was banned and watering by hand-held hose or dripper systems was only allowed from 4 pm to 10 pm on certain days of the week (Chong et al., 2009). As the drought progressed and dam levels fell, the New South Wales (NSW) State Government commissioned a desalination plant at Kurnell, at a cost of AU$2 billion (billion ¼ 109). The plant has a capacity of up to 90 gigalitres per year (about 15 per cent of current water use) and is designed with the potential for a doubling in capacity in the event of a future severe drought (Department of Environment, 2010). However, high rainfall levels that occurred since the drought broke in 2010 resulted in an overflowing Warragamba Dam by March 2012 (Sydney Catchment Authority, 2012b) and led to the plant being switched off in July 2012. In May 2012, the NSW Government leased the Sydney Desalination Plant Pty Ltd to a private international consortium. At the end of the 50-year lease agreement, the ownership will be transferred to the lessee (Sydney Water, 2012a).

2.2. Institutions and governance The governance arrangements underlying the supply of water to greater Sydney reflect the competition reform principles articulated in the National Competition Policy, as outlined in Section 3.

4

J. Chong / Water Policy 16 (2014) 1–18

Fig. 1. Actors involved in the provision and regulation of urban water services in Sydney. The responsibilities and Acts are examples and this diagram is not intended as a comprehensive guide.

Water is supplied by the Sydney Water Corporation (SWC), a statutory state-owned corporation (SOC) that operates in accordance with the Sydney Water Act 1994 (NSW) and its operating licence (Sydney Water, 2010). The responsibility for protecting the quality and quantity of water in catchment areas and for bulk water in the Sydney region lies with the Sydney Catchment Authority (SCA), established in 1998 under the Sydney Water Catchment Management Act 1998 (NSW) (McKay, 2005).

The multiple objectives for SWC, as outlined in its operating licence, include: being a successful business; having regard for the interests of the communities; complying with the principles of

J. Chong / Water Policy 16 (2014) 1–18

5

ecologically sustainable development and preventing the degradation of the environment; and protecting and reducing risks to public health (Sydney Water, 2010). The dual nature of the ‘private business’ and ‘public entity’ objectives underlies the intricacy of legislative, regulatory and governance arrangements required to enable achievement of these multiple goals.

The Independent Pricing and Regulatory Tribunal (IPART) is the economic regulator with a primary function, in the water sector, to determine the maximum prices that can be charged for water and wastewater services. IPART also monitors SWC’s and SCA’s compliance with their respective operating licences, and has regulatory and administrative functions with regard to licensing private-sector provision of water services.

The development of the Metropolitan Water Plan, which outlines the mix of measures planned to ensure water security for greater Sydney, is coordinated by NSW Government (Department of Environment, 2010). The 2010 Metropolitan Water Plan includes a portfolio of water supply options including desalination, recycling, stormwater harvesting, and drought-readiness measures. It emphasises the need for diversification as well as flexibility to respond and adapt to challenges presented by a highly variable climate, droughts, and climate change (Department of Environment, 2010).

3. Background to reform in the Australian urban water sector The process of strategic reform of the Australian water industry has its origins in the National Competition Policy (NCP), implemented from 1995 to 2005 by the Council of Australian Governments (COAG) (Productivity Commission, 2005). The NCP was based on the recommendations of the Independent Commission of Inquiry into National Competition Policy (the Hilmer Inquiry), which endorsed the principle that competitive markets would generally best serve the interests of consumers and the community, by creating incentives that promote supplier innovation and efficient operations (Committee of Inquiry into National Competition Policy, 1993; Productivity Commission, 2005). The guidelines established by the resultant Competition Principles Agreement concerning the structural reform and regulation of public monopolies underpinned the reform of government monopolies in various service-provision sectors, including ultimately the corporatisation of urban water businesses (as government trading enterprises) and the establishment of price-regulatory functions of government agencies.

In 1994, COAG adopted a strategic framework for water industry reform, which was subsequently linked to AU$16 billion of payments made available by the Commonwealth to incentivise State and Territory implementation of the NCP agreements over 1997–1998 and 2005–2008 (Lyster et al., 2009). A key element of the 1994 urban water reform package was the separation of policy setting, regulatory enforcement and service-delivery functions of government agencies (National Water Commission, 2011). The 1994 reforms also required a restructuring of water tariffs and reduction of cross-subsidies in line with cost-reflective pricing. The intention, although ultimately not the overarching outcome, was that governments would set clear policy objectives for the sector, and provide water utilities with the autonomy to deliver services that met these objectives.

Between 2004 and 2006, all Australian governments signed the Intergovernmental Agreement on a National Water initiative (NWI), which included the establishment of the National Water Commission (NWC) to advise COAG on water issues and progress under the NWI. The NWI commitments and its approach to addressing environmental considerations predominantly relate to non-urban water issues.

Nevertheless, the NWI commitments included a limited set of urban water actions, largely focused

6

J. Chong / Water Policy 16 (2014) 1–18

on pricing (Council of Australian Governments, 2004). In 2008, COAG agreed to a further set of actions to progress urban water reforms (Council of Australian Governments, 2008). The actions relevant to preparing the sector for climate change included publishing guidance on best-practice scenario planning for climate variability; promoting private-sector participation through third-party access; and establishing and funding major research centres for recycling and desalination.

In its 2009 biennial assessment of progress against the NWI, the NWC found that the NWI did not sufficiently guide urban water reform, and that ‘new challenges have presented themselves [that were] not as evident when the NWI was signed’ (National Water Commission, 2011, p. 57). These challenges included uncertainty about rainfall and run-off levels due to climate change. In its 2011 assessment of future directions for the urban water sector in Australia, the NWC found that, despite some progress, there are ongoing inadequacies in the policy and institutional settings for urban water, including unclear specification of responsibilities of the various agencies involved and a lack of agreed objectives for the sector.

4. Third-party access: recycling and the Water Industry Competition Act 2006 (NSW) 4.1. Origins of the NSW third-party access regime for water Regulatory arrangements to enable third-party access to ‘essential’ infrastructure services were a key pillar of promoting the competition objectives of the NCP. Part IIIA of the Trade Practices Act 1974 (Cth) (replaced in 2011 by the Competition and Consumer Act 2010 (Cth)) created a national access regime to enable access by parties, other than the infrastructure owner, to network facilities. The purpose of access regulation is to enable access to services provided by infrastructure or other facilities that are not easily obtainable through commercial negotiation because of the natural monopoly characteristics of the infrastructure and facilities in question. Statutory third-party access regimes also provide a framework for an independent regulatory agency to arbitrate the terms of access where the infrastructure owner and the third-party access seeker fail to agree on the terms of access (Gray & Gardner, 2008).

The effectiveness of Part IIIA in enabling access to urban water infrastructure was tested by the application by Services Sydney Pty Ltd. After failed negotiations with Sydney Water to gain access to its sewerage-regulation network in order to treat and recycle sewage water for non-potable purposes including agriculture, industrial and domestic uses and environmental flows, Services Sydney sought declaration of various sewerage networks under the TPA Part IIIA. Sydney Water opposed the application on many grounds including cost and concerns about consumer protection, public safety, and environmental standards, but after initial delays the Australian Competition Tribunal declared the service for a 50-year period. However, a dispute over access-pricing methodology followed. The results of arbitration found in Sydney Water’s favour, and Services Sydney did not subsequently take up access (Gray & Gardner, 2008; Abbott & Cohen, 2011).

This test case was a catalyst for the introduction of the Water Industry Competition Act 2006 (NSW) (WICA), which established a state-based access regime. The NSW Government’s motivation for introducing the regime was to facilitate competitive service provision and to promote innovative solutions such as sewer mining that had potential for supplying water throughout drought (Gray & Gardner, 2008; National Competition Council, 2009). As required by WICA, as 5 years have passed since its introduction, the Minister is currently reviewing the Act (NSW Government, 2011).

J. Chong / Water Policy 16 (2014) 1–18

7

4.2. Water Industry Competition Act 4.2.1. Features of WICA. The objectives of WICA and associated regulations are wide-ranging and ambitious. The core elements are the establishment of a licensing scheme for private-sector providers of reticulated drinking water, recycled water and sewerage services; the establishment of a third-party access regime for water and sewerage infrastructure; and provisions to authorise IPART to facilitate resolution of disputes over access to infrastructure (Water Industry Competition Act 2006 (NSW)). Unlike Part IIIA, there are no merit-review processes for access decisions under WICA, which can only be tested by judicial review (Crosbie, 2009). Since the introduction of WICA, no applications for sewernetwork access have been received or granted; the only access undertaking to date has been sought

by Sydney Water to enable access by the operators of the Sydney Desalination Plant to the drinking supply network (Independent Pricing and Regulatory Tribunal, 2012a). The adequacy of disputemechanism elements regarding access have thus not had to be tested in practice.

Regulation of private-sector participation in recycling has been enabled in a number of cases through WICA’s licensing provisions. Under part 2 of WICA, the two types of licences that may be applied for are a network operator’s licence and a retail supplier’s licence (Water Industry Competition Act 2006 (NSW) pts 3, 4; Explanatory Notes, Water Industry Competition Bill 2006 (NSW)). In applying for a network licence, the applicant must provide comprehensive statements with regard to: the activities; minimising risks to their ability to carry out the activities; technical details about the system including information on how safe and reliable supplies will be ensured if infrastructure is inoperable; and addressing relevant guidelines for managing health and environmental risks (Water Industry Competition (General) Regulation 2008 (NSW) s 6). Applicants for a retail supplier’s licence must address similar issues (Water Industry Competition (General) Regulation 2008 (NSW) s 10).

A raft of government agencies provide advice for each WICA licence approvals process. Licence applications are lodged with IPART, which invites and considers submissions from those Ministers administering the Public Health Act 1991 (NSW), the Water Management Act 2000 (NSW), the Environmental Planning and Assessment Act 1979 (NSW) (EPAA), the Protection of the Environment Operations Act 1997 (NSW), as well as the public (Water Industry Competition Act 2006 (NSW) s 9). Before granting a retail supplier’s licence, the Minister must be satisfied that activities will be carried out without presenting a significant risk of harm to the environment. Licence conditions for network operators and retail suppliers include compliance with plumbing codes and regulatory requirements for water quality; reporting of information relating to incidents that could threaten public health or the environment; the preparation, publication and auditing of operating and management plans; and other conditions imposed by the Minister (Water Industry Competition Act 2006 (NSW) s 13).

WICA requires licensees of recycling schemes to address the 12 elements of the risk-management framework detailed in the ‘Australian Guidelines for Water Recycling: Managing Health and Environmental Risks’ (AGWR) (Water Industry Competition (General) Regulation 2008 (NSW) s 6(1)(d)).

These Guidelines include rigorous validation requirements and are designed to ensure that users or the environment are not exposed to undue risks.

WICA also includes ‘retailer of last resort’ (RoLR) provisions that aim to protect customers and ensure continuity of services in the event that a private retail supplier fails to provide services. These arrangements are currently under review and the discussion paper released by the State Government has identified a number of gaps in the regulatory arrangements (NSW Department of Finance &

8

J. Chong / Water Policy 16 (2014) 1–18

Services, 2011). For example, the legislation does not specify commercial parameters or cost-recovery requirements, leaving SWC (and hence its customer base) potentially exposed to bearing the costs of stepping in to act as retailer in the event that a private retailer fails to provide services. Further, WICA does not include arrangements for ‘operators of last resort’ (OoLR), which would be designed to respond to network operator failure, including the protection of ‘off-grid’ communities that are not otherwise connected to mains networks. The consequences of supplier failure have fortunately not eventuated in practice, but the issues around RoLR and OoLR arrangements illustrate the crucial role of regulation of private-sector involvement.

4.2.2. The influence of WICA on water recycling: evidence and issues. A total of 19 licences have been granted by IPART for nine schemes across Sydney (Independent Pricing and Regulatory Tribunal, 2012b). One example is the 5.8 ha ‘Central Park’ urban village currently being developed near central Sydney. This development is expected to house approximately 5,000 residents and host more than 15,000 workers and visitors daily. A private company has applied for licences to own and operate drinking water, recycled water and sewerage infrastructure (Independent Pricing and Regulatory Tribunal, 2012c). A climate-ready character is reflected in the scheme’s diverse sources of water, including rainwater, stormwater, groundwater, sewage from onsite buildings, and via access to a SWC sewer main that lies underneath the site.

SWC also has an active role in many recycling schemes including public–private partnerships. One such example is the Rosehill Recycled Water Scheme, the first scheme licensed under WICA (Water Services Association of Australia, 2012b), which supplies recycled water for industry and irrigation in western Sydney. A private consortium built, owns, and operates the recycled water treatment plant and supply network (Independent Pricing and Regulatory Tribunal, 2012b); SWC has an agreement to purchase the recycled water from the private consortium and, in turn, sells water to industry customers (Sydney Water Corporation, 2009). As outlined above, WICA licensing focuses on ensuring safe and reliable service provision along with measures for protecting public health and the environment. The negotiating of the commercial terms of contracts between the various private and public parties, which can ultimately have implications in terms of risks and costs to the community, lies outside the economic regulatory framework.

To date, the schemes licensed under WICA represent a small proportion of the total number of recycling schemes and volume of water recycled across greater Sydney. SWC has developed and operates its own recycling schemes (Sydney Water, 2012b) and is required by its operating licences to comply with various health and environmental guidelines. Most recycling schemes in NSW are operated by councils, whose wastewater recycling operations (in non-metropolitan regions only) are regulated under section 60 of the Local Government Act 1993 (NSW) (LG Act). Although beyond the scope of this paper to address, there are a number of issues relating to the capacity of and support available to local councils to effectively apply the 12 steps of the AGWR (Power, 2010). There are also numerous private schemes that do not, or do not yet, require licensing under WICA. The regulatory and governance model for private-sector involvement underpinned by WICA directly reflects the principles of the pervasive competition and economic reform framework outlined in the NCP. The WICA review process (NSW Government, 2011) is expected to examine the extent to which private-sector involvement has yielded innovation and economic efficiency benefits, and to what extent health, environmental, amenity and other social considerations have been adequately protected by the broader regulatory frameworks.

J. Chong / Water Policy 16 (2014) 1–18

9

5. Development assessment, the new planning regime and urban water infrastructure ‘The reality is that two decades of governance reform has yet to keep politics out of the day to day management of water businesses, particularly during periods of crisis’ (Water Services Association of Australia, 2011, p. 12).

‘Major investment decisions made by governments remain beyond the scrutiny of economic regulators, even though those investments account for most increases in costs and hence price increases to customers’ (National Water Commission, 2011, p. viii).

Of the broad suite of planning laws relevant to the provision of urban water services, those addressing development controls are particularly relevant to assessing and approving urban water infrastructure.

The NSW planning regime is currently undergoing a major review, with expected significant changes to specific elements as well as to the overall architecture of the scheme (NSW Government, 2012).

Changes are expected to occur rapidly and be finalised by the end of 2013. During this current period of reform, it is timely to first revisit and critique the previous development-control provisions that facilitated the implementation of the major desalination infrastructure, as a basis of comparison with the emerging directions in the new planning regime as outlined in A New Planning System for New South Wales – Green Paper (NSW Government, 2012).

5.1. Development assessment of the Sydney desalination plant In Sydney, as in many other cities, the drought of the 2000s precipitated the State Government assuming control of the process of development assessment and approval for the desalination plant. In 2004, as part of the Metropolitan Water Plan, the NSW Government had allocated funding to investigate desalination as an option for securing water supplies during drought, the main benefit being that desalination is a rainfallindependent source (White et al., 2006). The newly appointed Premier of NSW announced in August 2005 that the desalination plant would be built ‘drought or no drought’ (Smith, 2005), but this stance proved unpopular and the position was reversed in response to the recommendations of an expert panel commissioned to review the 2006 Metropolitan Water Plan (White et al., 2006). As recommended by this review, in February 2006 the NSW Cabinet announced that the desalination plant would be integrated into an emergency drought-readiness strategy, and would be constructed only if dam levels dropped to 30 per cent (New South Wales Parliament, 2006). However, when dam levels fell to 35 per cent, the State Government announced that it had called for tenders for construction (White et al., 2008). Immediately following this, the drought broke and the 30 per cent trigger point was never reached (Sydney Catchment Authority, 2012c).

Regulatory changes in the period shortly preceding the initiation of the desalination plant enabled the NSW Government to assume control over the approvals process. A new clause was entered into the State Environmental Planning Policy (Sydney Metropolitan Water Supply) 2004 enabling SWC to develop the desalination plant without development consent under Part 4 of the EPAA (Lyster et al., 2012). The Minister exercised discretionary authority by declaring the plant a Part 3A project, of ‘State or regional environmental significance’, as well as ‘critical infrastructure’, which is ‘essential for social, economic or environmental reasons’ (Environmental Planning and Assessment Amendment (Infrastructure and Other Planning Reform Act 2005 (NSW) s 1).

10

J. Chong / Water Policy 16 (2014) 1–18

During its 6-year existence, Part 3A of the EPAA was criticised for the extent of discretion it allowed the Minister, its inadequate and uncoordinated assessment requirements, and for failing to achieve the objectives of the EPAA: specifically, that it did not require adequate consideration of ecologically sustainable development, climate change, consideration of alternatives, public participation or consultation, or appeals or judicial review (Ghanem & Ruddock, 2011; Carr, 2007; Minister for Planning v Walker (2008) 161 LGERA 423). Part 3A has since been repealed, but the current ‘critical state-significant infrastructure’ provisions are similar in process and scope.

During the process of developing the environmental assessment report (EAR), concerns about ecological impacts of the desalination plant, local environmental impacts, energy use and climate change impacts, inadequate public engagement and insufficient consideration of alternatives were raised by the community, non-government agencies, academics and businesses (Prince et al., 2006; NSW Government, 2006; Ghanem & Ruddock, 2011). However, an independent panel appointed by the Minister to review matters regarding the Kurnell desalination plant found that all issues raised had been adequately addressed by the EAR process or in the Department’s advice to the Minister (Prince et al., 2006).

The issue of cost and the resultant financial impacts on communities were beyond the reach and objectives of the EPAA (Grafton & Kompas, 2006). This relates to the specific ‘pulling the trigger early’ on the desalination plant, and concerns that opportunities for more cost-effective investment in urban water options had not previously been undertaken (UTS, 2007). While the critical infrastructure provisions of the EPAA could be interpreted as enabling Ministerial flexibility to intervene in order to avert a water shortage ‘crisis’, the processes did not themselves enable cost-effective planning for urban water, nor did they constitute a mechanism for community concerns about financial impacts to be considered, nor an appeal to be made on this or any other basis. The case of the Sydney desalination plant illustrates the difficulties in separating government’s roles as policy-maker and regulator from that of service provider.

5.2. A new planning regime: implications for urban water infrastructure In July 2011, the newly elected NSW Government embarked upon a major review of the state’s planning system. Recommendations of an Independent Review Paper were released in March 2012 and A New Planning System for New South Wales – Green Paper was released in July 2012 (NSW Government, 2012). The Green Paper proposes major changes to the planning system and presents several implications for strategic planning of urban water systems, environmental assessment and development control of water infrastructure, and private-sector participation in the provision of urban water services (NSW Government, 2012).

5.2.1. Strategic planning. The Green Paper proposes a shift of focus and attention from development assessment to strategic planning. To achieve this, it proposes major structural change at all levels of the planning system to result in a spatially nested framework of strategic planning: NSW planning policies, metropolitan/regional growth plans, subregional delivery plans and local land-use plans (NSW Government, 2012).

Within this proposed strategic planning framework, guidance for urban water infrastructure provision is included in the 20-year State Infrastructure Strategy (SIS) released in September 2012 (Infrastructure NSW, 2012). This is intended to link to the metropolitan/regional growth plans, although how this

J. Chong / Water Policy 16 (2014) 1–18

11

linking will be achieved is not articulated within the Green Paper. The water section of the SIS does not make apparent whether it will contribute to filling gaps in the planning for cost-effective, sustainable water service provision for the metropolitan water area under changing climate conditions. The SIS refers to the Metropolitan Water Plan (MWP), but makes only the limited recommendation that the review of the MWP take into account infrastructure options.

The Green Paper also proposes to introduce a legislative basis for strategic plans, although the aims and character of this legislative basis have not yet been specified (NSW Government, 2012). In the case of urban water, it is yet to be seen whether the new legislation will establish governance arrangements that appropriately support the process of planning and approvals for infrastructure. The Green Paper also proposes to include statutory requirements for community participation in the development of strategic plans, although it is also not clear whether this requirement will apply to metropolitan strategic plans, the level relevant to urban water service provision, or just to subregional delivery plans.

5.2.2. Environmental assessment and infrastructure approval. Despite relatively limited specific reference to water service provision, there are elements of the Green Paper relating to environmental assessment and approval of infrastructure that are of clear relevance to the sector. Firstly, the critical state-significant infrastructure provisions of the current EPAA (effectively similar to the critical infrastructure provision under the previous Part 3A) will be carried forward into the new regime (NSW Government, 2012), providing the Minister with the same flexibility to declare and fast-track infrastructure, including for water supply. This, however, leaves open the question about what mechanisms will ensure planning for cost-effective and sustainable water infrastructure.

Secondly, the Green Paper makes a clear statement that for ‘public priority infrastructure’ the place for balancing multiple objectives of water security, cost, and other environmental and social objectives lies at the strategic planning level, not within environmental assessment (EA) (NSW Government, 2012). Within the broad range of definitions and applications of EA, this proposal represents a relatively narrow scope. It does not, for example, require the EA to consider other options, and does not enable environmental impacts that are only realised through rigorous assessment at the EA stage to form the basis for declining development approval. This is likely to raise concerns similar to those that have previously arisen in relation to EPAA ‘concept stage’ approval processes for state-significant infrastructure; through such processes, approvals based on limited information about impacts and with reduced review rights have alienated communities (EDO NSW, 2012). As yet, there is little specific detail about appeal and review mechanisms available under the new system.

This placement of limits on the EA in the pursuit of streamlining development-approval processes mirrors a general trend across Australian jurisdictions to limit ‘green tape regulation’ in order to reduce the costs to business (Lowe, 2012). A question thus remains as to whether the strategic planning processes will be integrated with the Metropolitan Water Plan to provide a sound foundation for water infrastructure planning processes to adequately balance environmental objectives with the need to secure water supplies.

5.2.3. Contestability. The Green Paper also makes explicit a proposal to enhance contestability in order to encourage private-sector provision and operation of infrastructure networks, including urban water (NSW Government, 2012). It identifies that implementing contestability will lead to cost savings and encourage innovation. It proposes that new access arrangements should be established to facilitate private-sector participation. IPART has agreed that contestability will facilitate benefits including

12

J. Chong / Water Policy 16 (2014) 1–18

removal of a potential bias towards traditional capital-intensive solutions (Independent Pricing and Regulatory Tribunal, 2012d).

In the case of water, WICA provides the framework required for contestability outlined in the Green Paper. As described above, the frameworks for RoLR and potentially OoLR need to be further developed in terms of ensuring appropriate financial risk-sharing in the event of failure, so that customers can be assured of services for safe and secure water, and for wastewater. Overall, the contestability provisions of the Green Paper are wide-ranging but provide little specific detail. At least some industry leaders have previously argued that, for alternative sources of bulk water such as desalination and recycling, the potentially high costs of treatment and pumping for desalination and recycling and the large existing storages may not lend themselves to cost-effective private-sector involvement in service provision (Schott et al., 2008). The Green Paper does not raise the possibility that it may not be appropriate, nor economically efficient, for some functions of the urban water industry to be open for contestability.

6. Water conservation ‘As a result of the drought the community at large has changed the way it values water … Regulation is playing a key role in achieving water savings and will play a greater role in the future’ (Sydney Water, 2011).

‘Prescribing the use of water efficient appliances in buildings obliges all consumers to use water efficient appliances … by restricting consumer choice, policies that either prescribe or encourage (through moral suasion) water savings lead to inefficiencies’ (Productivity Commission, 2011).

Water savings initiatives for Sydney that are underpinned by legislation include targets for water use, establishment of water savings funds, restrictions of water use during drought, and mandatory water efficiency requirements.

6.1. Implementing water savings programmes The Energy and Utilities Administration Act 1987 (NSW) (the EUA Act) establishes the basis for SWC and the NSW Government to fund and implement water savings programmes. The objects of the EUA Act in relation to water apply to the area of operations of SWC, and include: to reduce the demand for water; stimulate investment in innovative water savings measures; increase public awareness and acceptance of the need to save water; improve access to a wider range of water savings technologies; and encourage the use of non-potable water (EUA Act s 5(3)). The EUA Act establishes the Climate Change Fund, incorporating the previous Water Savings Fund (Energy and Utilities Administration Act 1987 (NSW) s 34) and since 2005 the Minister has required SWC to make annual contributions. In 2010–2011 SWC contributed over AU$23 million to the Climate Change Fund and received AU$6.7 million for its Demand Management programme (NSW Department of Environment & Heritage, 2012). The EUA Act also requires local councils within the SWC area of operations to prepare and submit water savings plans for approval by the Minister (Energy and Utilities Administration Act 1987 (NSW) ss 34A, 34O).

Sydney Water also has specific water conservation conditions included within its operating licence.

The conditions include requirements to maintain the level of drinking water supplied from all sources to

J. Chong / Water Policy 16 (2014) 1–18

13

less than 329 litres per person per day, limit the level of water leakage from its drinking water supply system, undertake and promote water efficiency programmes, and meet targets set for water recycling consistent with the objectives of the metropolitan Water Plan including as directed by the Minister (2010).

These underpin a robust approach to facilitating water savings in the Sydney area, and the water efficiency and conservation programmes implemented by Sydney Water have been commended by the Australian water industry and internationally for their innovative design and water savings achievements, and for their associated energy and greenhouse-gas savings. However, the regulatory approach that mandates SWC’s implementation of water savings initiatives was found by the Productivity Commission to pose significant barriers to achieving economic efficiency, and to be a key area for further reform (Productivity Commission, 2011). The argument is that mandatory targets for water savings are pursued through water efficiency programmes that restrict consumer choice and require government to ‘pick winners’; that is, targets have the effect of arbitrarily favouring demand management, and preclude consideration and implementation of other potentially more cost-effective water supply options.

Nevertheless, without specific obligations mandating water savings programmes, it is not apparent how water efficiency, conservation, and other demand-management programmes that save water without requiring significant built infrastructure would be given equivalent consideration to supply-side options, irrespective of their relative cost-effectiveness. Water utilities as monopolies are in theory expected to act as profit maximisers, and hence the approach to price regulation is designed to prevent utilities from overpricing and seeking monopoly rents. In reality, however, utilities as government-owned enterprises are subject to political pressures and have been found by the Productivity Commission to tend towards underpricing and failing to generate adequate revenue for efficient investment in the systems (Productivity Commission, 2011). The underpricing of water means that, without mandatory requirements for investment in water conservation programmes, there is little economic incentive for residents and businesses to limit their water use to economically efficient levels, although undoubtedly household and business actions are also influenced by considerations other than price.

Furthermore, as outlined earlier, the principal objectives of SWC as outlined in the Sydney Water Act 1994 (NSW) are to ‘operate a successful business’ and to earn a required rate of return on assets. IPART in its price determinations is required to consider whether services are being met in the most economically efficient way, and under recent price determinations the focus of efficiency gains is on minimising operating costs (Independent Pricing and Regulatory Tribunal, 2012e). Water conservation programmes rarely create significant assets but nevertheless require ongoing expenditure. With prices set to recover expenditure on assets, it is not clear how the incentives created by economic regulation would encourage utilities to implement cost-effective water savings programmes, if they were not otherwise mandated by regulation.

6.2. Restricting water use during drought The Sydney Water Regulation 2011 (NSW) provides the Minister with discretionary powers to regulate or restrict the purposes, times, quality or methods of water use, if the Minister is of the opinion that it is necessary in the public interest and for the purpose of maintaining water supply (s. 18). These powers were exercised across NSW, including Sydney, during the restrictions’ period of 2004 to 2006 (Chong et al., 2009).

14

J. Chong / Water Policy 16 (2014) 1–18

During the 2000s in Sydney and across other Australian cities and towns, there was widespread debate about the appropriateness of water restrictions in place. A divergent range of perspectives emerged from the community, urban water managers, government officials, and elected representatives (Chong et al., 2009). Community surveys indicated support for at least the less stringent restrictions, and showed that these restrictions engendered a sense of community pride and were perceived to underpin equality of participation in saving water during drought (Chong et al., 2009). The previous Chairman of IPART noted ‘there is a high cost to drought proofing a city so that drought never has an impact’ (Keating, 2006); and many in the water industry considered that water restrictions were a flexible mechanism to reduce demand quickly during drought, and delay the need for investment in costly infrastructure (Chong et al., 2009). Nevertheless, the resultant overarching policy message since the drought is that any restriction on water use imposes costs on consumers, curtails economic efficiency and should not be included as a drought-response option (National Water Commission, 2011; Productivity Commission, 2011). Given the apparent unassailability of this stance, it is not apparent that Ministerial discretion will be exercised to restrict water use in order to manage climate variability in the future.

6.3. Mandating water conservation in the home: BASIX About 73 per cent of all water in Sydney is used in the residential sector, and water use in homes represents significant potential for saving water (Department of Environment, 2010). Water efficiency and conservation initiatives also reduce energy use and greenhouse-gas emissions associated with pumping water through the system; where water savings are achieved through a reduction of hot water use (such as through water efficient savings), the avoided energy costs and greenhouse-gas emissions can be substantial (Chong et al., 2008). Under changing and potentially drier climate conditions, water conservation measures would thus seem to be critical to achieving the goals of water security as well as reducing greenhouse-gas emissions.

In NSW, water conservation inside the home is mandated by the BASIX (Building Sustainability Index) scheme, which was established in 2004 and incorporated into the Environmental Planning and Assessment Act 1979 (NSW). BASIX is incorporated into the SEPP – Building Sustainability Index (BASIX) (2004) and the Environmental Planning and Assessment Amendment (Building Sustainability Index (BASIX) Regulation 2004. BASIX requires new residences, additions and alterations to be designed and built to use 50 per cent less mains water supply and produce 25 per cent lower greenhousegas emissions than standard building stock. The water savings achieved in BASIX via water efficiency of fittings and some fixed appliances were enabled by the introduction of the national Water Efficiency Labelling and Standards Scheme, which requires certain products to be registered and labelled with their water efficiency in accordance with the standard set under the national Water Efficiency Labelling and Standards Act 2005 (Cth).

On its introduction in 2004, BASIX was welcomed as an innovative mechanism to achieve sustainability outcomes, including water savings, in homes (Thorpe & Graham, 2009). However, savings targets have since been criticised for being too low and of limited scope, for example, by setting weaker targets for multi-unit housing and excluding renovations costing less than AU$50,000, and for not applying to the non-residential sector. Where the bar is set, in terms of water savings required, is determined by the BASIX scheme, because the SEPP also precludes any local environmental plan or development-control plan from imposing higher standards for energy or water consumption (Ghanem & Ruddock, 2011).

J. Chong / Water Policy 16 (2014) 1–18

15

Whether and how BASIX will be incorporated into the new planning regime and what scope local governments will have to set standards remain to be seen.

Voluntary programmes have the potential to deliver significant water savings. The voluntary, national Green Star scheme, administered by the not-for-profit Green Building Council of Australia, has been instrumental in driving developers to install water savings initiatives and recycling schemes to achieve five-star rating for their residential and commercial building and precinct developments. In recent years, achieving this rating has been critical to developers quickly securing premium tenants in an otherwise slow market. Nevertheless, the Green Star scheme has far less influence over smaller or individual residential developments or renovations, which are regulated by the minimum standards of BASIX.

7. Conclusion The legislative and institutional arrangements underpinning the supply of urban water services in Sydney comprise many actors whose rules, roles, and responsibilities are interlinked but not always effectively integrated. In particular, there are various tensions between public organisations’ multiple responsibilities as policy-makers, economic regulators, protectors of public health and the environment, planners, system constructors and operators, and facilitators of competition.

Given the heightened uncertainty of rainfall and run-off conditions expected under climate change, long-term, coordinated planning of urban water options by an entity charged with representing the public interest would seem paramount to effective balancing of water security, cost, sustainability, and other objectives. Planning and development-control laws as an instrument for fast-tracking essential infrastructure, for example, to respond to drought, would not be a required element to ensure water security if the overall governance framework supports an adaptive management approach to planning for and responding to climate uncertainty. There is little detail yet available for the forthcoming NSW planning regime to indicate whether or how an integrated approach to planning urban water systems will be supported.

WICA provides a legal framework for regulating potentially innovative private-sector solutions to recycling water as a rainfall-independent source of supply. What is clear is that private-sector involvement in establishing and operating a recycled water scheme requires extensive government oversight in regulating environmental and public health risks. Private-sector involvement in sewer mining and other schemes also potentially creates a complex role for government or utilities, as negotiators of commercial contracts on behalf of the public ratepayers. Whether the resultant risk-sharing agreements between ratepayers and private-sector participants, and the extent of government involvement in regulation, represent best value for money for the public has not been tested empirically. Furthermore, there is still an ongoing need for government involvement in integrated planning and investment for water security infrastructure. Outstanding questions remain about how coordination of public and private service provision will be achieved.

Programmes that aim to reduce water use would appear to successfully achieve water security, climate change mitigation, and, by delaying the need for system augmentation, potential cost savings. Mandatory obligations on utilities, local government and major water users to achieve water savings have been criticised because they, in principle, favour water conservation programmes over other potentially more cost-effective measures. While this may be the case in theory, the reality is that, in practice, without the

16

J. Chong / Water Policy 16 (2014) 1–18

regulatory requirement for utilities to invest in water conservation programmes, there would be limited economic incentive for major suppliers of water to encourage reduction in water use.

Ultimately, the introduction and expansion of the number of actors delivering urban water services are an ambitious experiment that continues to have great influence on the governance of the sector and the ways in which water services are planned for and provided. The original NCP-based hypothesis did not position competition per se as the ultimate goal, but rather as a means to achieve economically efficient delivery of water security. In terms of evidence for or against, it remains to be determined whether water sector security under climate change is adequately achieved through the complex sum of public investment, private-sector innovation, government agency planning, ministerial discretion over infrastructure approvals, administration of private-sector involvement, economic price regulation, and regulation to ensure that health, environmental and other public objectives are met. What is clearer, however, is that the experiment is continuing, having developed a self-perpetuating momentum of its own.

Acknowledgements This article is based on research conducted by the author towards the degree requirements of the Master of Environmental Law, University of Sydney.

References Abbott, M. & Cohen, B. (2011). Competition in urban water and sewage: the case of Sydney, Australia. Urban Policy and Research 29(2), 167–181.

Bayliss, K. (2011). A cup half full: the World Bank’s assessment of water privatisation. In: The Political Economy of Development: The World Bank. Bayliss, K., Fine, B. & Waeyenberge, E. (eds). Neoliberalism and Development Research, Pluto Press, London, UK. Carr, Y. (2007). Does Pt 3A of the environmental planning assessment Act 1979 (NSW) undermine the objects of that Act? Local Government Law Journal 12, 240.

Chong, J., Kazaglis, A. & Giurco, D. (2008). Cost Effectiveness Analysis of WELS: the Water Efficiency Labelling and Standards Scheme (prepared for Australian Government Department of the Environment, Water, Heritage and the Arts), Institute for Sustainable Futures, University of Technology, Sydney, Australia, pp. 1–91.

Chong, J., White, S. & Herriman, J. (2009). Review of Water Restrictions. Institute for Sustainable Futures, University of Technology, National Water Commission, Sydney.

Committee of Inquiry into National Competition Policy (1993). National Competition Policy Review Report. Australian Government Publishing Service, Canberra, Australia.

Council of Australian Governments (2004). Intergovernmental Agreement on a National Water Initiative. See http://www. environment.gov.au/water/australia/nwi/ (accessed 17 November 2013).

Council of Australian Governments (2008). COAG Work Program on Water – November 2008 – Agreed Actions See http:// www.environment.gov.au/water/australia/coag/work-program.html (accessed 17 November 2013).

Crosbie, F. (2009). Access to water infrastructure in New South Wales: An effective state access regime? Trade Practices Law Journal 17, 199.

Cubby, B. (2012). Plan to drill 150 gas wells across water catchment. The Sydney Morning Herald, 7 September 2012.

Department of Environment, Climate Change and Water (2010). 2010 Metropolitan Water Plan. NSW Government, Sydney.

EDO NSW (2012). Submission on A New Planning System for NSW – Green Paper.

François, D., Correljé, A. F. & Groenewegen, J. P. M. (2010). Cost recovery in the water supply and sanitation sector: a case of competing policy objectives? Utilities Policy 18, 135–141.

J. Chong / Water Policy 16 (2014) 1–18

17

Ghanem, R. & Ruddock, K. (2011). Are New South Wales’ planning laws climate-change ready? Environment and Planning Law Journal 28, 17.

Grafton, R. Q. & Kompas, T. (2006). Sydney Water: Pricing for Sustainability. The Australian National University.

Gray, J. & Gardner, A. (2008). Legal access to sewage and the ‘re-invention‘ of wastewater. The Australasian Journal of Natural Resources Law and Policy 12(2), 115.

Independent Pricing and Regulatory Tribunal (2012a). Sydney Water undertaking for access to drinking water network 2012.

IPART Reviews: Access.

Independent Pricing and Regulatory Tribunal (2012b). WICA Licence Register – Private Sector Licensing. IPART Licensing: Private sector.

Independent Pricing and Regulatory Tribunal (2012c). WICA Licensing: Private sector.

Independent Pricing and Regulatory Tribunal (2012d). NSW Planning System Review: IPART Submission on the Green Paper.

Independent Pricing and Regulatory Tribunal (2012e). Review of Prices for Sydney Water Corporation’s Water, Sewerage, Stormwater Drainage and Other Services: From 1 July 2012 to 30 June 2016.

Infrastructure NSW (2012). State Infrastructure Strategy 2012–2032.

Keating, J. (1992). The Drought Walked Through: A History of Water Shortage in Victoria. Department of Conservation & Environment, Victoria.

Keating, M. (2006). Water Pricing and its Availability. Speech to Centre for Economic Development of Australia, 29 November 2006.

Lowe, I. (2012). Queensland’s big step back from environmental assessment. The Conversation, 13 September 2012.

Lyster, R., Lipman, Z., Franklin, N., Wiffen, G. & Pearson, L. (2009). Environmental & Planning Law in New South Wales.

The Federation Press, Sydney, Australia.

Lyster, R., Lipman, Z., Franklin, N., Wiffen, G. & Pearson, L. (2012). Environmental & Planning Law in New South Wales.

The Federation Press, Sydney, Australia.

McKay, J. (2005). Water institutional reforms in Australia. Water Policy 7, 35–52.

National Competition Council (2009). Application for Certification of the NSW Water Industry Infrastructure Services Access Regime: Final Recommendation See http://www.ncc.gov.au/images/uploads/Final_Recommendation_WICA.pdf (accessed 17 November 2013).

National Water Commission (2010). Coal Seam Gas and Water: Position Statement See http://nwc.gov.au/_data/assets/ pdf_file/0003/9723/Coal_Seam_Gas.pdf (accessed 17 November 2013).

National Water Commission (2011). Urban Water in Australia: Future Directions. Commonwealth of Australia, Canberra.

New South Wales Parliament, Legislative Council (2006). A Sustainable Water Supply for Sydney.

NSW Department of Environment & Heritage (2012). Climate Change Fund Annual Report 2010–2011.

NSW Department of Finance & Services (2011). Retailer of Last Resort and Operator of Last Resort Arrangements under the Water Industry Competition Act 2006.

NSW Government (2006). Major Project Assessment – Kurnell Desalination Plant and Associated Infrastructure.

NSW Government (2011). Review of the Water Industry Competition Act. Water 4 Life, NSW Government.

NSW Government (2012). A New Planning System for NSW – Green Paper.

NSW Office of Water (2010). Climate Change and its Impacts on Water Supply and Demand in Sydney. Department of Environment, Climate Change and Water, State of New South Wales.

Parry, M., Canziani, O., Palutikof, J. P., van der Linden, P. J. & Hanson, C. E. (eds). (2007). Contribution of Working Group II to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change: Impacts, Adaptation and Vulnerability, Cambridge University Press, Cambridge, UK. Power, K. (2010). Recycled Water Use in Australia: Regulations, Guidelines and Validation Requirements for a National Approach. National Water Commission, Canberra, Australia.

Prince, R., Wright, T. & Cox, G. (2006). Independent Panel – Major Project: Kurnell Desalination Plant and Associated Infrastructure.

Productivity Commission (2005). Review of National Competition Policy Reforms, Report No. 33, Canberra, Australia.

Productivity Commission (2011). Australia’s Urban Water Sector, Report No. 55, Final Inquiry Report, Canberra, Australia.

Schott, K., Wilson, S. & Walkom, S. (2008). Urban water reform: an industry perspective. The Australian Economic Review 41(4), 413–419.

Smith, S. (2005). The Future of Water Supply, NSW Parliamentary Library Briefing Paper No. 4/2004 (2005).

Sydney Catchment Authority (2012a). Warragamba Dam Facts.

18

J. Chong / Water Policy 16 (2014) 1–18

Sydney Catchment Authority (2012b). Warragamba Dam Spill – 2 March 2012.

Sydney Catchment Authority (2012c). Water Storage and Supply Reports.

Sydney Water (2009). The History of Sydney Water: Sydney’s Water Supply and Wastewater Treatment.

Sydney Water (2010). Sydney Water Operating Licence 2010–2015.

Sydney Water (2011). Water Efficiency Annual Report 2010–2011.

Sydney Water (2012a). Successful Lease of the Sydney Desalination Plant Announced.

Sydney Water (2012b). Recycling.

Sydney Water Corporation (2009). Rosehill Camellia Recycled Water Project Contract Summary.

Thorpe, A. & Graham, K. (2009). Green buildings – are codes, standards and targets sufficient drivers of sustainability in New South Wales? Environment and Planning Law Journal, 26, 486.

University of Technology, Sydney (UTS), ‘Desalination Plant a Costly Mistake’. (Media Release, 13 September 2007.)

Water Services Association of Australia (2011). A Sustainable Future for the Urban Water Industry: Report Card 2011.

Water Services Association of Australia (2012a). Climate Change Adaptation and the Australian Urban Water Industry, Occasional Paper 27.

Water Services Association of Australia (2012b). Innovative Solutions from the Australian Urban Water Industry: A Series of Contemporary Case Studies.

White, S., Campbell, D., Giurco, D., Snelling, C., Kazaglis, A. & Fane, S. (2006). Review of the Sydney Metropolitan Water Plan. Institute for Sustainable Futures, University of Technology, Sydney.

White, S., Noble, K. & Chong, J. (2008). Reform, risk and reality: challenges and opportunities for Australian urban water management. The Australian Economic Review 41(4), 428–434.

Received 7 April 2013; accepted in revised form 18 July 2013. Available online 24 September 2013

Water Policy 16 (2014) 19–38

Why Eastern Himalayan countries should cooperate in transboundary water resource management Golam Rasul Theme Leader, Livelihoods, International Centre for Integrated Mountain Development, GPO Box 3226, Khumaltar, Lalitpur, Kathmandu, Nepal. E-mail: grasul@icimod.org

Abstract Bangladesh, Bhutan, India, and Nepal in the Eastern Himalayas are interconnected by the common river systems of the Ganges, Brahmaputra, and Meghna (GBM). The GBM basin is home to approximately 700 million people, comprising over 10% of the world’s population. The economy and environment of the region depend on water, but while the need for water is increasing, poor management and climate-related effects are making water supplies erratic. Upstream–downstream interdependencies necessitate developing a shared river system in an integrated manner through collaboration of the riparian countries. This paper examines the opportunities for, and potential benefits of, regional cooperation in water resource management. It suggests that the benefits can increase considerably when a regional (river basin) perspective is adopted that promotes optimum use of water resources for consumptive and non-consumptive use. Regional cooperation can bring additional economic, environmental, social, and political benefits through multi-purpose river projects, which help by storing monsoon water, mitigating the effects of floods and droughts, augmenting dry season river flows, expanding irrigation and navigation facilities, generating hydropower, and enhancing energy and environmental security. A broader framework to facilitate regional cooperation in transboundary rivers in the Eastern Himalayan region is suggested.

Keywords: Benefits of regional cooperation; Eastern Himalayas; Ganges, Brahmaputra and Meghna basin; Transboundary water resource management

1. Introduction The Eastern Himalayan countries of Bangladesh, Bhutan, India, and Nepal are interconnected by the common river systems of the Ganges (or Ganga), the Brahmaputra (known as Yarlung Tsangpo in China, and Jamuna in Bangladesh), and the Meghna. The combined river basins of these three river systems are often referred to as the Ganges–Brahmaputra–Meghna (GBM) basin. The GBM basin covers an area of 174.5 million hectares, across parts of the four Eastern Himalayan countries of Bangladesh and India (all three rivers), Bhutan (the Brahmaputra), and Nepal (the Ganges), and the Tibet doi: 10.2166/wp.2013.190 © IWA Publishing 2014

20

G. Rasul / Water Policy 16 (2014) 19–38

Autonomous Region (TAR) of China (the Brahmaputra and Ganges) (Figure 1, Table 1). With an average run-off of around 1,200 km3, it constitutes the third largest hydrological region in the world (Ahmad et al., 2001). Table 1 summarises some of the major features.

At present about 700 million people live in the GBM basin, comprising more than 10% of the world’s population. These river systems are rich in water, land, and forest resources. They provide fertile agricultural flood plains and feed into one of the most productive estuarine ecosystems in the world, the Sundarbans, which sustains the lives and livelihoods of millions. Despite such richness in natural resources, the region is one of the poorest in the world. The overall socioeconomic conditions of the Eastern Himalayan countries sharing the basin are low. The Human Development Indices range from 119 to 138 (IFAD, 2010; World Bank, 2010) and in terms of social indicators such as education, health, malnutrition, child mortality, access to safe drinking water, and energy, this region lies well behind other regions of the world (Biswas, 2004).

Water is crucial to both the economy and the environment, not only for the survival of humans and the ecosystem, but also for agricultural production, industrial development, electricity generation, transportation, environmental conservation, and regional peace and security (Brown & Lall, 2006). Rapid population growth, the pace of urbanisation, and economic development have increased the pressure on the demand for this finite resource (Ahmad et al., 2001; Bandyopadhyay & Ghosh, 2009).

Water resources in the region are distributed very unevenly over time and space. About 84% of the rainfall occurs between June and September, and 80% of the annual river flow takes place in the 4 months between July and October. Huge amounts of water during the monsoon period trigger floods

Fig. 1. The Ganges, Brahmaputra, and Meghna river basin (map by Sagar Ratna Bajracharya, ICIMOD).

G. Rasul / Water Policy 16 (2014) 19–38

21

Table 1. Salient features of the GBM basin.

Ganges–Brahmaputra–Meghna Drainage area (million ha) Population in millions (2001 census) Arable land (million ha)

China (TAR)

India

Bhutan

Nepal

Bangladesh

Total

32.6 3 Negligible

110.5 408 67.2

4.5 0.7 0.2

14.0 22 2.6

12.9 123 9.1

174.5 556.7 79.1

Sources: Rangachari & Verghese (2001); Brichieri-Colombi & Bradnock (2003); Rahaman (2009a).

and other hazards, whereas in the dry season the water is insufficient to meet the requirements for irrigation, navigation, and maintaining minimum environmental flow in the rivers (Figure 2).

Whereas the need for water has increased rapidly, water supplies have become more erratic as a result of both poor management and climatic effects. Upstream–downstream interdependencies and geographical linkages necessitate the development of a shared and integrated river system through collaboration among the riparian countries. Water is the most important natural resource in the region, and it is imperative that it be developed and managed in a rational, efficient, and equitable manner to act as an engine for socioeconomic development (Verghese, 1990; Verghese & Iyer, 1993; Verghese & Ramaswamy, 1993; Ahmad et al., 2001; Rangachari & Verghese, 2001; Ahmad, 2004; Biswas, 2004).

Fig. 2. Seasonal variation in river discharge in the Eastern Himalayas (drawn by Sagar Ratna Bajracharya, ICIMOD; source: Bandyopadhyay, 1995).

22

G. Rasul / Water Policy 16 (2014) 19–38

In the past, a fragmented and uncoordinated approach has been taken to transboundary water management that ignores the synergistic aspects of, and mutual benefits to be gained from, regional development (Crow et al., 1995; Rangachari & Verghese, 2001: 137). The importance of regional cooperation in transboundary water resource management has gained increasing attention in international forums such as The Hague Forum, the Bonn Conference, and the World Summit on Sustainable Development. However, so far, there has been little effort to show how regional cooperation can provide additional economic and social benefits to riparian countries.

This paper briefly examines the opportunities for cooperation and regional development through integrated water resource development and management in the GBM basin. It looks at the role of cooperation in enlarging the basket of benefits from shared water resources, and the contribution these benefits can make to economic and social development in the region.

2. Is regional cooperation in transboundary water resource management worthwhile?

Cooperation is fundamental to human existence, and philosophers argue that social cooperation is the key to improving the human condition (Rawls, 1971; Heath, 2006) and achieving ever higher goals (Gauthier, 1985; Heath, 2006; Bakker, 2009; Feiock et al., 2009). Water is fundamental to life and livelihood security, and human civilisation is strongly influenced by water availability and the way water is managed (Priscoli, 1998). Water does not follow national boundaries, thus international transboundary cooperation can enhance the capacity and resilience needed to respond to new challenges and opportunities (Priscoli, 1996; Priscoli & Wolf, 2009).

The value of the water resource lies not so much in its availability but in its efficient use and management. In transboundary water management, cooperation at basin level enlarges the planning space and horizon, increasing economies of scale and enabling actions that create an optimum effect (Sadoff & Grey, 2002; Grey & Sadoff, 2007). Moreover, knowledge and information can be shared and can highlight opportunities for better and more cost-effective solutions (Bakker, 2009). The greater the perceived and potential gains, the higher the possibility for cooperative management (Gardner et al., 1994).

In the GBM basin, the abundant water during the monsoon leads to hazards such as flooding and other natural disasters, but could be converted into benefits through innovative and collaborative efforts.

Storage of even a fraction of the huge flow of monsoon water in the Himalayan rivers to generate hydroelectricity, for example, could go a long way to bringing about economic transformation and development in the region (Ahmad et al., 2001; Rangachari & Verghese, 2001).

The benefits of regional cooperation in managing transboundary water resources lie in both consumptive and non-consumptive uses, which are linked to the ‘specific use value’ and the ‘system value’. The specific use value arises from a single, specific use of water, for example, the use of water for irrigation or hydroelectricity. The system value is the aggregate value that a unit of water generates as it moves through the river system before it is consumed or lost. For example, a cubic metre of water flowing through the Himalayan rivers from upstream Nepal, through India, and on to Bangladesh can generate hydropower at different sites and provide irrigation and drinking water. The system value is the sum of benefits accrued to all the countries involved through all uses of water – hydropower, irrigation, navigation, fisheries, and so forth – within a river basin. Regional cooperation using a river basin approach is essential to achieve the maximum system value of transboundary water resources (Crow & Singh, 2000,

G. Rasul / Water Policy 16 (2014) 19–38

23

2009; Grey & Sadoff, 2007; Rahaman, 2009a), and a move from single-purpose to multi-purpose planning is necessary to capture the full range of potential benefits (Barrow, 1998).

Cooperation in international rivers is also increasing in other parts of the world (Wolf, 1998; Uitto & Duda, 2002). Eleven European states collaborate in the maintenance and development of the water resources of the Danube, which drains a quarter of Europe. Coordinated efforts have developed the hydroelectric capacity of the river, expanded navigation, and opened the Rhine–Main–Danube waterway, which connects hundreds of inland ports and has turned the Danube into an artery for the continent ( Jansky et al., 2004). Efforts towards collaboration have also started in a number of river basins in Africa (see Howell & Allan, 1994; Klaphake & Scheumann, 2006; Wolf & Newton, n.d.). Lesotho–South Africa collaboration on the Orange river basin demonstrates that ‘even with power disparity, there is possibility for agreement over water resources through economic benefits’ (Wolf & Newton, n.d.). Four countries in the lower Mekong region – Cambodia, Laos, Thailand, and Vietnam – established the Mekong River Commission (MRC) in 1995 for sustainable development of the Mekong river basin. Joint efforts have been undertaken to develop hard and soft infrastructure to mitigate flood damage and develop the economic potential and benefits of the shared water resources ( Jacobs, 2002; Chheang, 2010). The Mekong sub-regional cooperation is now moving towards a Greater Mekong Subregion Economic Cooperation (Chheang, 2010). Experience gained from South East Asia, Central Asia, Africa, and Latin America suggests that, if properly orchestrated, cooperation in transboundary water management can increase economic, environmental, and social benefits for the riparian countries (Crow & Singh, 2000, 2009; Jacobs, 2002; Sadoff & Grey, 2002, 2005; Uitto & Duda, 2002; Jansky et al., 2004; Grey & Sadoff, 2007; Chheang, 2010).

Cooperation can take different forms, ranging from communication and notification to joint action and investment (Sadoff & Grey, 2002); similarly, there are different types and levels of incremental benefits – from economic and environmental to social and political. Sadoff & Grey (2002 and 2005) identified four types of benefits: benefits to the river; benefits from the river; a reduction in costs because of the river; and benefits beyond the river. Phillips et al. (2006) grouped them into three broad categories: economic, environmental, and security. Joint development of common water resources can provide additional benefits to all riparian countries by enhancing economies of scale, increasing planning horizons, bringing efficiency, reducing costs, and attracting the investments required for water resource development (Crow & Singh, 2000, 2009; Biswas, 2004; Grey & Sadoff, 2007; Rahaman, 2009a).

3. Potential benefits of regional cooperation in the Eastern Himalayan region The GBM basin is very rich in water resources, but this potential has remained largely untapped. This section briefly explores the potential for increased cooperation and associated benefits in the GBM basin through mitigation of flood damage, augmentation of dry season water availability, hydropower, energy security, water transport, and sociopolitical change.

3.1. Mitigating flood damage The high precipitation in the summer monsoon season (June to September/October) renders the Eastern Himalayan countries vulnerable to natural hazards such as floods and landslides. The Ganges–Brahmaputra

24

G. Rasul / Water Policy 16 (2014) 19–38

Table 2. Economic and social costs of natural disasters in the GBM countries.

People affected per annum on average Mortality 1971– 2008 (Average no. of people dying annually) Bangladesh 5,673 India 2,497 Nepal 137

Economic loss (annual average from 1971 to 2008)

Drought (’000)

Floods and storms (’000)

Share of population (%)

Droughts (million US$)

Floods and storms (million US$)

Largest loss per event (% of GDP)

658 25,294 121

8,751 22,314 87

9.1 7.2 2.0

0 61.6 0.3

445.6 1,055.4 25.8

9.8 2.5 24.6

Source: World Bank (2010).

basin is one of the most flood-prone regions in the world (Ahmad et al., 2001; Dixit, 2003; Dhar & Nandargi, 2003) with flooding a recurring phenomenon in Bangladesh, India, and Nepal.

Table 2 shows the huge cost of flood damage. The loss of human life is highest in Bangladesh (on average around 6,000 people per year) and the number of people affected by floods is highest in India (more than 22 million per year). About one-tenth of the population of Bangladesh is affected by floods or drought annually. Bihar, Assam, and other north Indian states are highly vulnerable to floods. In Bihar alone, the annual average economic damage is to the tune of 458 million Indian rupees (Table 3). The frequency and severity of floods, and the population affected, have been increasing in all the GBM countries (Mirza et al., 2001, 2003; Dhar & Nandargi, 2003), which has placed enormous constraints on the development potential of these countries.

While floods cannot be completely avoided, the damage can be minimised through the joint efforts of governments and those living in the major river basins. The lead time for flood forecasting can be increased substantially through exchange of real-time data on river flow from upstream areas of the basins. Downstream states also depend on the upstream neighbours for data and warnings about below average precipitation or drought. Flood warnings can provide a grace period – 2 to 14 days depending on the size of the river – between upstream rainfall and the augmentation in river flow downstream (Zawahri, 2008). This grace period enables arrangements to be made such as evacuating people and transporting food, water, and medicine to the region. Sharing of real-time hydro-meteorological data is crucial for better flood forecasting and reduction of flood damage. However, notwithstanding a number of initiatives made by International Centre for Integrated Mountain Development (ICIMOD) and South Asian Association for Regional Cooperation (SAARC), real-time data sharing among the South Asian countries remains limited (Thakkar, 2006). Billions of dollars could be saved by regional cooperation in flood mitigation (Ahmad, 2004).

Table 3. Average annual damage due to floods in Bihar.

Period

Total area affected in thousand ha

Crop area affected in thousand ha

Average annual economic damage in thousand Indian rupees (at constant prices)

1950–65 1966–70 1971–78

881 1,082 2,130

443 585 885

86,192 118,408 458,857

Source: Thakkar (2006).

G. Rasul / Water Policy 16 (2014) 19–38

25

3.2. Augmenting dry season water availability The fundamental problem with water resource management in the GBM basin lies in the seasonal concentration of rainfall and spatial variation in its distribution, as well as unreliability in water supplies.

These characteristics of water availability mean that water should be stored when it is abundant and redistributed when and where required within a framework of regional understanding and cooperation.

Perusal of the literature on potential sites for storage reservoirs in India and Nepal reveals that there is great potential for storage of monsoon water in the GBM basin; summary details and key references are given in Tables 4 and 5. This could provide substantial hydropower benefits and support the process of industrialisation in the region, provide clean energy, and deliver other socioeconomic and environmental benefits (Vaidya et al., 2010). The proposed Sunkoshi dam in Bhutan could also store water for hydropower (4,000 MW) and augment flow in the dry season (Ahmad, 2004). Hydropower storage projects often have better economic performance than irrigation projects; multi-purpose water storage projects providing both services can offer energy benefits as well as improve food security in an economically and financially feasible manner.

Table 4. Potential storage reservoirs identified in the Ganges basin (in Nepal).

Name of basin and potential reservoir Kosi Basin Sapta Koshi Tamor 1 Sunkoshi 2 & 3 Gandaki Basin Andhi Khola Kali-Gandaki 1 & 2 Seti (Central) 1 Karnali Basin Karnali (Chisapani) Bheri 3 & 4 Karnali IB West Seti 1& 6 Mahakali Basin Mahakali (Pancheswar) Poornagiri Other Basins Trisuli Ganga Kankai Bagmati West Rapti (at Bhalubang) Kamala Sharada TOTAL

Catchment area (km2)

Dam height (m)

Gross storage capacity (million m3)

Live storage capacity (million m3)

Hydropower generation capacity (MW)

54,100 5,085 15,916

269 153 306

13,450 1,890 5,520

9,370 760 3,590

3,300 696 1,646

420 20,490 2,740

130 404 165

940 12,000 4,000

800 8,600 1,900

180 2,260 320

42,890 17,231 14,417 10,668

270 540 294 455

28,200 25,800 7,400 3,740

16,200 14,700 4,200 2,550

10,800 3,928 ** 1,440

12,100

315

12,260

6,560

6,480

15,000

156

3,400

1,240

1,000

16,260 1,190 2,700 3,680

230 85 117 93

11,000 1,130 3,000 1,390

6,700 730 2,100 970

** 60 180 107

1,450 860 237,197

51 85

713 260 136,093

493 220 81,683

32 49 32,478

**Indicates information is not available.

Sources: Bandyopadhyay & Gyawali (1994); Sharma (1997: 110–112); Rangachari & Verghese (2001: 113); Rahaman (2009b).

26

G. Rasul / Water Policy 16 (2014) 19–38

Table 5. Potential storage reservoirs identified in the Brahmaputra and Meghna basins in Bhutan and India.

Reservoir

Catchment area (km2)

Dam height (m)

Gross storage capacity (million m3)

Live storage capacity (million m3)

Hydropower generation capacity (MW)

Lohit Dibang Subansiri Jia Bhareli Dihang Tipaimukh Pagladiya Sankosh

19,000 10,350 27,000 9,980 247,500 12,758 5,550 10,820

296 236 257 211 296 161 26 265

5,160 6,200 14,000 6,500 47,000 15,900 NA 6,325

3,310 4,700 10,000 5,100 35,500 9,000 NA 4,456

8,600 4,500 4,800 2,000 7,600 1,500 3 4,000

Sources: Rangachari & Verghese (2001: 118); Rahaman & Varis (2009); Sharma (1997); Rao (2006); Sharma (2004). See also Bandyopadhyay & Gyawali (1994); Sankosh data from THDC India Ltd, http://thdc.gov.in/Projects/english/Scripts/ Prj_Introduction.aspx?vid=171 (accessed 11 April 2011).

The seasonal and spatial variation in water supplies could be tackled to some extent by storing monsoon rainwater in lakes and reservoirs through multi-purpose river development projects. This would help to meet the demand for water in the dry season and to control floods in the wet season. It could also lead to substantial economic and social benefits by generating hydroelectricity, facilitating irrigation, improving navigation, and maintaining the ecological balance of the region. The agricultural sector uses most of the water and the main crops (such as rice and wheat) are heavily dependent on freshwater: dependence on these water-thirsty crops has placed the GBM basin in a difficult position in terms of agricultural production and food security.

Water supplies need to be augmented in the dry season through storage of monsoon water, but this requires huge investments, sound technologies, and an appropriate scale of operations: these can only be realised through a basin-wide approach. Regional cooperation can be instrumental in harnessing monsoon water for more productive use in the dry season.

3.3. Hydropower and clean energy Abundant rain-fed and snow-fed water resources and topography with a favourable relief for hydropower generation provide an excellent opportunity for generating an enormous amount of hydropower.

The energy requirements of the region could be met and the surplus exported. For example, the theoretical hydropower potential of glacial rivers in Nepal is estimated to be 83,000 MW, in Bhutan 21,000 MW, and in north-east India about 58,971 MW. It is estimated that the GBM river systems have about 200,000 MW of hydropower potential, of which half or more is considered to be feasible for harnessing (Chalise et al., 2003).

Many projects proposed in Nepal have a large water storage potential, for example, the Mahakali (Pancheswar), Sapta Koshi, and Karnali (Chisapani) (Table 4). Several multi-purpose projects with large reservoir storage capacities have also been identified in India in the Brahmaputra and Meghna basins (Table 5).

Despite the potential, per capita energy consumption in the GBM basin is low. There are persistent shortages of energy supplies in all the basin countries, apart from Bhutan. Whereas the average per capita electricity consumption in the world in 2001 was 2,159 kWh, in Bangladesh it was a mere 94 kWh, in India 365 kWh, and in Nepal 336 kWh. Less than 60% of the region’s population have direct access to electricity, or other modern forms of energy, either because of lack of supply or due to high prices that poor people cannot afford.

G. Rasul / Water Policy 16 (2014) 19–38

27

Demand for power will increase because of the rapid pace of industrialisation, urbanisation, and increased demand for food for ever-growing populations. Availability of an uninterrupted supply of power is a major challenge to sustainable development in the region. Energy is needed not only to sustain the region’s growth, but also to improve socioeconomic conditions and human development.

Shortage of energy is a severe constraint to economic growth, poverty alleviation, and improving standards of living (Lama, 2000; Dhungel, 2008).

At present, cross-border energy trade in the region is minimal and exists only at the bilateral level.

Hydropower trade between India and Bhutan and between India and Nepal is taking place on a small scale. There is vast potential for joint development of the hydropower sector to support accelerating economic growth, reduce poverty, and enhance human development (Lama, 2000; Srivastava & Misra, 2007; Dhungel, 2008; Malla, 2008; Crow & Singh, 2009; Karki & Vaidya, 2009). Realisation of hydropower potential in Nepal and Bhutan can help to meet the growing demand for power and replace, to some extent, existing biomass and fossil fuel energy as well as provide additional environmental benefits to all through carbon offset (Malla, 2008).

Harnessing the huge untapped hydropower resources in Nepal, Bhutan, and north-east India could transform this region from poor to prosperous. While developed countries use up to 70% of their hydropower potential (for example, Switzerland 70% and Norway 68%), the Eastern Himalayan countries use negligible amounts: for example, Nepal 1.7%, Bhutan 2%, and north-east India less than 2%.

The principal impediments to the exploitation of hydropower potential include lack of financial resources to develop the necessary infrastructure and technical capacity, along with the risks associated with single buyers. Regional cooperation can help to overcome these impediments by harnessing the necessary financial resources, suitable technologies, guaranteed markets, and institutional mechanisms for sharing the costs and benefits of joint efforts. This would facilitate the exploitation of untapped hydropower resources for the benefit of the entire region (Lama, 2000; Dhungel, 2008). Initial efforts in this direction have already begun. An intergovernmental task force of senior government officials from the water and energy sectors of Bangladesh, Bhutan, India, and Nepal has been formed and a meeting was held in New Delhi in 2008. The task force agreed to develop regional hydropower resources within the framework of sub-regional cooperation and promote a ‘South Asia Growth Quadrangle’ initiative (Malla, 2008). An Energy Research Centre has been established in New Delhi under the Bay of Bengal Initiative for Multi-Sectoral Technical and Economic Cooperation (BIMSTEC) to promote regional grid development, and a Memorandum of Understanding has been signed between Nepal, India, and the World Bank to develop cross-border transmission lines.

3.4. Regional energy security Establishment of an inter-country power grid could facilitate the integration of different power systems and the export of excess hydropower from Nepal and Bhutan to India and Bangladesh. A grid link with Bangladesh would substantially shorten the transmission distance between north-east India and West Bengal compared to the route via Siliguri.

Cross-border grid interconnections are increasing all over the world. Such interconnections already exist in North America, Europe, and southern Africa; they include the Nord Pool (Denmark, Finland, Norway, and Sweden) and the South African Power Pool (12 countries). The establishment of power grid networks and introduction of power trading can promote the use of untapped water resources and increase energy security in the region, a factor crucial for economic development and improving living standards.

28

G. Rasul / Water Policy 16 (2014) 19–38

3.5. Improving regional water transport The Ganges, the Brahmaputra, and the Meghna rivers flow into Bangladesh from three directions and merge into a single outlet that constitutes a vast water network; this provides an opportunity to develop an integrated water transport system. Two countries in the basin, Bhutan and Nepal, are landlocked, and this is an obstacle to their industrial growth and overall economic development. Currently only 130,000 tonnes of freight are shipped out by boat along the Karnali, Gandaki, and Koshi basins in Nepal (Ahmad, 2004). Most of the freight from Nepal is transported by road, which incurs high transport costs. It costs NRP66 million (approximately US$943,000) to transport 100,000 tonnes of freight from Kolkata to Kathmandu, whereas transporting by water would cost only NRP6.7 million (approximately US$96,000), one-tenth only (Upreti, 2006). Access to the Bay of Bengal via the Brahmaputra could also help Bhutan in expanding regional and international trade.

There is a potential for developing water transportation: rivers like the Karnali, Gandaki, and Koshi in Nepal; the Ganges and Brahmaputra in India; and the Brahmaputra and Meghna in Bangladesh support inland water transport facilities. It is technically feasible for Nepal and Bhutan to gain direct access to the sea. Boats could travel from the Hooghly (in India) along the Ganges (in Bangladesh) via Farakka and Kanpur in India to several points in Nepal, such as Bhardaha on the river Koshi, Narayanghat on the Gandaki, and Chisapani on the Karnali (Malla et al., 2001; Sainju & Shrestha, 2002; Upreti, 2006).

Water transport is cost-effective compared to other forms of transport, particularly for bulk commodities. A navigational link to the Ganges and Brahmaputra could change the economy of Nepal and Bhutan through increased export and import and by triggering industrial development. Regional cooperation can facilitate development of a regional water transport network to foster integration of regional economies, better use of resources, and economic growth.

3.6. Political benefits Transboundary water resources have become a contentious issue in the GBM region, as in other parts of the world. With the right perspective, transboundary water resources can become a source of understanding, of regional cooperation, and peace (Libiszewski, 1995; Wolf, 1998). Through cooperative development of water resources, current tensions between neighbouring countries can be reconciled to a great extent, and this would bring political benefits to all the countries involved through building trust and increasing regional security and economic growth. The issue of transboundary water has been a hindrance to promotion of trust and cooperation, but development of water resources brings new opportunities and stimulates cooperation through sharing data and knowledge, flood management, inland water transportation, energy cooperation, and industrialisation, along with increased understanding, interdependencies, and confidence and peace building (Rasul & Manandhar, 2009).

4. Challenges and opportunities for cooperative development of water resources in the Eastern Himalayan region 4.1. Challenges Although the potential benefits of collaborative development of the transboundary water resources in the Eastern Himalayan region are huge, there are a number of challenges that must be met and impediments

G. Rasul / Water Policy 16 (2014) 19–38

29

to overcome before these benefits can be realised. This section briefly examines the key factors constraining the collaborative development of transboundary water resources in the Eastern Himalayan region.

4.1.1. Hydropolitics. Hydropolitics among the Himalayan countries can be traced back to the 19th century with the signing of the Sugauli Treaty in 1816 between the British East India Company and Nepal, which delimited the boundary along the Mahakali river in Nepal. Disputes over water intensified with the partition of India in 1947. When Pakistan became a separate state, the intra-nation water issue became an international transboundary water issue between India and Pakistan and, subsequently, India and Bangladesh, involving upstream–downstream rights and responsibilities.

The Ganges water treaty between India and Bangladesh was signed in 1996, but little progress has been made in augmenting dry season water availability in the Ganges due to disagreement between the countries on mechanisms of augmentation. While India wants to divert water from the Brahmaputra to the Ganges to increase dry season water availability, Bangladesh prefers an agreement with Nepal to store water in the Nepal Himalayas (Crow & Singh, 2000). The Tipaimukh dam on the Barak river proposed by India and equitable sharing of water from the Teesta and other transboundary rivers have remained constant sources of tension between Bangladesh and India.

Similarly, there is strong disagreement between India and Nepal on the mode of development of water resources, sharing of costs and benefits, and water transportation and navigation. There is a general feeling that costs and benefits were shared unequally between India and Nepal in earlier water resource development projects, and initiatives taken by India and Nepal to develop water resources such as the Koshi, the Karnali, and the Mahakali have not achieved their goals (Upreti, 2006; Thapa, 2013).

Although the relationship between Bhutan and India is relatively good, Bhutan aims to expand and diversify the hydropower market to make it more competitive. The deterioration of water quality in Assam, India, due to mining activities upstream in Bhutan, also raises some concerns. Similarly, there is growing concern across the region about the potential diversion of water by China from the upper Brahmaputra and the potential ecological threat downstream due to unsustainable exploitation of the natural resources of the Tibetan Plateau. All these differences have created an atmosphere of mistrust in the region, which is a major impediment to collaborative development of transboundary water resources (Upreti, 2006; Bandyopadhyay & Ghosh, 2009). Accommodating diverse interests and overcoming mistrust remain the most critical challenges in enhancing regional cooperation in the Eastern Himalayan region.

4.1.2. Conventional perspectives. For optimal development, a river basin needs to be managed through an integrated basin-wide approach (Verghese, 1990; Ahmad et al., 2001; Biswas, 2004; Upreti, 2006; Crow & Singh, 2009). Transboundary water resources in the Eastern Himalayan region are generally seen from a national perspective, with a focus on problems of sharing water rather than expanding the benefits through joint resource development. This narrow perspective often leads to bilateralism and encourages unilateral and fragmented decisions, with transboundary water resource development seen as a ‘zero sum game’ in which the gains of one country must mean losses for another, and negotiations become deadlocked (Crow & Singh, 2009). Moreover, transboundary water resource management has become a purely diplomatic matter, with little space for civil society, nongovernmental organisations (NGOs), private sector, and other stakeholders, who are directly and indirectly involved in water management. This is another major obstacle to cooperative development of transboundary water resources.

30

G. Rasul / Water Policy 16 (2014) 19–38

4.1.3. Inadequate knowledge. Reliable data and information can advance knowledge and provide a basis for shared understanding of water problems as a foundation for negotiated management and policy solutions (Timmerman & Langaas, 2005; Crow & Singh, 2009). But there is a dearth of reliable data and information and limited cross-border exchange of information (Shahjahan, 2008). In the absence of such information, different groups interpret issues from different perspectives, which misleads those involved, fuels suspicion among the co-riparian countries and hinders cooperation (Iyer, 2006; Shahjahan, 2008; Bandyopadhyay & Ghosh, 2009).

4.1.4. Inadequate institutional arrangements and capacities. To facilitate cooperation it is critical to have adaptive institutions (Molden et al., 2005) that are able to recognise both common interests and differences of co-riparian countries and provide a safe space for meeting and discussion (Priscoli, 2001). There is no institution in the Eastern Himalayan region able to bring riparian countries together on a regular basis to facilitate dialogue and development of shared understanding. Existing mechanisms between India and Bangladesh, in the form of the Indo-Bangladesh Joint Rivers Commission, and between India and Nepal on the Mahakali only work bilaterally and have a limited capacity to generate impartial data and information, mediate conflicts, or foster basin-wide cooperation.

4.2. Opportunities Although there are challenges, the opportunities for collaboration are also growing as civil society and other non-state actors emerge, and new forces for cooperation grow.

4.2.1. Driving forces for cooperation. Civil society, NGOs, scientific communities, and other new forces for regional cooperation are emerging. Ensuring food, water, and energy for growing populations, protecting the environment, promoting socioeconomic development, and alleviating poverty have become common concerns for all the countries in the region. There is a growing realisation and demand for collaborative development of transboundary water resources to address common concerns. Scientists and civil society are increasingly working together to this end. The Abu Dhabi dialogue is one example, where government representatives and experts from Afghanistan, Bangladesh, Bhutan, China, India, Nepal, and Pakistan come together to exchange views with the aim of fostering cooperation on water in the greater Himalayan region. Civil society organisations and other non-state actors are actively working to make transboundary water governance and negotiations more inclusive and create a space for civil society and scientific communities (Bandyopadhyay, 1995; Ahmad et al., 2001; Rangachari & Verghese, 2001; Upreti, 2006), which has opened up new opportunities for collaborative development of transboundary water resources. Global warming and the resultant impacts on the flows of the Himalayan rivers, shifting monsoon patterns, increased uncertainty of water availability, and increased frequency of extreme events, as well as growing interdependence for energy security, water security, agricultural sustainability, and food security, are making regional cooperation between upstream and downstream countries an imperative to enable better adaptation, enhance resilience, and provide the capacity to respond to new challenges, and exploit opportunities.

4.2.2. Changing policy environment. The policy environment in the region is becoming more favourable for cooperative development of transboundary water resources. The cooperation between Bhutan and India on hydropower development is a good example. Similarly, the signing of the Mahakali treaty between India

G. Rasul / Water Policy 16 (2014) 19–38

31

and Nepal and the treaty for sharing Ganges water between Bangladesh and India have opened up opportunities for collaboration in regional water resource development. The declarations of the South Asian Association for Regional Cooperation (SAARC) summit in Male in 1997 and in Colombo in 1998 endorsed sub-regional cooperation among the SAARC countries. It is now important to move from bilateralism to multilateralism for optimal use of water resources (Ahmad et al., 2001; Biswas, 2004).

4.2.3. Positive shift. There has been a positive shift in the region towards cooperative water management.

Although slow, efforts are ongoing to resolve differences over water issues. The Bangladesh–India Ganges Treaty states that both countries will work together to augment the river flows in the upstream and share such waters. This provision opens a path for regional cooperation to harness the water resources of the GBM basin. Efforts are also ongoing in the Koshi basin. The Prime Ministers of India and Bangladesh signed a framework agreement on cooperation in 2011 calling for enhanced efforts in cooperative water management, which clearly indicates a positive shift. The SAARC summit and the Climate Summit for a Living Himalayas in 2011 also called for enhanced transboundary cooperation. Bangladesh has agreed in principle to allow transit from Bhutan, India, and Nepal to use the Mongla and Chittagong ports. A waterway transit for Bhutan to Mongla port is under consideration. A feasibility study is under way on developing a navigable canal waterway extending from Chatra in Nepal to Kursella in Bihar, India for linking the Koshi navigational canal to the National Waterway No. 1 of India (Thapa, 2011).

4.2.4. Cross-border electricity grid connection. The Bay of Bengal Initiative for Multi-Sectoral Technical and Economic Cooperation (BIMSTEC), which has Bangladesh, Bhutan, India, and Nepal as members, established an Energy Centre in Delhi in 2011 to coordinate, facilitate, and strengthen cooperation in the energy sector in the region. This will help to develop the cross-border energy trade and a regional market for hydropower from Nepal and Bhutan. India and Bhutan already have grid connections. Work is progressing on a Bangladesh–India grid connection, and trade in power between India and Bangladesh is expected to start in July 2013 (Financial Express, 1 February 2013). India has agreed to export 250 MW of electricity to Bangladesh. Cross-border 400 kV electricity transmission lines between Nepal and India are under construction. The physical and institutional basis for energy cooperation is growing, which will provide a further basis for cooperation in transboundary water.

5. Discussion and conclusions The GBM basin is unique in terms of water resources, seasonality, upstream–downstream linkages, and socioeconomic conditions. Water is abundant in the monsoon season, creating hazards, but scarce in the dry season. The challenge the region faces is how best to use the available water resources to facilitate rapid economic and social development and reduce the adverse impacts of floods, droughts, and other water-related stresses and hazards. River basins are naturally occurring landscape units and thus provide logical units for water management. Integrated river basin management is essential for the sustainable development of water resources, but it is only possible through collaboration.

Management of transboundary water resources has evolved gradually in accordance with changing societal needs, increasing demands for water resources, growing water-related stress, and climatic effects. Institutional innovations, technological progress, and growing knowledge have triggered a more collaborative development of water resources than before and a move from single-purpose to

32

G. Rasul / Water Policy 16 (2014) 19–38

multi-purpose development to meet diverse socioeconomic and environmental needs (Biswas, 2004).

Collaborative efforts can unleash the potential of industrial growth and economic development and improve the socioeconomic conditions of the countries in the region. Development of water resources in the GBM basin can transform lives through improved flood management, hydropower development, augmentation of dry season flow, irrigation, and navigation. There is growing evidence that a regional perspective could increase benefits considerably, leading to a win–win situation for all the countries involved (Table 6).

The GBM river systems cut across several countries, and solutions to floods, dry season water shortages, and economic growth lie in collaborative efforts. The river systems need to be managed and developed within an integrated regional framework for the region’s future development. Water provides the most appropriate entry point for cooperative regional development. In the past, unilateral and bilateral approaches have hindered maximising the benefits of water resources management. Moving to a basin-wide approach and from single-objective to multi-objective development will help maximise benefits throughout the basin. Upstream hydropower plants can also support downstream irrigation and flood control, benefiting both upstream and downstream countries.

Steps towards resolving water issues will help improve the political climate in the region: moving from hydropolitics to hydro-economics is in the best interests of poverty alleviation and sustainable development, and an improved political climate could help forge cooperation in other areas. All the countries in the region have a stake in the economic prosperity and political stability of their neighbouring countries, and there are significant opportunity costs from delaying regional cooperation, for example, the annual cost of floods. Collaboration can bring about improvement in regional navigational systems and reduce transport costs. Strong political will and commitment, negotiation, financial resources, and meaningful technical cooperation will lead to successful transboundary cooperation in water management.

6. Towards a framework for cooperative development of transboundary water resources in the Eastern Himalayan region A framework is needed for the cooperative development of transboundary water resources to transform conflict into cooperation. Development of a comprehensive framework is beyond the scope of this paper; however, some of the major points for a conceptual and explorative framework are outlined in this section, with the aim of stimulating critical thinking rather than providing definite answers.

• Promote a multi-purpose basin-wide approach for optimum use of Himalayan water resources in an integrated manner to create dynamic complementarities that transcend national boundaries in order to prevent and mitigate harmful impacts, such as floods and drought, and to protect the environment and public health. Cooperation could be at bilateral and multilateral levels in parallel, depending on the nature and local conditions (Chen et al., 2013). The starting point could be cooperation in flood control, as flooding is a common issue for all the countries of the region.

• Shift the focus from sharing water to sharing the benefits of water. To overcome the impasse in negotiations on transboundary water management, the focus could be shifted to sharing the benefits derived from the common rivers to provide a more flexible framework for negotiation and increase the possibilities for cooperation (Sadoff & Grey, 2002; Dinar, 2008). Focusing on the benefits provided incentivises collaborative management in other initiatives (Rahaman & Varis, 2009).

G. Rasul / Water Policy 16 (2014) 19–38

33

Table 6. Benefits of collaborative management of transboundary water resources.

Country

Economic

Environmental

Social

Political

Bangladesh Reduced floods and flood damage, dry season water, increased navigation, increased regional trade

Bhutan

India

Nepal

Maintenance of minimum Improvement in public Better relationships river flows and health, reduced social with neighbouring preservation of ecosystems, tension countries, less control of saline intrusion, political tension reduced river bank erosion, access to hydrometeorological data Increased power Electricity market expanded, More sustainable use of water Employment, low-cost through move power to villages, and other natural resources enhanced bargaining from bilateral to improved living through access to regional power, access to sea regional standards and welleconomy, technology, and through waterways cooperation being, capacity science development through management of mega projects Better relationships More surface irrigation water Reduction of environmental Reduced flood damage, with neighbouring increased well-being, pollution, reduced floods and groundwater pumping, countries, less increased responsibility and flood damage flood control, inexpensive tension, less towards neighbours power, increased life of coal military and hydrocarbon reserves, expenditure, sale of hydropower greater recognition machinery and equipment, in international use of skilled labour, forums, better reduction in industrial loadreputation, stable shedding, reduction in oil and prosperous imports, foreign exchange neighbours savings, increased trade with Tibet Autonomous Region of China Increased power Substitution of fuel-wood by Employment, low-cost Increased resource value through move power for villages, electricity leading to through royalties, reduced from bilateral to improved health, reduction of pressure on submergence and regional reduced time spent forests and improved water environmental costs, cooperation, gathering wood, and air quality irrigation water, reduced overcoming the reduced water-related flood damage, reduction in constraints of a diseases, capacity industrial load-shedding, landlocked state development through higher price for and management of mega income from electricity projects (through increased bargaining power resulting from downstream benefits of flood moderation and increased dry season water flow), inland and sea navigation benefits, reduced transport costs, increased regional trade (Continued.)

34

G. Rasul / Water Policy 16 (2014) 19–38

Table 6. (Continued.)

Country

Economic

Environmental

Social

Political

China

Use of untapped water resources of Yarlung Tsangpo (Brahmaputra) river, increased trade with South Asian countries, potential access to Tibet Autonomous Region of China through proposed Koshi canal

More sustainable use of water Improved living standards Improved and other natural resources and economic wellrelationship with through access to regional being, less social South Asian economy, technology, and tension, increased countries, better science responsibility for reputation, stable neighbours and prosperous neighbours

• Link water-sector strategies with broader national and regional development goals; formulate •

•

• •

•

appropriate policies and actions to promote management of water as an economic, environmental, and social resource; and shift the focus from hydro-diplomacy to a hydro-economic perspective.

Build trust. A concerted effort is required to build trust and confidence so that negotiations and discussions can start. An alternative way for dispute resolution suggested by Priscoli (1996) could be explored to find creative solutions for building trust among the riparian countries. Mistrust is partly due to poor understanding of the benefits and costs of collaborative development. There is a need to promote joint research on transboundary water management issues; on integrated hydrologicaleconomic modelling to explore alternative development options; and on quantification of the benefits and costs of regional cooperation, and the social, economic, and environmental costs of noncooperation. Social awareness needs to be created of the benefits of cooperation and of the costs of non-cooperation. Mechanisms for fair sharing of the costs and benefits from the cooperative development of transboundary water resources need to be developed. Informal intergovernmental cooperation processes should be fostered and used to convince policy-makers of the need to share information.

Facilitate multi-track diplomacy. Efforts should be made to facilitate cross-border exchange among civil society organisations, NGOs, academic and scientific communities, and government officials to deepen relationships and foster informal intergovernmental cooperation processes (Crow & Singh, 2009).

Undertake joint research by the Eastern Himalayan countries to produce credible information and knowledge, explore development potential and options, assess risks, costs, and benefits of cooperative management to support sound decision-making.

Establish a mechanism and institutional arrangements to coordinate, facilitate, and strengthen cooperation in water, hydropower, and flood management in the GBM basin, which is able to bring together scientists, development practitioners, and policy-makers to share, discuss, negotiate, and develop a common understanding. Developing such an institution is time-consuming, and a regional organisation like ICIMOD, which is already working on regional water issues, could be entrusted to provide such a platform and assist national governments to develop an appropriate legal and institutional framework for establishing a regional collaboration.

Explore joint projects for development of transboundary water resources for short-, medium-, and long-term measures based on mutual understanding and priorities. Joint action plans could focus on both structural and non-structural measures.

G. Rasul / Water Policy 16 (2014) 19–38

35

• Establish a basin-wide data bank and system for timely sharing of meteorological, hydrological, economic, and environmental data and information among the countries sharing the basin. Transboundary technical teams and committees are needed to monitor river behaviour and determine strategies.

Acknowledgements My sincere thanks to the anonymous reviewers for their helpful comments and suggestions. I have benefitted from discussions with, and comments from, my colleagues David Molden, Gopal Rawat and Arun B. Shrestha. Valuable comments were received from Jacqueline A. Ashby (CIAT), and Claudia Sadoff (World Bank) on the earlier version of this paper. Prem Manandhar helped me in collecting some information. This study was partially supported by the Department for International Development (DFID) of the United Kingdom through core funding to the International Centre for Integrated Mountain Development (ICIMOD) and the Koshi Basin Programme of ICIMOD funded by the Australian Agency for International Development (AusAID) of Australia. The views expressed are mine and do not necessarily reflect those of ICIMOD or other organisations mentioned above.

References Ahmad, Z. U. (2004). Water development potential within a basin-wide approach: Ganges-Brahmaputra-Meghna (GBM) issues. In: Water as a Focus for Regional Development. Biswas, A. K., Unver, O. & Tortajada, C. (eds). Oxford University Press, New Delhi, pp. 83–113.

Ahmad, Q. K., Biswas, A. K., Rangachari, R. & Sainju, M. M. (eds) (2001). Ganga-Brahmaputra-Meghna Region: A Framework for Sustainable Development. The University Press Ltd, Dhaka.

Bakker, M. H. N. (2009). Transboundary river floods and institutional capacity. Journal of the American Water Resources Association 45(3), 553–566.

Bandyopadhyay, J. (1995). Water management in the Ganges-Brahmaputra basin. Water Resources Development 11(4), 411–442.

Bandyopadhyay, J. & Ghosh, N. (2009). Holistic engineering and hydro-diplomacy in the Ganges-Brahmaputra-Meghna basin.

Economic and Political Weekly 64(45), 50–60.

Bandyopadhyay, J. & Gyawali, D. (1994). Himalayan water resources: ecological and political aspect of management.

Mountain Research and Development 14(1), 1–24.

Barrow, C. J. (1998). River basin development planning and management: a critical review. World Development 26(1), 171–186.

Biswas, A. K. (2004). Water and regional development. In: Water as a Focus for Regional Development. Biswas, A. K., Unver, O. & Tortajada, C. (eds). Oxford University Press, New Delhi, pp. 1–13.

Brichieri-Colombi, S. & Bradnock, R. W. (2003). Geopolitics, water and development in South Asia: cooperative development in the Ganges-Brahmaputra. The Geographical Journal 169(1), 43–64.

Brown, C. & Lall, U. (2006). Water and economic development: the role of variability and a framework for resilience. Natural Resources Forum 30, 306–317.

Chalise, S. R., Kansakar, S. R., Rees, G., Croker, K. & Zaidman, M. (2003). Management of water resources and low flow estimation for the Himalayan basins of Nepal. Journal of Hydrology 282(1–4), 25–35.

Chen, S., Pernetta, J. C. & Duda, A. M. (2013). Towards a new paradigm for transboundary water governance: implementing regional frameworks through local actions. Ocean and Coastal Management, 85(Part B), 244–256. http://www.sciencedirect. com/science/article/pii/S096456911200302X#FCANote (accessed 22 October 2013).

Chheang, V. (2010). Environmental and economic cooperation in the Mekong region. Asia Europe Journal 8, 359–368.

Crow, B., Lindquist, A. & Wilson, D. (1995). Sharing the Ganges: The Politics and Technology of River Development. The University Press Ltd, Dhaka, Bangladesh.

36

G. Rasul / Water Policy 16 (2014) 19–38

Crow, B. & Singh, N. (2000). Impediments and innovation in international rivers: the waters of South Asia. World Development 28(11), 1907–1925.

Crow, B. & Singh, N. (2009). The management of international rivers as demands grow and supplies tighten: India, China, Nepal, Pakistan, Bangladesh. India Review 8(3), 306–339.

Dhar, O. N. & Nandargi, S. S. (2003). Hydrometrological aspects of floods in India. Natural Hazards 28, 1–33.

Dhungel, K. R. (2008). Regional energy trade in South Asia: problems and prospects. South Asia Economic Journal 9, 173.

Dinar, S. (2008). International Water Treaties: Negotiation and Cooperation along Transboundary Rivers. Routledge, London.

Dixit, A. (2003). Floods and vulnerability: need to rethink flood management. Natural Hazards 28, 155–179.

Feiock, R. C., Steinacker, A. & Park, H. J. (2009). Institutional collective action and economic development joint ventures.

Public Administration Review 69(2), 256–270.

Gardner, R., Ostrom, E. & Walker, J. M. (1994). Social capital and cooperation: communication, bounded rationality, and behavioral heuristics. In: Social Dilemmas and Cooperation. Schulz, U., Albers, W. & Mueller, U. (eds). SpringerVerlag, New York.

Gauthier, D. (1985). Bargaining and justice. Social Philosophy and Policy 2, 29–47.

Grey, D. & Sadoff, C. W. (2007). Sink or swim? Water security for growth and development. Water Policy 9(6), 545–571.

Heath, J. (2006). The benefits of cooperation. Philosophy and Public Affairs 34(4), 313–351.

Howell, P. P. & Allan, J. A. (eds) (1994). The Nile: Sharing a Scare Resource. Cambridge University Press, Cambridge, UK. IFAD (2010). Rural Poverty Report 2011. IFAD, Rome. http://www.ifad.org/rpr2011/index.htm (accessed 11 March 2013).

Iyer, R. R. (2006). South Asian Water Concerns. South Asian Studies 6, 22–56.

Jacobs, J. (2002). The Mekong River Commission: transboundary water resources planning and regional security. Geographical Journal 168(4), 354–364.

Jansky, L., Pachovaa, N. I. & Murakami, M. (2004). The Danube: a case study of sharing international waters. Global Environmental Change 14, 39–49.

Karki, M. & Vaidya, R. (2009). Adaptation to climate change impacts and regional cooperation on water and hazards in the Himalayan Region. South Asia Water Journal 1(2), 117–131.

Klaphake, A. & Scheumann, W. (2006). Understanding transboundary water cooperation: evidence from Africa. Working Paper on Management in Environmental Planning 014/2006, Technical University of Berlin, Berlin.

Lama, M. P. (2000). Economic reforms and the energy sector in South Asia: scope for cross-border power trade. South Asian Survey March 2000 7(1), 3–22.

Libiszewski, S. (1995). Water Disputes in the Jordan Basin Region and their Role in the Resolution of the Arab-Israeli Conflict.

Occasional Paper No. 13, Center for Security Studies and Conflict Research, ETH Zurich/ Swiss Peace Foundation, Berne.

Malla, S. K. (2008). Towards a Regional Energy Market in South Asia. Paper No. 16, South Asia Centre for Policy Studies, Kathmandu.

Malla, S. K., Shrestha, S. K. & Sainju, M. M. (2001). Nepal’s water vision and the GBM basin framework. In: GangaBrahmaputra-Meghna region: A Framework for Sustainable Development. Ahmad, Q. K., Biswas, A. K., Rangachari, R. & Sainju, M. M. (eds). The University Press Ltd, Dhaka, pp. 143–197.

Mirza, M. M. Q., Dixit, A. & Nishat, A. (2003). Flood Problem and Management in South Asia, Preface. Kluwer Academic Publishers, London.

Mirza, M. M. Q., Warrick, R. A., Ericksen, N. J. & Kenny, G. J. (2001). Are floods getting worse in the Ganges, Brahmaputra and Meghna basins? Global Environmental Change, Part B, Environmental Hazards 3(2), 37–48.

Molden, D., Sakthivadivel, R., Samad, M. & Burton, M. (2005). Phases of river basin development: the need for adaptive institutions. In: Irrigation and River Basin Management: Options for Governance and Institutions. Svendsen, Mark (ed.).

CABI; IWMI, Wallingford, UK; Colombo, Sri Lanka, pp. 19–29.

Phillips, D., Daoudy, M., McCaffrey, S., Öjendal, J. & Turton, A. (2006). Transboundary water cooperation as a tool for conflict prevention and broader benefit sharing. Global Development Studies No. 4. Ministry of Foreign Affairs, Stockholm.

Priscoli, J. D. (1996). Conflict Resolution, Collaboration and Management in International Water Resource Issues. Alternative dispute resolution series, IWR Working Paper 96–ADR-WP-6, Institute for Water Resources, US Army Corps of Engineers, Alexandria, USA. Priscoli, J. D. (1998). Water and civilization: using history to reframe water policy debates and to build a new ecological realism. Water Policy 1, 623–636.

Priscoli, J. D. (2001). Participation, River Basin Organizations and Flood Management. ESCAP, Bangkok.

G. Rasul / Water Policy 16 (2014) 19–38

37

Priscoli, J. D. & Wolf, A. T. (2009). Managing and Transforming Water Conflicts. Cambridge University Press, Cambridge, New York.

Rahaman, M. M. (2009a). Integrated Water Resources Management: Constraints and Opportunities with a Focus on the Ganges and the Brahmaputra River Basins. Dissertation for the degree of Doctor of Science in Technology, Helsinki University of Technology, Helsinki.

Rahaman, M. M. (2009b). Integrated Ganges basin management: conflict and hope for regional development. Water Policy 11, 168–190.

Rahaman, M. M. & Varis, O. (2009). Integrated water management of the Brahmaputra basin: perspectives and hope for regional development. Natural Resources Forum 33, 60–75.

Rangachari, R. & Verghese, B. G. (2001). Making water work to translate poverty into prosperity: The Ganga-BrahmaputraBarak region. In: Ganga-Brahmaputra-Meghna Region: A Framework for Sustainable Development. Ahmad, Q. K., Biswas, A. K., Rangachari, R. & Sainju, M. M. (eds). The University Press Ltd, Dhaka, pp. 81–96.

Rao, R. V. V. (2006). Hydropower in the North East: Potential and Harnessing Strategy Framework. Background Paper No. 6, Central Electricity Authority, New Delhi, India.

Rasul, G. & Manandhar, P. (2009). Prospects and problems in promoting tourism in South Asia: a regional perspective. South Asia Economic Journal 10(1), 187–207.

Rawls, J. (1971). A Theory of Justice. Harvard University Press, Cambridge, MA. Sadoff, C. W. & Grey, D. (2002). Beyond the river: the benefits of cooperation on international rivers. Water Policy 4, 389–403.

Sadoff, C. W. & Grey, D. (2005). Cooperation on international rivers: a continuum for securing and sharing benefits. Water International 30(4), 1–8.

Sainju, M. M. & Shrestha, S. K. (2002). Cooperation in the Ganga-Brahmaputra-Meghna (GBM) basin: role of transport infrastructure. In: Contemporary Issues in Development. Biswas, A. K., Brichier-Colombi, J. S. A., Chowdhury, A. I. & Rasheed, S. (eds). Bangladesh Unnayan Parishad, Dhaka.

Shahjahan, M. (2008). Integrated River Basin Management for the Ganges: Lessons from the Murray-Darling and Mekong River Basins (A Bangladesh Perspective). PhD Thesis, The University of Adelaide, Adelaide, Australia.

Sharma, C. K. (1997). A Treatise on Water Resources of Nepal. MASS Printing Press, Kathmandu.

Sharma, P. C. (2004). Development of power. In: The Brahmaputra Basin Water Resources. Singh, V. P., Sharma, N. & Ojha, C. S. P. (eds). Kluwer Academic Publishers, The Netherlands, pp. 436–472.

Srivastava, L. & Misra, N. (2007). Promoting regional energy co-operation in South Asia. Energy Policy 35(6), 3360–3368.

Thakkar, H. (2006). What, Who, How and When of Experiencing Floods as a Disaster. South Asia Network on Dams, Rivers & People, New Delhi.

Thapa, A. B. (2011). New Spotlight Magazine 27 May 2004 (23).

Thapa, Y. B. (2013). Legal aspects of regional co-operation for basin optimization with reference to Koshi River. Himalayan Times, February 2013.

Timmerman, J. G. & Langaas, S. (2005). Water information: what is it good for? The use of information in transboundary water management. Regional Environmental Change 5(4), 177–187.

Uitto, J. I. & Duda, A. M. (2002). Management of transboundary water resources: lessons from international cooperation for conflict prevention. The Geographical Journal 168(4), 365–378.

Upreti, T. (2006). International Watercourses Law and Its Application in South Asia. Pairavi Prakashan, Kathmandu.

Vaidya, R. A., Eriksson, M., Bhandari, B. & Ouyang, H. (2010). Accessing Available Freshwater: Expanded Capture and Storage. Lead Paper for Thematic Area 4.3, presented at Water Crisis and Choices: ADB and Partners Conference 2010, October 11–15, Manila, Philippines.

Verghese, B. G. (1990). Waters of Hope: Himalaya-Ganga Development and Cooperation for a Billion People. Oxford & IBH Publishing Co.Pvt. Ltd, New Delhi.

Verghese, B. G. & Iyer, R. R. (1993). Harnessing the Eastern Himalayan Rivers: Regional Cooperation in South Asia. Konark Publishers Pvt. Ltd, New Delhi.

Verghese, B. G. & Ramaswamy, R. I. (1993). Harnessing the Eastern Himalayan Rivers: Regional Cooperation in South Asia.

Konark Publishers Pvt. Ltd, New Delhi.

Wolf, A. (1998). Conflict and cooperation along international waterways. Water Policy 1(2), 251–265.

38

G. Rasul / Water Policy 16 (2014) 19–38

Wolf, A. T. & Newton, J. S. (n.d.). Case Study Transboundary Dispute Resolution: the Lesotho Highlands Water Project. http:// www.transboundarywaters.orst.edu/research/case_studies/Documents/lesotho.pdf (accessed 13 April 2011).

World Bank (2010). World Development Report 2010. The World Bank, Washington, DC. Zawahri, N. A. (2008). International rivers and national security: The Euphrates, Ganges–Brahmaputra, Indus, Tigris, and Yarmouk rivers. Natural Resources Forum 32(4), 280–289.

Received 12 November 2011; accepted in revised form 24 June 2013. Available online 27 August 2013

Water Policy 16 (2014) 39–61

A participatory decision making process for community-level water supply V. H. Honkalaskara, M. Sohonia and U. V. Bhandarkarb a

CTARA, Electrical Annex Building, Indian Institute of Technology Bombay, Powai, Mumbai, 400076, India Corresponding author. Office MECH 303C, Department of Mechanical Engineering, Indian Institute of Technology Bombay, Powai, Mumbai, 400076, India. E-mail: bhandarkar@iitb.ac.in

b

Abstract This paper describes outcomes of a 3-year participatory action research project which involved community-level decision making to choose between various technologies to supply domestic water to a tribal village. Six technology alternatives were considered, which were ranked by adopting the analytical hierarchy process (AHP). At each stage, starting from project identification to project synthesis, people’s participation was sought in a true sense. This required design of novel strategies embedded in local culture, values, and language. The overall process yielded a participatory decision making method for a community, which would uphold people’s involvement, a sense of ownership, and control at each step, which is required for the successful implementation and sustainable operation of the project.

Keywords: AHP; Asia; Community-level decision making; Community-level water supply; India; People’s participation

1. Introduction The dominant development paradigm adopted by the mainstream development practitioners is often associated with reductionist analytical assumptions, top–down and centralized planning, and standardized universal methods to fix perceived universal problems (Chambers, 2007). Broadly, it strives to ascertain the development either through growth led by industrialization and globalization or through breakthroughs in science, technology, and policies to mitigate multiple causes and symptoms of underdevelopment (Leach & Scoones, 2006). Although this paradigm has been seen as an effective chariot to escalate a set of developed countries to achieve a high per-capita gross domestic product and human development index, it has been inadequate to address the local, complex, diverse, dynamic, uncontrollable, and unpredictable realities (Chambers, 1995) of most of the rural poor in developing and underdeveloped countries (Leach & Scoones, 2006). Notable examples include problems of poverty/ unemployment and the agrarian crisis in India (Dhas, 2009). doi: 10.2166/wp.2013.113 © IWA Publishing 2014

40

V. H. Honkalaskar et al. / Water Policy 16 (2014) 39–61

This led development practitioners and thinkers to evolve a paradigm that strives to address people’s interest and their needs in a more democratic and decentralized way without undermining the equity, sustainability, and environmental concerns. This paradigm assumes a central role of people’s participation (PP) in the process of community-oriented development projects that are typically promoted by governments, volunteer organizations, and civil society actors. A community-level development project usually involves six stages, namely, (1) problem identification, (2) analysis of available resources, (3) choice of technology alternatives, (4) designing and planning of the project, (5) implementation of the solution, and (6) successful operation of the project. Insistence on PP at each of these stages serves to strengthen the bottom-up approach (Oakley, 1998) of the new paradigm to problem solving.

The definitions of PP are broadly categorized into three different forms, depending upon the goal of the development projects, that emphasize (1) people’s involvement (Uphoff & Cohen, 1979; Chowdhury et al., 1996; Oakley, 1998), (2) people’s control (Pearse & Stiefel, 1980; Westergaard, 1986; Paul, 1987; Bank, 1994; Cornwall, 2000), and (3) people’s empowerment (Samad, 2002; Chambers, 2007) in the community-level development initiatives. These are not mutually exclusive definitions. One characteristic leverages the other two.

Broadly, the term PP involves people’s organized involvement in the community-level development initiatives that ensures their control over decision making and resources, which would eventually empower the people to achieve their own well-being. Participation is seen both as a means as well as an end for the development initiatives (Oakley, 1998). As a means, PP is utilized to improve the effectiveness and the efficiency of development projects. It is looked upon as a project input that is required to achieve the project’s objectives. In the perspective where PP is seen as an end, the emphasis is upon participation as a process in which confidence and solidarity in the community are built up. This process strengthens any development project and empowers the community to sustain the operation of the project. It improves the efficient use of the resources for the development projects (Oakley, 1991). It improves effectiveness of the project by involving the community in determining objectives, obtaining support for project administration, and making local knowledge, skills and resources available (Oakley, 1991). Improvement of the level of participation of the disadvantaged groups (rural poor, in the present context) leads to the reduction of social and economic inequality (Centre for the Study of Higher Education, 2008).

Thus, to ascertain the above-mentioned attributes, a development project must be extended by employing a systematic method for ‘participatory decision making’, which involves making choices among available alternatives. In this paper we consider a drinking-water supply project in the village of Gawand-wadi (population: 293, 120 km east of Mumbai (19°040 54.1000 N, 73°270 19.3400 E)) by interacting with the villagers over a period of 3 years. Technology adoption and sustainable management of water supply are determined by complex social forces and social relations that shape people’s choice of technology (Vincent, 2003). When choice of technology is made by an outside agency, community demands are often not met, even when such demands have been duly assessed (Singh, 2005). An approach that ensures sustainability of rural water supply schemes mobilizes the assets and insights of different social actors to influence decision making at all stages, including the design and choiceof-technology stages. The local people must be encouraged to negotiate, communicate, learn and arrive at joint decisions that reflect community choices and preferences (Gleitsmann et al., 2007; Ademiluyi & Odugbesan, 2008). Therefore, the project is developed in a participatory action research (PAR) mode in which, at each stage of decision making, the analytical hierarchy process (AHP) is adopted. Elements of PAR and AHP are described in Section 2. Section 3 introduces the assets and activities of the people in Gawand-wadi. The main part of this paper, which involves execution of a

V. H. Honkalaskar et al. / Water Policy 16 (2014) 39–61

41

participatory analytical hierarchy process (PAHP) exercise in the village, is described in Section 4. It includes participatory selection of development projects, study of the water-fetching activity and the available resources, selection of the six candidate technology alternatives, and the execution of the PAHP exercise with the people. This participatory execution required several novel context-specific interventions. Section 5 describes a set of observations made by the facilitator regarding the PAHP exercise. A sensitivity analysis that was carried out to see how the rankings of the six alternatives change with the removal of different attributes one at a time is explained in Section 6. Finally, the main findings and conclusions are listed in Section 7.

2. AHP and PAR 2.1. AHP AHP is a multi-criteria decision making approach that can be used to make a choice of technology from the available alternatives (Saaty & Vargas, 1982; Saaty, 2008, 1990). Its usefulness has been considered to be better than other rating methods such as multi-attribute value theory, the simple weighted average method, the elimination and choice translation algorithm (ELECTRE), and the preference organization method for enrichment evaluation (PROMETHEE) (Raju et al., 1995; Scholl et al., 2005; Pohekar & Ramachandran, 2004). The AHP methodology has been accepted by the international scientific community as a robust and flexible multi-criteria decision making tool for dealing with complex decision problems (Elkarmi & Mustafa, 1993; Srdjevic, 2005).

AHP is carried out in the following four steps: (1) define the problem and find out the knowledge requirement for further analysis, (2) structure the complex decision problem as a hierarchy of goal, attributes, and alternatives (see Figure 1), (3) pair-wise comparison of elements at each level of the hierarchy with respect to each criterion of the preceding level, and (4) vertically synthesize the judgements over the different levels of the hierarchy (Saaty & Vargas, 1982; Tiwari & Banerjee, 2001).

The synthesis part of the AHP exercise is usually carried out in the following four steps.

1. Let there be m attributes for comparison. Determine their respective weights wi by using a standard scale, where i ¼ 1, … , m.

2. For each attribute i, compare the n alternatives and determine by pair-wise comparisons attribute weights Wij, where ( j ¼ 1, … , n), with respect to attribute i.

Fig. 1. Structure of the complex decision problem in AHP exercise.

42

V. H. Honkalaskar et al. / Water Policy 16 (2014) 39–61

3. Determine the final rankings by finding Rj (see Table 1; here [w]¼ [w1, w2, …, wm], and [Wj] ¼ [W1j, W2j, …, Wmj]) by using the following relation: Rj ¼ W1jw1 þ W2jw2 þ . . . þ Wmjwm

(1)

4. The alternatives are then ordered by the values of Rj, with the most preferred alternative having the largest Rj. AHP has been extensively utilized to make decisions in corporate as well as in government sectors (Saaty & Vargas, 1982). Most often, the task is carried out by experts of AHP, and the stakeholders participate in the process of data sharing only. On the other hand, for community-level decision making, there are relatively few examples (Kangas, 1994; Ananda, 2007; Schmoldt et al., 2007). In the references, participation was sought during the second step of pair-wise comparisons only. For example, Soma (2003) considered the case of teaching AHP to the stakeholders to increase the level of participation during the second stage of the AHP for a fisheries project that has been termed as mere ‘technical action research’ (Grundy, 1982) that maintains the dominant role of the facilitator in the whole process. In contrast, PAR offers a platform to enable the stakeholders to have a dominant role in the whole process of decision making (Grundy, 1982).

2.2. PAR with AHP Hall (1981) described PAR as an integrated activity that combines social investigation, educational work, and action. Many researchers thereafter, namely, Maguire (1987), Reason & Rowan (1981), and Greenwood et al. (1993), have coined different definitions for PAR. Despite differences in definitions and applications, there are a few commonalities, which include: 1. PP requirement in all phases of the PAR process; 2. participants’ control in the process (in the context of this study, ‘participants’ implies people from the community); 3. mutual respect for different provinces of knowledge that the team members have; 4. bidirectional education of the researchers and participants (White et al., 2004).

Table 1. Attribute weight and alternatives’ weight matrix.

Technology alternative (comparison values) Attributes

Attribute weights

T1

T2

A1 A2

Ai

Am

W1 W2

Wi

Wm ∑w ¼ 1

W11 W21

W12 W22

R1

R2

Tj

Tn

Rj ¼ [w][Wj ]T

Wmn Rn

V. H. Honkalaskar et al. / Water Policy 16 (2014) 39–61

43

The common element of PAR projects is the fact that PAR is regarded as an orientation rather than a method. In the present work, ideas gathered from the PAR are used to improve the level of participation during every step of the AHP exercise. The PAR approach is utilized to formulate a participatory AHP to help in deciding the best alternative for water supply in Gawand-wadi. Contextual aspects such as traditions, beliefs, skills, capabilities, and experiences are utilized to teach the AHP methodology to ensure people’s participation in the whole process, which increases efficiency, effectiveness, and sustainability of the development initiative.

3. The project area Gawand-wadi is a tribal village, located in the Karjat tribal block of Raigad district in the state of Maharashtra in western India. The major sources of livelihood for the people are rain-fed agriculture (mainly paddy, finger millet, and vegetables); collection and selling/consumption of forest produce such as gums, fruits, vegetables, roots, and oil seeds; and agricultural labour in neighbouring irrigated areas. Occasionally, the villagers also find paid labour work in road construction and infrastructure projects in neighbouring areas. The nearest market place is located at a distance of 30 km.

The average family size is 5.63 with nearly 63% of households having a size between four and eight people. The level of education in the youth (18–25) is approximately seventh standard of the stateboard syllabus, while for adult men (25–50) it is fifth standard. Total village land is 143.55 ha. Most of the land is sloping and forest land. Total cattle head is 212 with 4.07 cattle head per family, which is higher than the national average of 1.73 (National Sample Survey Organization, 2010). The surrounding forest serves as a resource to collect fuel-wood, wood, and other forest produce. It comprises both villageowned and government-owned forest land. The main water source is a small earthen dam near (at a distance of 200 m) the village residential area. The reservoir capacity is 67,000 m3 and its water lasts for the whole year. There is a small river at a distance of 1.5 km from the village residential area. The river carries water until the end of March. Thereafter, water remains in a few small ponds in the river basin. Apart from this, the river and reservoir also serve as breeding and raising grounds for fish and crabs.

The villagers mainly depend upon local ecological resources for their livelihoods and therefore their daily activities vary seasonally. Most of what they produce is for self-consumption. Their livelihood activities can be grouped into four types as follows: rain-fed agricultural activities (paddy, finger millet, and vegetable cultivation, carried out from late summer up to early winter); livestock raising (mainly done in the monsoon season); domestic activities (firewood fetching, water fetching, cooking, fish/crab catching, plinth preparation, space heating1, clothes washing etc., mostly carried out through the year); and employment and trade (forest collection, wage labour, forest cutting, carpentry, liquor making, sand collection, bamboo work, moha seeds collection, brick making, etc.).

4. PAHP in the village PAHP methodology seeks to involve the people at all four stages of the AHP exercise (see Section 2).

The methodology was evolved over a number of collaborative meetings, surveys, and studies in the village. The following subsections detail all four stages of a PAHP exercise carried out in the village.

1

Space heating is required during the winter season to warm the house during the night and early morning.

44

V. H. Honkalaskar et al. / Water Policy 16 (2014) 39–61

4.1. Participatory goal setting and study of the problem A shared development agenda was chosen by employing a participatory study of the context through a set of participatory rural appraisal (PRA) (Chambers, 1994) exercises, sample surveys, personal interviews, and group discussions, which ensures sharing, reflection, and analysis of the people2.

Problems faced by the villagers (see Table 2) were listed through a set of focused group discussions among women and men. A further survey was conducted among 37 men and women (11 young men, 8 young women, 10 adult men, and 8 adult women) in 20 households. Average hierarchies for different groups were found by assigning a weight to each problem, equal to the reciprocal of its rank assigned by an individual followed by a weighted average (see Table 2). This study reveals people’s perception that the drudgery in firewood fetching and water fetching, and lack of adequate and secured earning sources are the major problems faced by the village. Presently, both the firewood fetching and water fetching activities are carried out by women, carrying loads (around 20–32 kg) on their head and walking on sloping/undulating terrain. These activities lead to considerable health-related impacts. A survey of 40 women across various age groups (age distribution – below 30: 22; 30–50: 12; above 50: 6) was carried out to identify the health hazards associated with these two laborious activities. The survey revealed that most women face problems of backache, neck ache, calf-muscle ache, and fatigue.

All the livelihood activities (mentioned in Section 3) carried out in the village over the year involve exchange of time, material and energy resources, and some of them involve exchange of money. Human hours and human energy values associated with these activities were found by carrying out a survey in six households, selected by considering their even distribution across the variation of family size with land holding.

Contribution of men, women, and children for all activities is 37.9, 58.9, and 3.2% respectively.

The percentage distribution of human work hours for different activities is as follows: agriculture (30.4%), animal raising (9.6%), domestic (44.6%), employment and trade (15.3%). Domestic activities, which demand most of the human work hours, are mainly carried out by women. Firewood fetching and water fetching activities demand 13.4 and 7.3% of the total human work hours Table 2. Overall problem ranking.

Problems faced

Young men

Adult men

Adult women

Young women

Drudgery in firewood fetching Lack of secured and adequate income source Drudgery in water fetching Lack of health care in village Low level of education Liquor addiction Lack of toilets Lack of transportation facilities Sporadic electricity supply

2 1 3 8 4 6 9 5 7

3 2 5 1 6 4 7 8 9

1 3 2 4 8 7 5 6 9

1 7 2 3 9 6 3 5 8

2

A detailed document of goal setting is being written separately. It includes an elaborative account of the PRA exercise, time and energy utilization survey, problem ranking exercise, and analysis and reflection with people. This paper simply lists the outcomes of this exercise.

V. H. Honkalaskar et al. / Water Policy 16 (2014) 39–61

45

spent over all the livelihood activities carried out in the village for 1 year. Using the average human energy demands for different livelihood activities (Date, 1989), average per-capita energy expenditure in each of these activities was calculated and is listed in Table 3. It can be seen that water fetching and firewood fetching activities demand significant amounts of human energy throughout the year.

The sharing and discussion of the analysis based on the information generated through the PRA exercise and other studies helped to refine the perceptions of the local context for the facilitator as well as for the local people. Finally, the following three areas were selected, which were found to be of prime importance to the people: 1. Addressing the problem of drudgery associated with the firewood fetching activity.

2. Addressing the problem associated with the water fetching activity.

3. Livelihood generation.

The problems of addressing the drudgery associated with the firewood fetching activity and livelihood generation are being tackled differently. The PAHP exercise focused on selecting the best technology to address the problem of drudgery associated with water fetching in the village.

4.2. Domestic water fetching and its sources The water fetching activity was further studied by carrying out surveys in the six households. The village has three wells, one on the downstream side of the dam and two on the upstream side. Water lasts in the two upstream wells until the end of February. The downstream well is recharged from the dam and thus retains water for the whole year. An average overhead load of domestic water of 32 kg per trip per person is fetched during winter and summer from the downstream well (see Figure 2), which is located at a distance of 300 m from the central community hall in the village and at a vertical depth of 25 m. In the monsoon season, however, water is fetched from a nearby stream that flows during this period. Average water fetched per household of six people varies from 140 litres per day during the monsoon season to 240 litres per day during summer. The water resources sufficient to fulfil the domestic water need of the village throughout the year were identified with the help of local people. It was

Table 3. Average human energy expenditure for different livelihood activities per year.

Serial No. 1 2 3 4 5 6 7 8

Activities Animal rearing Agriculture Water fetching Firewood fetching Cooking Clothes washing and plinth preparation Fishing Other activities

Annual per-capita energy expenditure (Gcal/year) 15.9 68.0 45.8 24.8 20.4 9.9 3,091 64,471

Percentage of energy expenditure (%) 6.37 26.26 18.30 9.91 8.15 3.98 1.23 25.77

46

V. H. Honkalaskar et al. / Water Policy 16 (2014) 39–61

Fig. 2. Water resources of Gawand-wadi – earthen dam and percolation dug well (Google Maps, March 2011).

found that a perennial well (diameter 6 m, depth 6.4 m) located on the downstream side of the check dam (Figure 2) and heavy rainfall during the monsoon season (3,000 mm (Date, 1989)) are the only reliable water resources. To find out the ability of the perennial well to supply domestic water, a yield test (Bureau of Environmental Protection, 2005) was carried out during the second week of May 2010. Variation of the height of the water column with time was noted as shown in Table 4.

The yield test gives the variation of recharge rate of the well with water column height (see Figure 3).

This relation is used to calculate maximum possible recharge of water during a day. The volume of water that can be recharged from the minimum water column height, say 0.5 m, (to ensure safe working of the pump) during the whole day is 75,400 litres during the month of May. This is much higher than the maximum requirement (12,500 litres) for daily domestic water during summer season for the village.

The following subsection elaborates the second step of the PAHP exercise, which lent a proper structure to the problem in the form of goal, alternatives, and attributes.

Table 4. Yield test data for a well at Gawand-wadi (May 2010).

Actual time

Time duration (minutes)

Height of water column (m)

10.00 a.m.

12.30 p.m.

6.30 p.m.

9.00 a.m.

0 150 510 1,380

2.50 2.76 3.14 3.43

V. H. Honkalaskar et al. / Water Policy 16 (2014) 39–61

47

Fig. 3. Variation of recharge rate with water column height.

4.3. Structuring the complex problem The basic concepts of AHP were taught to the people with the help of a multi-criteria decision making process that is carried out regularly in the village, namely, the activity of fixing a marriage. If parents want to select a groom for their daughter, certain attributes of the potential candidate are considered, namely, personality, character, land ownership, and the distance from their village. They usually look for three or four such youngsters (alternatives) with some of these attributes and choose one based on some comparison of these attributes. The idea of considering the different criteria or attributes for decision making and to select one among many alternatives was explained with the help of this familiar example. For the meeting venue, a blackboard and sufficient space to accommodate more than 40 people were needed. The village school was chosen as the venue. As far as the invitations were concerned, the traditional method of inviting people to a village-level meeting was adopted. In this method a person, called Kotval, is assigned the task to go out to each part of the village and announce the time and venue of the meeting and its purpose.

Four technology alternatives were identified by the people, and two additional alternatives were identified by the facilitator. The unknown alternative of a biomass-operated steam engine was explained to the people via a video clip and pictures. The other five known technology alternatives were discussed with the people. The facts and figures of these six different alternatives were explained on the blackboard.

4.3.1. Technology alternatives. Based on the water resources and available technologies, six different technology alternatives (T1–T6) were chosen, as depicted in Figure 4, to accomplish the goal. For the pump and pipe alternatives (T1, T2, and T3), a system curve (Figure 5) showing variation of total head with water flow rate is drawn by considering different available pipe diameters. For the pedal pump alternative (T4), gravitational head increases by 1.5 m because of the increase in the average elevation of the delivery side, so the system curve shifts upward by 1.5 m. A suitable pump is selected by matching the pump performance curves provided by the local dealers to the system’s requirement at its higher efficiency range. The six alternatives are listed in the following subsections.

48

V. H. Honkalaskar et al. / Water Policy 16 (2014) 39–61

Fig. 4. Technology alternatives.

Fig. 5. System curve for pumping water to the village (DÂź pipe diameter).

V. H. Honkalaskar et al. / Water Policy 16 (2014) 39–61

49

4.3.1.1. T1 Electric pump to lift water from the well to a water tank in the village. A single-phase submersible pump (power output: 1 hp; flow rate: 1.2 litres/s at 34 m of total head; pipe diameter: 1.25 inch (1 inch ¼ 25.4 mm)) can be utilized to pump water (2.16 hours of daily pumping to cater for average daily domestic water need, i.e. 9,360 litres, for the village) from the well to a constructed tank (10,000 litres) in the village. A ferro-cement water tank can be built by utilizing locally available sand and stones, but cement and steel must be purchased from the local market. An electricity connection must be provided near the well. This alternative involves operation and maintenance of an electric pump (Rs. 9,192 per year for electricity bill and labour charges). A village-level water committee has to take responsibility for the monthly collection of charges from each household for the daily operation and maintenance of this project. Total estimated cost for this alternative is around Rs. 91,290.

4.3.1.2. T2 Gasoline engine pump to lift water from the well to a water tank in the village. This alternative is similar to the first alternative, where, instead of an electric motor, a gasoline engine is used to run the pump. The power output of the engine is 1.5 HP and it is normally operated with one-sixth gasoline and five-sixths kerosene. The flow rate is 1.7 litres/s at 33 m of total head, and the pipe diameter is 1.5 inch (3.8 cm). This would require an additional small room near the well to house the gasoline engine.

Daily pumping for 1.5 hours is needed to cater for the average daily domestic water need for the village.

Fuel will have to be procured from the market, situated 12 km away, by a village-level water committee.

The estimated cost of the project erection is Rs. 96,690. An estimated operation and maintenance cost is Rs. 33,360 per year.

4.3.1.3. T3 Steam engine to lift water from the well to a water tank in the village. In this alternative, water would be pumped by a reciprocating steam engine power plant (1.5 hp – the only available 1–2 hp engine in the market places close by; flow rate: 1.7 litres/s at 33 m of total head; pipe diameter: 1.5 inch), operated on biomass fuel. It requires a shed in which to house and operate the steam engine and to store biomass. Biomass (approximately 12 kg/day) would be contributed by each household.

A skilled operator is required to run the steam engine plant. Estimated cost for the erection of the project is Rs. 194,190. An estimated operation and maintenance cost is Rs. 39,600 per year.

4.3.1.4. T4 Pedal-operated water-lifting device. A pedal-operated centrifugal pump (coupled with a pedal-operating device with belt and pulley) can lift water from the well to the village residential area. A healthy man can deliver an average of around 250 W of human power while pedalling for 20 minutes (Ikonen, 2011). Thus, the required power to lift water by the pedal pump should be below 250 W for 20 minutes of continuous operation. A total head (H ) of 32.8 m is obtained for a flow rate (Q) of 0.18 litres/s through a 1 inch (2.5 cm) diameter pipe. Owing to the unavailability of such a pump in the market (i.e. one providing the required head and flow rate for a limited power input of 250 W), two pumps delivering water at half the total head can be coupled. This combination of total head (this is double the head of a single pump) and flow rate matches both system and pump performance. Efficiency (η) of the pump and pulley-belt pedal drive mechanism can be assumed as 0.26 (efficiency of selected pump for the flow rate of 0.18 litres/s is around 0.27;

50

V. H. Honkalaskar et al. / Water Policy 16 (2014) 39–61

efficiency of the mechanical drive can be assumed as 0.95). The required human energy input is

rgHQ 100ðkg=m2 Þ9:8ðm=s2 Þ32:8ðmÞ0:00018ðm3 =sÞ ¼ ¼ 224 W h 0:26

(2)

where ρ is the density of water and g is gravitational acceleration. This is below the average power that can be delivered (250 W) by a healthy person for 20 minutes. Thus, at 0.18 litres/s, a person can pump around 200 litres of water in 20 minutes. It would require 14.5 hours to pump the average domestic water demand of the village. People prefer to pump water early in the morning as they want to carry out other livelihood activities during the rest of the day. Thus, several pumps are required. Ten families can have a common pedal-operated pump so that one group of 10 families can pump the total daily requirement of water within 2.8 hours. It is a simple device with very low maintenance. The entire village would require five such pumps. A small room is required to house these pumps near the well. Total estimated cost is Rs. 0.12 million. An estimated maintenance cost is Rs. 600 per year.

4.3.1.5. T5 Rainwater collection and storage in ferro-cement tanks near houses. Average rainfall in Gawand-wadi is 3,000 mm/year. This alternative involves building ferro-cement water tanks (10,000 litres) to collect rainwater falling on the roof of the tank during the monsoon season. The average amount of rainwater received per tank (diameter: 2.7 m; height: 1.75 m; wall thickness: 5 mm; volume: 1 m3) is 3 (avg. rainfall, m) 3.14 (value of π) 1.352 (tank radius, m) ¼17.1 m3.

Thus, the tank of above-mentioned dimensions receives an ample amount of rainwater to fill it completely during the monsoon season. Harvested rainwater in the tanks during the monsoon season can be directly used for domestic water supply. Harvested water in the tanks at the end of the monsoon season can be utilized for the rest of the year to meet the daily water requirement. An average water requirement per household of six persons is 50,000 litres for winter and summer (8 months). Therefore, there is a need to build five such tanks that would provide up to the capacity of 50,000 litres, to meet the daily household need for every household during winter and summer. The cost of the water tank (10,000 litres) is Rs. 30,290. Thus, the total cost for building the tanks to supply every household is Rs. 7.5 million.

4.3.1.6. T6 Water pond in bedrock. A water storage pond, dug in hard rock, forms a water reservoir. If there is a perennial underground stream or sufficient depth of pond, it can hold water until the end of summer. This alternative has been implemented by a non-governmental organization called Shashwat on the eastern side of the Western Ghats. It was carried out by ensuring people’s participation and led to a successful water storage alternative for the entire year (Kapur, 1995). In the present case, there is a rocky terrain at the same level as the village, located at a distance of around 150 m from the village centre.

There is a seasonal stream flowing through this terrain. A pond of size of 30 30 6 m can store 3,840,000 litres of water, taking into account evaporation losses. This is sufficient for 240 days of the non-monsoon period. Domestic water can be fetched manually from the pond built in that space.

Estimated cost for the project is Rs. 2.3 million.

It is to be noted that the first four alternatives (T1, T2, T3, and T4) require external energies. T1 and T2 require an electric motor and a gasoline pump for which energy must be purchased from outside the village. T3 requires locally available biomass fuel. T4 involves manual pumping. T5 and T6 require

V. H. Honkalaskar et al. / Water Policy 16 (2014) 39–61

51

no external energy, being implemented near homes and the village residential area, respectively, with negligible elevation difference. Hardware redundancy was not considered for any of the options, to reduce the total cost of erection of the project.

4.3.2. Identification of attributes. A definition of an attribute was framed in the local language with the help of the villagers, translated as: ‘Attributes are different points that should be taken into consideration while deciding the best choice among the given alternatives to seek our common goal.’

Based on this, the villagers were asked to identify the possible attributes to select an alternative for the water supply project. Whenever any ambiguity occurred, the marriage example was utilized to make it clear. People identified a total of ten attributes for the water supply project. Six more attributes were proposed by the facilitator, four of which were accepted by the villagers. These are listed in Table 5.

The rejected attributes were ‘requirement of social capital’ and ‘social capital generation’. People argued that these two attributes are covered by the remaining fourteen. Each attribute was expressed in the local language to reduce the confusion that develops during comparison of attributes. These were further classified into quantitative (A1–A6) and qualitative (A7–A14) attributes as shown in Tables 6 and 7 respectively.

The quantitative attribute values shown in Table 6 are calculated as follows.

• A1 Space: This is the total space required for setting up the infrastructure of an alternative.

• A2 Erection time: This comprises total erection time in number of days required for the building of total infrastructure for an alternative in the present context.

Table 5. Attributes for water supply project alternatives.

Serial No.

Attribute

Particulars

1

Space

2 3 4 5 6 7 8

Erection time Energy required to operate project Operation time Erection expenses Operation expenses Operation ease Effects on other livelihoods

9 10 11

Self-reliance to operate Purity of water Water storage

12 13 14

Durability Reliability Ease of erection

Infrastructure for an alternative requires certain space for its erection Time required to complete the project Fuel energy resources ( joules/litre of water) required to get water Time required to collect the water for whole village Cost of erection of the project infrastructure Operation cost and maintenance cost for daily water supply Comfort as per skills and work hours to collect water Negative impact of this project on other livelihood activities in the village Self-dependency of energy resources to get water Cleanliness of water Sufficiency of the water resource for drinking-water supply throughout a year This indicates life of the infrastructure Ability of project to supply water without failure Ease of the project implementation

Nature of variable Quantitative Quantitative Quantitative Quantitative Quantitative Quantitative Qualitative Qualitative Qualitative Qualitative Qualitative Qualitative Qualitative Qualitative

52

V. H. Honkalaskar et al. / Water Policy 16 (2014) 39–61

Table 6. Quantitative attributes.

Attribute A1 A2 A3 A4 A5 A6

2

Space (m ) Erection time (days) Energy required to operate (kWh/day) Operation time (minutes/day, total human minutes for the villager per day) Erection expenses (Rs.) Operation expenses (Rs./year)

T1

T2

T3

T4

T5

T6

11.5 13 3.34 2,286

20 14 10.41 2,293

30 16 52.41 2,395

10 10 3.88 1,040

2,210 60 0.30 780

961 100 2.72 3,120

91,290 96,690 194,190 111,750 9,192 33,360 39,600 600

7,500,000 2,288,647 0 0

• A3 Energy required to operate: The total average water required for the village is L litres/day. The total pumping head is taken as H m (as indicated in Figure 5). Energy supply rate (E) is evaluated as

E¼

L(m3=s)r(kg=m3 )g(m=s2 )H(m) h

(3)

where ρ (kg/m3) is density of water, and η (device efficiency) is found by multiplication of pump efficiency and machine (motor or engine) efficiency values. Values for pumping efficiency of the first four alternatives are estimated from the efficiency versus flow rate curves provided by manufacturers.

Values for machine efficiency are assumed (electric motor: 0.9; gasoline engine: 0.25; steam engine: 0.05). Human energy costs to fetch domestic water from the source are added to this energy value. To calculate human energy costs, the average time required to carry water from the source to homes for each trip is assumed to be 2 minutes (tank would be located 160 m from the village centre: thus, it takes around 2 minutes to walk from the source to home at 5 km/h) for the first three alternatives, 1 minute for the rainwater harvesting alternative, and 3 minutes for the water pond alternative (proposed location is 250 m from the village centre). It is assumed that the pipe connection from the pedal-operated pump can be directly taken to the home, to avoid the energy requirement for manual collection of water from the tank in the village. Human energy cost to fetch water is assumed as 5 kcal/min (Date, 1989).

• A4 Operation time: This is the total time required to fetch the average domestic water requirement (by using necessary forms of energy) from the source to home for the whole village per day. This includes total human minutes required in reaching the source, collecting water, and carrying it back to the home in the case of all the alternatives other than the pedal-power pump (T4). For pedal-power pumping, it includes the time required to pump the water to homes. It also includes the time required to fill the tanks (for the first three alternatives).

• A5 Erection expenses: These involve total estimated cost for the construction of the total infrastructure for an alternative.

• A6 Operation expenses: These involve daily expenditure on operation, and average maintenance cost for an alternative per month. In a village-level meeting, the labour charge for the first alternative was decided as Rs. 500/month, and for the second and third alternatives as Rs. 1,500/month. The steamengine power plant operation requires two persons and higher maintenance cost.

Table 7. Qualitative attributes.

T2

T3

T4

T5

T6

A7

Operation ease

Just switching on and off the electrical machine

Gasoline fetching and running gasoline engine

Manual operation for 15–20 minutes per household

Collecting water from tank near home

Fetching water overhead from water tank in rock near the village

A8

Effects on other livelihoods

People cannot use well water for irrigation

People cannot use well water for irrigation

Running steam engine plant, feeding wood fuel to the boiler and monitoring the system People cannot use well water for irrigation

People cannot use well water for irrigation

Farming land of one or two farmers is required to build water tank in rock

A9

Self-reliance to operate Purity of water

Dependence on gasoline Good

Dependence on firewood Good

Manual operation

A10

Dependence on electricity Good

Space required for the construction of tanks for each family may demand a few people’s farming land around the village residential area Manual operation

A11

Water storage

Adequate

Adequate

Adequate

Adequate

Have to boil water before drinking Less, depending on tank capacity

A12

Durability

Gasoline engine

Steam engine

Simple mechanism

A13

Reliability

Electrical machinery Dependence on grid electricity, repair is not local Laying of cables and pipes, building of water tank

Dependence on gasoline, repair can be local Laying of pipes, building of water tank, constructing a small room to house gasoline engine

Dependence on wood fuel, repair is not local Laying of pipes, building of water tank, constructing a shed to erect the set-up of steam engine

Repair can be local, human-powered machine

Have to boil water before drinking Less, depending on water structure capacity and construction A tank in rock, the most durable Least maintenance, does not depend on any energy resource Digging, blasting, and construction work, lots of human work hours and collective work are required

A14 Ease of erection

Good

Laying of pipes and constructing small room to house pedaldriven machine

Cement concrete construction Least maintenance, does not depend on any energy resource Building of around six water tanks for each family

Manual operation

53

T1

V. H. Honkalaskar et al. / Water Policy 16 (2014) 39–61

Attribute

54

V. H. Honkalaskar et al. / Water Policy 16 (2014) 39–61

4.3.3. Sub-division of attributes. It is not possible to compare all the attributes simultaneously as people can deal with information involving only a few facts – seven, plus or minus two (Saaty, 1990). Similarly, comparisons of attributes in pairs require that they must be homogeneous, otherwise significant errors may be encountered (Saaty, 1990). For these reasons, the attributes were divided into three major groups: resources (M1), infrastructure (M2), and expenditure (M3). These three groups serve as a second level of attributes in the hierarchy structure of the second stage of AHP (see Figure 6).

Resources (M1) involve all kinds of natural and other resources required for building and operation of the project. Hence, it is divided into a third level of attributes: energy (M4), water (M5), space (A1), and effect of the utilization of the resources on other livelihoods (A8). Energy (M4) is further divided into two attributes: energy required to operate (A2), and self-reliance of the alternative in terms of energy resource (A9). Water (M5) is further categorized into its quality (A10), and availability (A11). Infrastructure (M2) involves all kinds of physical structure required for the building and operation of the project. It is further classified into four categories of durability (A12), reliability (A13), erection (M5), and operation (M6) of the infrastructure.

Erection and operation are further classified into the following: ease and time required for the erection (A14, A3) and operation (A7, A4) respectively. Expenditure (M3) is the total expense of the building and operation of an alternative. It is categorized into erection expenses (A5), and operation expenses (A6).

These attributes are listed in Table 8. Figure 6 depicts the hierarchy of attributes. Here, the attributes, denoted as Ai, are divided into homogeneous groups by introducing a few new attributes, denoted as Mi. 4.4. Pair-wise comparisons Different technology alternatives or attributes for the water supply project were allotted comparison weights by plotting a ladder structure on the blackboard as shown in Figures 7 and 8. It has nine rungs,

Fig. 6. Hierarchical sub-division of attributes.

V. H. Honkalaskar et al. / Water Policy 16 (2014) 39–61

55

Table 8. Characteristics of technology alternatives for water supply project (qualitative attributes).

M1 M2 M3 M4 M5 M6 M7

Attribute

Particulars

Natural resources Infrastructure Expenditure Energy resources Water resources Erection Operation

Natural resources utilized in the project erection, operation, and maintenance Infrastructure to be built for the particular alternative Expenditure for the erection, operation, and maintenance of the project Energy required to run the project Water resource used to supply the drinking water Erection of the project structure Operation and maintenance of the project

Fig. 7. Ladder structure to compare attributes and technology alternatives.

Fig. 8. Deduction of the qualitative comparisons with the people.

56

V. H. Honkalaskar et al. / Water Policy 16 (2014) 39–61

adopted from Saaty’s fundamental scale for pair-wise attribute comparison (Saaty, 1990). A comparison matrix was filled by comparing each possible pair of different alternatives or attributes with the help of this ladder. A technology alternative was written in front of the middle rung. Each other alternative (which was not compared with the middle one in any earlier exercise) was compared and written in front of a rung according to its importance. Attribute-to-attribute comparisons, at each level, were carried out in a similar way.

4.5. Synthesis The process of calculating the final priorities was elaborated on the blackboard. The villagers were asked if all these calculations were possible for a human mind. It was concluded that these calculations were best done with the help of a computer. Priority vectors (Wij and Wi) were worked out and their normalized values are given in Table 9. Finally, rankings Rj ¼ ∑WiWij were evaluated and are reported in the last row of Table 9. Technology T1 (electric pump) emerged as the most preferred alternative.

Consistency ratios (RI; Saaty (1990)) for all the comparison matrices were below 0.1.

5. Observations 1. It was noticed that, when the village-level meeting was conducted in the traditional way, the villagers showed more participation in terms of quantity and quality. A contextual decision making example of a marriage helped the villagers to easily understand the basics behind the AHP methodology. These initiatives improved the feeling of ownership of the project among the villagers.

Table 9. Final attribute weight and normalized attribute matrix.

Attributes A1 A2 A3 A4 A5 A6 A7 A8 A9 A10 A11 A12 A13 A14

Space Erection time Energy required to operate project Operation time Erection expenses Operation expenses Operation ease Effects on other livelihoods Self-dependence to operate project Purity of water Water storage Durability Reliability Ease of erection Ranking – Rj ¼ ∑wiWjj (Rank)

Attribute weights

Technology alternative (comparison values) T1

T2

T3

T4

T5

T6

0.0126 0.0210 0.0221

0.2621 0.2171 0.1432

0.1747 0.1977 0.0477

0.1311 0.1445 0.0320

0.3609 0.3568 0.0955

0.0293 0.0545 0.4055

0.0418 0.0295 0.2811

0.0429 0.0262 0.0785 0.3004 0.0401 0.0193

0.1229 0.3658 0.0724 0.3682 0.2477 0.0311

0.0891 0.2923 0.0256 0.2630 0.2650 0.0513

0.0592 0.1107 0.0239 0.1117 0.1599 0.0892

0.2217 0.1799 0.1904 0.0569 0.2650 0.1672

0.4740 0.0214 0.3438 0.1625 0.0377 0.3786

0.0331 0.0299 0.3438 0.0378 0.0248 0.2827

0.1606 0.0229 0.2093 0.0563 0.0070

0.2194 0.2293 0.0959 0.0568 0.2035 0.2179 (1)

0.2194 0.2293 0.0461 0.0479 0.1257 0.1652 (5)

0.2194 0.2293 0.0278 0.0266 0.0887 0.1028 (6)

0.2194 0.2293 0.1940 0.1415 0.5155 0.1669 (4)

0.0403 0.0294 0.2030 0.2851 0.0442 0.1819 (3)

0.0820 0.0536 0.4332 0.4420 0.0224 0.1845 (2)

V. H. Honkalaskar et al. / Water Policy 16 (2014) 39–61

57

2. A healthy discussion among the villagers was facilitated before freezing the comparison ratios. A few choices of weights did not match the facilitator’s perception. For example, people allotted more importance to the attribute ‘Operation ease’ than to other attributes, because their main interest in this project is to have an easy way to fetch the domestic water. ‘Ease’ is equated with ‘elimination of drudgery’ in water fetching.

3. Relating the project to their own traditional activities increased the sense of belonging and thus the level of people’s participation.

6. Sensitivity analysis In order to check if any one attribute dominated the rankings, a sensitivity analysis was carried out to see how the rankings change with the removal of different attributes one at a time. When an attribute is removed, a new priority vector for the other attributes is obtained by deleting that attribute from the original pair-wise comparison matrix and calculating the new priority vector for the remaining attributes (Raju et al., 1995). The ranking values for the fourteen cases are summarized graphically in Figure 9.

Case 0 corresponds to the original results, i.e. considering all the attributes. The results of Case 1, Case 2 … Case 14 are obtained by deleting the major attributes A1, A2 … A14 respectively. Sensitivity analysis results (Figure 9) show that the attribute A7 ‘operation ease’ has played a major role in determining the final rankings. If it is removed, the rank of the electric pumping alternative would go down to fourth position. This confirms the significance of the attribute ‘operation ease’. The electric pumping alternative is not that sensitive to the other attributes, and it retains its first position for the other cases. Similarly, for the rock pond alternative (T6), durability is the key attribute in determining its position. In the absence of this attribute, the rock pond goes down to the lowest position. Attributes A4, A5, A7, A9, A10, and A12 vary the positions of manual pump (T4), rainwater harvesting (T5), and rock pond (T6) alternatives. Thus, comparison values belonging to these attributes should be carefully allotted, to obtain trustworthy rankings.

Fig. 9. Sensitivity analysis results.

58

V. H. Honkalaskar et al. / Water Policy 16 (2014) 39–61

7. Main findings and conclusion 7.1. Main findings 1. Villagers have advantages over the external agencies in their knowledge of their complex and diverse livelihoods and local ecology, and the management of local resources to earn their livelihoods (Chambers, 1990). So, once they understood the decision making process, they could compare the technology alternatives and attributes in a better way than any external agency. For ‘operation ease’, people could imagine the scattered locations of the rainwater harvesting structures around the village, due to paucity of space within the residential area. Hence, they offered a lower weight to the rainwater harvesting alternative than to the electric pumping alternative, which has a centralized water tank.

2. The electric pumping alternative obtained the highest ranking. Two attributes, ‘water quality’ and ‘operation ease’, have contributed mainly to the final ranking. In fact, the reason behind its choice is the higher importance assigned to ‘operation ease’. In terms of ‘reliability’, ‘durability’, and ‘self-dependence to operate the project’, this alternative is relatively poor. For the remaining attributes, this alternative was assigned average comparison values. Nearly two decades after the move towards globalization, there is an increased drive towards unification of cultures and sameness of lifestyle (Lie & Lund, 1999; Pratheep, 2010). Households in the urban areas nearest the village receive domestic water either through taps in their homes or through nearby water standpipes, operated and maintained by the local municipal corporation. A local gynaecologist, Dr Nilakantha Phadke, revealed that over the past three decades, after the introduction of hybrid varieties of rice, the nutritional level of the food has deteriorated. This has been because the villagers have stopped eating ragi (which has high nutritional content) since it is more labour-intensive to grow than the hybrid rice. This, along with changing lifestyles, has resulted in increased health-related problems among the tribal women. Reduction of the local forest area has again increased the level of drudgery associated with the firewood fetching activity. The effect of this entire changed scenario on the villagers has been the changed aspiration of the tribal people to seek the easiest alternative of an electric pump and tank irrespective of the increased dependency on an external electricity supply.

3. Manual pump, rock pond, and rainwater harvesting are the next three rankings in the final hierarchy levels. These three alternatives do not differ much in terms of their final sum of weights. These alternatives are ranked better for ‘operational expense’, ‘durability’, and ‘reliability’. These three attributes were allotted average importance in the relative weights of attributes. The manual pump is also better for the ‘water quality’ attribute. Rainwater harvesting is better in terms of ‘operation ease’.

4. The gasoline pump holds fifth position in the final ranking list. It performs averagely in terms of ‘operation ease’, ‘water quality’, and ‘water purity’.

5. Biomass-fuelled steam engine pumping stands lowest in the final ranking list. It has been assigned a poor weight for all the attributes other than ‘water quality’ and ‘water storage’.

7.2. Conclusions In this study, the PAHP was formulated for community-level decision making by combining a few insights from PAR with AHP. We therefore make the following conclusions.

V. H. Honkalaskar et al. / Water Policy 16 (2014) 39–61

59

An experience of the community-level decision making process for the domestic water supply at the tribal village of Gawand-wadi has shown that it is possible to teach the basic concepts behind AHP methodology to enable the people’s participation in process of multi-criteria decision making. PAHP essentially strives to ensure people’s participation at different stages of the AHP process, namely, goal setting, comparisons, and synthesis.

By using local language and by invoking local culture, traditions, and values in the process of teaching AHP, the decision making process is simplified and the feeling of ownership of the overall process is increased.

Based on the experience and analysis of the two AHP iterations, the following methodology for the participatory AHP is proposed. (a) Project identification must be carried out through prioritization of people’s needs. People’s perceptions and desires should be captured to identify the community-level project. This should be followed by a rigorous analysis of the chosen problem to support and to integrate its importance and to plan further stages of the project. This increases the conviction of both the facilitating agency and the community towards completion of the project. (b) Teaching AHP methodology is a very crucial step. Only after increasing people’s curiosity about the scientific methodology for multi-criteria comparisons, can the basic concepts of AHP be taught with the help of a local example of a decision making process. This increases clarity and people’s participation in the whole process. (c) Different technology alternatives should be collectively selected as probable solutions to the problem in question. Each technology alternative should be discussed thoroughly. The alternatives that are not known to the people should be demonstrated to participants by showing pictures, video clips, and comparative assessments. Understanding the different technology alternatives is very important to enable participants to allot comparison values. (d) Different attributes for the alternative comparisons should be identified with the help of the people. For this activity, the facilitator should have a detached attitude and disposition. In the end, different attributes that are not identified should also be introduced. All attributes must be stated in terms of the local language so that they are understood by all the participants without ambiguity. (e) A ladder structure (Figures 7 and 8) can be utilized to compare technology alternatives and attributes. While assigning the comparative values, a discussion should be facilitated among the participants before freezing on the final locations over the ladder structure. This discussion not only increases the clarity for the participants, but also helps the facilitator to gain a few insights into people’s perceptions. (f) The process of calculation of the normalized comparative weights should be introduced to the people to convey the idea of the impossibility of mental calculations. It should be collectively decided to use computerized methods to carry out further processing and to prevent mystification.

The feeling of ownership so generated would be reflected in the final stage of acceptance of the results of the AHP exercise and their credibility for implementation. (g) The hierarchy order and sensitivity analysis should be communicated back to the people. Sensitivity analysis would show up the attributes that have a major contribution in the final hierarchy of alternatives. It is advisable to get feedback from the people and further iterate.

60

V. H. Honkalaskar et al. / Water Policy 16 (2014) 39–61

Following the community-level decision to implement the electric pump–pipe alternative, a committee of eleven members selected by the people in a village-level meeting was registered as a public charitable trust named ‘Gawand-wadi Vikas Sanstha’. A local volunteer organization, Jalavardhini Pratishthan, took responsibility to build a ferro-cement tank in the village. Villagers contributed in terms of human labour and local natural resources (sand from the local river, and space to erect the tank and a small room). Villagers contributed nearly 25% of the total cost. Households in the village were divided into three groups to carry out the labour work to build the tank and a small room. A total of 888 human labour hours were contributed by the people. Currently, the project is being operated by the trust through a monthly collection of Rs. 50 contributed by each household.

Acknowledgements The authors are grateful to the people of Gawand-wadi village for their participation and support throughout the research work.

References Ademiluyi, I. A. & Odugbesan, J. A. (2008). Sustainability and impact of community water supply and sanitation programmes in Nigeria: an overview. African Journal of Agricultural Research 3, 811–817.

Ananda, J. (2007). Implementing participatory decision making in forest planning. Environmental Management 39, 534–544.

Bank, W. (1994). The World Bank and Participation. SIDA Studies 2. Operations Policy Department, World Bank, Washington, DC. Bureau of Environmental Protection (2005). Well Yield Test Procedures. Drinking Water Program. http://publichealth.lacounty. gov/eh/docs/ep_dw_well_yield_procedures.pdf (accessed 23 October 2013).

Centre for the Study of Higher Education (2008). Participation and Equity. Technical Report. University of Melbourne, Melbourne, Australia.

Chambers, R. (1990). Participation and Equity. Technical Report. Sustainable Agriculture and Rural Livelihoods Program, International Institute for Environment and Development, Gatekeeper Series No. 22.

Chambers, R. (1994). The origins and practice of participatory rural appraisal. World Development 2, 953–969.

Chambers, R. (1995). Poverty and livelihoods: whose reality counts? Environment and Urbanization 7, 173–204.

Chambers, R. (2007). Ideas for Development. Earthscan, London, UK. Chowdhury, M., Begum, S. & Ahmed, T. (1996). Community Participation in Promoting Literacy Programme. Journal of Rural Development 26, 32–52.

Cornwall, A. (2000). Beneficiary, Consumer, Citizen: Perspectives for Participation for Poverty Reduction. SIDA Studies 2.

SIDA, Stockholm.

Date, A. W. (1989). Energy utilization pattern of Shilarwadi. Indian Journal of Rural Technology 1, 33–63.

Dhas, R. A. C. (2009). Agricultural crisis in India: The root cause and consequences. Munich Personal RePEc Archive Paper No. 18930, available at http://mpra.ub.uni-muenchen.de/18930/ (accessed 28 October 2013).

Elkarmi, F. & Mustafa, I. (1993). Increasing the utilization of solar energy technologies (SET) in Jordan: Analytic hierarchy process. Energy Policy 21, 978–984.

Gleitsmann, B. A., Kroma, M. M. & Steenhuis, T. (2007). Analysis of a rural water supply project in three communities in Mali: Participation and sustainability. Natural Resources Forum 31, 142–150.

Google Maps (March, 2011). http://maps.google.com/.

Greenwood, D. J., Whyte, W. F. & Harkavy, I. (1993). Participatory action research as a process and as a goal. Human Relations 46, 175–192.

Grundy, S. (1982). Three modes of action research. Curriculum Perspectives 2, 23–24.

Hall, B. (1981). Participatory research, popular knowledge, and power: a personal reflection. Convergence: An International Journal of Adult Education 14, 6–19.

V. H. Honkalaskar et al. / Water Policy 16 (2014) 39–61

61

Ikonen, M. (2011). About the Possibilities to Power Vehicles with Human Muscles. A lecture delivered at LLP Erasmus Intensive Program on Powering the Future with Zero Emission and Human Powered Vehicles.

Kangas, J. (1994). An approach to public participation in strategic forest management planning. Forest Ecology and Management 70, 75–88.

Kapur, A. (1995). Rural Water Supply. Seminar organized by The Institutes of Engineers, Nashik.

Leach, M. & Scoones, I. (2006). The Slow Race, Making Technology Work for the Poor. Technical Report. Demos, London.

Lie, M. & Lund, R. (1999). Globalisation, place and gender. AI & Society 13, 107–123.

Maguire, P. (1987). Doing Participatory Research: A Feminist Approach. The Center for International Education, University of Massachusetts, Amherst, MA, USA. National Sample Survey Organization (2010). Press Note on Income, Expenditure and Productive Assets of Farmer Households (January–December 2003). http://www.mospi.nic.in.

Oakley, P. (1991). The concept of participation in development. Landscape and Urban Planning 20, 115–122.

Oakley, P. (1998). Strengthening people’s participation in rural development. Occasional Paper Series no.1. Society for Participatory Research in Asia.

Paul, S. (1987). Community participation in development projects. Discussion paper 6, Centre for Development Research, Copenhagen.

Pearse, A. & Stiefel, M. (1980). Inquiry into Participation: A Research Approach. Technical Report. United Nations Research Institute for Social Development, Geneva.

Pohekar, S. D. & Ramachandran, M. (2004). Multi-criteria evaluation of cooking energy alternatives for promoting parabolic solar cookers in India. Renewable Energy 29, 1449–1460.

Pratheep, P. (2010). Globalisation, identity and culture: Tribal issues in India. In: LSCAC 2010 Proceedings. Faculty of Humanities and Social Sciences, Mahasarakham University, Thailand.

Raju, U. S., Nagraj, N. & Date, A. W. (1995). The influence of development perspectives on the choice of technology.

Technological Forecasting and Social Change 48, 27–43.

Reason, P. & Rowan, J. (1981). Human inquiry: a Sourcebook of New Paradigm Research. Wiley, New York.

Saaty, T. S. (1990). How to make a decision: The analytic hierarchy process. European Journal of Operational Research 48, 9–26.

Saaty, T. L. (2008). Decision making with the analytic hierarchy process. International Journal of Services Sciences 1, 83–98.

Saaty, T. S. & Vargas, L. G. (1982). The Logic of Priorities: Applications in Business, Energy, Health, and Transportation.

Kluwer-Nijhoff, Boston, MA, USA. Samad, M. (2002). Participation of the Rural Poor in Government and NGO Programs. Mowla Brothers, Dhaka, Bangladesh.

Schmoldt, D. L., Peterson, D. L. & Smith, R. L. (2007). The analytical hierarchy process and participatory decision making. In: Decision Support 2001. In: 17th Annual Geographic Information Seminar, Resource Technology 94 Symposium, Toronto, Vol. 1.

Scholl, A., Manthey, L., Helm, R. & Steiner, M. (2005). Solving multi attribute design problems with analytic hierarchy process and conjoint analysis: an empirical comparison. European Journal of Operational Research 164, 760–777.

Singh, U. K. (2005). Community Participation in the Management of Public Good: Myth or Reality. Case Study of Two Villages in India. Technical Report. ISS Research Paper, The Hague, The Netherlands.

Soma, K. (2003). How to involve stakeholders in fisheries management: A country case study in Trinidad and Tobago. Marine Policy 27, 47–58.

Srdjevic, B. (2005). Combining different prioritization methods in the analytic hierarchy process synthesis. Computers & Operations Research 32, 1897–1919.

Tiwari, M. K. & Banerjee, R. (2001). A decision support system for the selection of a casting process using analytic hierarchy processes. Production Planning & Control 12, 689–694.

Uphoff, N. T. & Cohen, J. (1979). Feasibility and Application of Rural Development Participation: A State of Art Paper.

Technical Report. Cornell University.

Vincent, L. F. (2003). Towards a smallholder hydrology for equitable and sustainable water management. Natural Resources Forum 27, 108–116.

Westergaard, K. (1986). People’s Participation, Local Government and Rural Development: The Case of West Bengal, India.

CDR Research Report 8. Centre for Development Research, Copenhagen.

White, G. W., Suchowierska, M. & Campbell, M. (2004). Developing and systematically implementing participatory action research. Archives of Physical Medicine and Rehabilitation 85, 3–12.

Received 20 January 2013; accepted in revised form 25 June 2013. Available online 4 September 2013

Water Policy 16 (2014) 62–78

Effects of the manager’s value orientation on stakeholder participation: at the front line of policy implementation Filip Aggestam European Forest Institute Central-East European Regional Office (EFICEEC), InFER – Institute of Forest, Environmental and Natural Resource Policy, University of Natural Resources and Life Sciences, Feistmantelstr. 4, 1180 Wien, Austria E-mail: filip.aggestam@boku.ac.at

Abstract Managers who implement stakeholder participation often have to navigate a complex subsystem of actors, policy-making institutions, and varying problem definitions. This paper examines how these managers’ values affect decision-making and the operationalisation of stakeholder participation, and how the institutional framework in which the managers are embedded affects these values. It is based on the inside views of 23 managers and expert consultants involved in nine projects implemented by international organisations. Their values and preferences were captured through a review of project documents and interviews. The results demonstrate that the managers’ personal value orientations affect the participatory process when there is a lack of control and support from their commissioning organisation, and also in cases where policy is ambiguous. The decision-making freedom accorded to the project manager defines whether they design stakeholder participation in accordance with personal value orientations, the organisation or policy. This study suggests that more stringent regulations and guidelines, as well as improved educational and awareness-raising activities, are required to resolve this problem.

It is also suggested that evaluation tools should be improved to account for the impact that stakeholders have. This may encourage managers to become more actively involved in the use of stakeholder input.

Keywords: Integrated water resources management; Project management; Stakeholder participation; Value orientations

1. Introduction It is often argued that real-world problems require the involvement of stakeholders, particularly within the context of natural resources management. In line with this argument, several international and

doi: 10.2166/wp.2013.135 © IWA Publishing 2014

F. Aggestam / Water Policy 16 (2014) 62–78

63

European policies require that stakeholders1 are invited and integrated into the process of managing natural resources (Sadoff & Grey, 2002; Reed, 2008). As stated in the Rio Declaration, ‘environmental issues are best handled with the participation of all concerned citizens, at the relevant level’ (part of Principle 10)2, while organisations such as the United Nations Development Programme (UNDP) and the Global Environment Facility (GEF) have increasingly promoted stakeholder participation.

Underlying this trend is the belief that participatory methods can improve how we manage natural resources (Failing et al., 2004), which reflects a system of management that is replacing the conception of environmental management as a technical process best left to experts (Creighton et al., 1998; Hare & Pahl-Wostl, 2002; Reed, 2008).

Despite efforts to move away from a technocratic approach to natural resource management, many obstacles still emerge with stakeholder participation in practice. These include issues such as shortage of time and resources (Taut, 2008), data or value conflicts (Creighton et al., 1998), and biased stakeholder involvement (Urwin & Jordan, 2008; Vugteveen et al., 2010). The obstacles illustrate a gap between theory about participation, on the one hand, and actual practice and policy outcomes, on the other. For instance, in practice, the project manager (PM) (or ‘street-level bureaucrat’) who implements stakeholder participation policy often has to navigate a complex subsystem of actors and policy-making institutions that hold conflicting goals and problem definitions (Lipsky, 1980; Sandström, 2011). This contextual background to policy implementation (e.g. existing belief coalitions and policy beliefs) may affect how the PM frames stakeholder participation (e.g. valuation of stakeholder input) (Sabatier, 2007; Buijs, 2009). This may, in turn, influence the manager’s operational decision-making on such issues as how stakeholder participation is designed (Hill, 2003; May & Winter, 2007).

The aim of this paper is to improve our understanding of how PMs interpret policies on stakeholder participation and, more specifically, how value orientations affect the operationalisation of stakeholder participation. The objectives are to analyse how PMs perceive stakeholder participation, establish whether value orientations affect the design of the participatory process, and examine the role stakeholders are allowed to play in projects. The analysis also investigates effects of the institutional design and corporate culture on decision-making and the discretion provided to the manager when implementing policy.

2. Theoretical and conceptual background There are many international and European policies and conventions that address stakeholder participation in water resources management. One prominent example is the EU Water Framework Directive (WFD)3 requiring projects to take into account stakeholders’ views. Other examples are the Aarhus Convention4 and the RAMSAR Wetlands Convention5. Yet, despite legal requirements, stakeholder participation remains weakly defined in international and European policies. The WFD, for instance, The ‘stakeholder’ is defined as any person, group or organisation that has an influence on or interest in a project and/or is affected directly (or indirectly) by its decision-making (Freeman, 2010).

2 See http://www.unep.org/Documents.Multilingual/Default.asp?documentid=78&articleid=1163.

3 See http://ec.europa.eu/environment/water/water-framework/.

4 See http://ec.europa.eu/environment/aarhus/.

5 See http://www.ramsar.org.

1

64

F. Aggestam / Water Policy 16 (2014) 62–78

does not define how stakeholders’ views should be taken into account. As a consequence, the diffuse wording in policy may allow for variations in implementation. The presumption is that policy implementation in the area of stakeholder participation may be framed by the PM or organisation in charge (Dewulf et al., 2004; Urwin & Jordan, 2008). Framing refers to the values, preferences and perspectives that PMs and/or organisations rely on to understand and define stakeholder participation (Buijs, 2009). It is thus the PM, at the front line of policy implementation, who defines what these policies mean at an operational level (Hill, 2003; May & Winter, 2007).

However, before entering into further discussion on stakeholder participation, it is also relevant to introduce a distinction between what is meant by the PM (‘street-level bureaucrat’) in charge of conceptualising and managing the project and the expert employees or consultants (‘front-line workers’) in charge of operationalising policy (May & Winter, 2007). Both have an impact on stakeholder participation: the streetlevel bureaucrat in terms of interpreting policy and designing the project and the participatory method, the frontline worker in terms of the actual hands-on implementation of the (pre)designed project. As both street-level bureaucrats and frontline workers have an effect on policy implementation, the analysis will distinguish between them in cases where different persons take these roles.

2.1. Stakeholder participation at the front line It was argued by Lipsky (1980) that the street-level bureaucrat (a role fulfilled here by the PM) creates the policy that citizens will experience. This process is shaped by both the implementer’s knowledge (Yanow, 1996) and by contextual factors such as the institutional framework (Sabatier, 2007). Project activities are thus influenced by different belief coalitions (e.g. diverging coalitions in a region, host organisation or project) that generate competing views on policy problems (vertical and horizontal complexity) and varying interpretations of the legal framework and policy solutions for stakeholder participation (Sandström, 2011). Analysing how stakeholder participation is implemented allows for an assessment of how policy is interpreted in practice (not the actual wording of legislation) and how it is influenced from the bottom-up (the policy implementers’ perspective).

Here it is presumed that stakeholder participation is introduced into practice as a response to policy demand (external adaptation) and is not necessarily internally enforced due to prevailing organisational practices (internal integration) (Schein, 1985). This means that an organisation may adhere to changing policy demands out of necessity (e.g. operational requirements), rather than having an actual belief in the added value of changing practice (May & Winter, 2007). The institutional framework (e.g. formal rules, distribution of power and organisational values) might therefore affect how the PM integrates stakeholders into the process (Hill, 2003). Lipsky (1980) contends that it is the nature of the work itself that empowers the manager with policy-making abilities, and that the conditions of the project environment dictate whether the manager can implement policy according to personal preferences.

This refers to what Wierzbicki et al. (2000) call a soft decision-making approach, often applied by managers when making decisions on a project: the manager attempts to ‘perceive the whole picture’ by observing the problem area from various angles and making decisions based on expert intuition, such as on how stakeholders should be involved (Bruno & Lay, 2008; Moxley et al., 2012). The freedom that this approach affords might enable managers, even when embedded within the same institutional framework, to make different decisions from one another (Hill, 2003). This suggests that the PM also makes value judgements and executes decisions based on personal values (Hall & Davis, 2007).

F. Aggestam / Water Policy 16 (2014) 62–78

65

2.2. How values shape stakeholder participation Although values can be the primary source of environmental conflicts (Creighton et al., 1998), they are difficult to conceptualise in project management, particularly as values are prominently discussed in the literature, in which many definitions and effects are described. This study adopts the Meglino & Ravlin (1998) definition of values, namely that they provide the basis for an individual’s internalised beliefs about how a person should behave (their value orientation). Values are effectively the basic components of beliefs that form an individual’s value orientation (Kaltenborn & Bjerke, 2002) and which influence the interpretation of experiences, facts and events (Stern et al., 1999; Vugteveen et al., 2010) and shape the motivational structures (e.g. preferences and perspectives) that subsequently affect decision-making (Bruno & Lay, 2008).

The relationship between value orientations and policy implementation is less clear. PMs can, for instance, switch between the values on which they put an emphasis. This could depend on contextual factors (e.g. personal or professional values) and the importance assigned to specific value orientations (e.g. organisational or public good), which makes them decide differently (Bruno & Lay, 2008; Aggestam, 2013). The project environment, as defined by the contextual background and policy, also has an impact on how the PM can (or is allowed to) express values in decision-making (Appelstrand, 2002). It is, nonetheless, clear that the manager has to interpret directives from policy, and balance stakeholder interests, project objectives and the host organisation’s interests. The PM effectively becomes the ‘normative’ gatekeeper (or filter) that determines how policy is implemented at the front line (see Figure 1).

As illustrated by Figure 1, the inclusion of new values and knowledge (organisational or public) can have an impact on the manager’s preferences and perspectives. The introduction of new values could, for example, expose biased perspectives that in turn affect the PM’s decision-making (Raymond et al., 2010; Moxley et al., 2012). This means that the level of involvement corresponds to a value statement by the PM, and, more specifically, how open the manager is to stakeholder input (unless dictated by the host organisation or policy). Value orientations can thus be expressed through different levels of involvement, ranging from information provision to active involvement during project implementation (Arnstein, 1969; Vugteveen et al., 2010). This contextual background for stakeholder participation presupposes freedom (or discretion) that has been provided for in the design of the participatory process.

Fig. 1. Conceptual model: normative gatekeeping by the manager.

66

F. Aggestam / Water Policy 16 (2014) 62–78

The manager is assumed to be striving to control the process in accordance with their own value orientations in order to maintain what they perceive as the status quo, such as a technocratic approach to project implementation (Stern et al., 1999; Castelletti & Soncini-Sessa, 2007). The emphasis in the analyses is on how the PM constructs a meaning out of policy, which shall provide insights into how their value orientations impact policy implementation.

3. Materials and methods 3.1. Background and case studies This paper reviews projects implemented by the UNDP, the International Commission for the Protection of the Danube River (ICPDR), the International Institute for Applied Systems Analysis (IIASA), and partner organisations. Nine projects were selected and grouped into three separate cases on the basis of who implemented the projects: more specifically, the same host organisation(s), the same PM(s) (or street-level bureaucrats) and the same project teams (or front-line workers). All projects were at the forefront of stakeholder participation, where they took a leading position on participatory methods. Individual projects will be highlighted only to illustrate specific issues as regards stakeholder participation.

Table 1 presents background details for Case I.

• Case I. ICPDR and UNDP projects based on the TDA and SAP methodology Five of the nine projects in the Danube and Tisza river basin were implemented by the UNDP and ICPDR. All applied the transboundary diagnostic analysis (TDA) and strategic action programme Table 1. Study background.

ICPDR and the Danube basin Background

Function

Stakeholder Participation

Format

ICPDR is a transnational body (or commission) established to implement the Danube River Protection Convention. It comprises a delegation of all contracting parties to the convention.

The commission functions as the platform for implementing the transboundary aspects of the EU Water Framework Directive (WFD).

ICPDR recognises that participation is multidimensional and their strategy focuses on spatial scales (international, national and local level) and on progressive levels of involvement (ICPDR, 2006).

The progressive levels of involvement are based on WFD Article 14 and defined as information supply, consultation and active involvement (e.g. policy deliberations) (ICPDR, 2009).

Transboundary diagnostic analysis (TDA) and strategic action programme (SAP) methodology The TDA is a diagnostic and adaptive management tool used to identify the cause-and-effect relationships for transboundary water problems.

The TDA functions as a tool for the development of a SAP that is then used for project implementation.

The TDA should include full consultation and stakeholder participation (GEF/C.7/6.).

The method prescribes/defines steps, starting from project conceptualisation to joint fact-finding and the SAP. The TDA/SAP does not operate as a prescriptive methodology but leaves room for flexibility to be adopted to local conditions (UNDP, 2002; Teng, 2006).

F. Aggestam / Water Policy 16 (2014) 62–78

67

(SAP) methodology, and received financing from GEF (see Table 1 for more details). The five projects6 focused on integrated land and water management, or rural water management, and/or the conservation and restoration of biodiversity in the Danube and Tisza river basin. They were also transboundary and participatory. The analysis will focus on how the PMs applied the TDA and SAP method to achieve operational requirements for stakeholder participation.

• Case II. IIASA research-based projects Two further projects in the Tisza river basin were implemented by IIASA. The first project explored the set-up of an adaptive management framework for the re-naturalisation of the Tisza river basin by allowing stakeholders and scientists to collaborate in the research into and the revision of policy and local practices (Sendzimir et al., 2006, 2007; Krolikowska et al., 2007). The second project explored stakeholder views on flooding in the Upper Tisza river basin. The results were to be integrated into a flood catastrophe model to develop policy alternatives for flood insurance (Vári, 2001; Vári et al., 2003; Linnerooth-Bayer & Vári, 2006).

• Case III. Caspian Sea and Kura–Aras basin Finally, two projects in the Caspian Sea and Kura–Aras river basin were implemented by the UNDP. They were included in the analysis based on input provided by frontline workers from Case I. The first project was a stakeholder analysis for the Caspian Environment Programme (CEP), as part of its participatory strategy, which sought endorsement among the Caspian states. It further sought to define the environmental problems facing the region (CEP, 2002). The second project aimed to reduce environmental degradation in the Kura–Aras basin (UNDP/GEF, 2005) and was designed to involve stakeholders in improving the provision of clean drinking water and to reduce the degeneration of natural resources.

Table 2 presents a summary of the main elements of the case studies.

Table 2. Summary of case studies.

Type of projects Background and format

Stakeholder participation

6

Case I

Case II

Case III

Development-driven projects.

Research-driven projects.

Development-driven projects.

Stakeholders were engaged as part of the TDA/SAP methodology and operational requirements from the funding agent.

Information provision, manipulation and consultation.

Stakeholders were engaged in the conducting of research into participatory methods.

Active involvement.

Stakeholders were engaged as part of the TDA/SAP methodology.

Distinct from Case I as it concerns different geographical regions.

Information provision and active involvement.

The five projects are Tisza Community-Led Demonstration Project for Sustainable Development and Integrated Land and Water Management; Integration of Rural Water Management in River Basin Management in the Tisza Basin; Establishment of Mechanisms for Integrated Land and Water Management in the Tisza River Basin; Conservation and Restoration of the Globally Significant Biodiversity of the Tisza River Floodplain through Integrated Floodplain Management; and Transboundary River Basin Management of the Körös/Crisuri River Project. See http://waterwiki.net for more information.

68

F. Aggestam / Water Policy 16 (2014) 62–78

3.2. Data collection and analysis This study draws on document analysis, a series of face-to-face interviews, and a focus-group discussion through an online forum. Secondary data were collected through project documents, published information on the projects, and peer-reviewed articles. Data were collected in 2006 through a crossregional UNDP conference ‘Stakeholder Management in Water Projects’ (held on 17 November 2006 for the UNDP Water Knowledge Fair). This conference provided access to and input from experts in the field, as well as a platform for the online forum. This was followed up with another round of data collection throughout 2007 and 2008 that focused on interviewing PMs and expert consultants. Some stakeholders were also interviewed. However, as this study analyses a rather small sample (see Step 2 below) of projects/cases, it refrains from generalisation beyond these projects.

The interviewees and focus-group participants were PMs (street-level bureaucrats) and expert consultants (front-line workers) who hold, in common, an expertise (e.g. technical knowledge on stakeholder participation) and leadership (e.g. control of policy implementation). All the participants worked for either UNDP, ICPDR, IIASA, the Regional Environmental Centre for Central and Eastern Europe, Nimfea Environment and Nature Conservation Association, the CEP or the Tisza Biodiversity Programme. The managers’ leadership position forms the basis for investigating how value orientations are expressed in the design of participatory processes. Data were collected and analysed in three steps: Step 1. Comparative analysis of public project documents and publications

• Document review and comparative analysis aimed to define the problem area (e.g. direction provided by policy and the host organisation). It was also carried out to identify PMs and expert consultants, and to find information in preparation for the interviews. Project documents were identified through the host organisations and project websites (e.g. http://waterwiki.net and http://caspian.iwlearn.org/). External information was also collected through peer-reviewed articles using science-based search engines (e.g. www.sciencedirect.com) and specific key terms (e.g. Danube basin, water resources management).

Step 2. Semi-structured interviews and forum discussion

• Individuals participated in face-to-face interviews, and an additional 11 participants took part in the forum-based focus-group discussion. Data collected through the forum provided information on how stakeholder participation was being implemented by the host organisation(s). The interviews were semi-structured and covered three thematic areas: (a) project environment (e.g. design and objectives); (b) project implementation (e.g. how stakeholders were engaged); and (c) value orientations associated with the other two thematic areas (e.g. why stakeholders were engaged). Questions were not asked in a standardised manner, and, instead, the respondents’ reactions and answers guided the conversation, which allowed for in-depth discussions. Representatives of stakeholder groups were also included. A typical interview lasted between 1 and 2 hours.

Step 3. Interview transcription and analysis

• Forum discussion and transcribed interviews were merged to produce a structured document containing critical responses. This was achieved by dividing all the responses into categories that correspond to the

F. Aggestam / Water Policy 16 (2014) 62–78

69

study objectives. The category system covered: (i) project success and problems; (ii) views on the project and the institutional framework; and (iii) views on stakeholder participation. General observations and the degree of importance allocated to each statement were also noted. The interviewees were asked about the degree of importance (very to least important) during the interviews. After the initial categorisation, critical responses were identified and clustered according to similarities. This approach provided a hierarchy of responses within the dataset and a comprehensive representation of value statements across the cases.

4. Results 4.1. How do PMs perceive stakeholder participation?

Before proceeding to the cases, the managers’ definition and rationale for a stakeholder analysis will be considered briefly. The purpose is to address one of the study’s objectives: to shed light on how PMs perceive participation and the role of stakeholders therein. When asked to define a stakeholder analysis, PMs provided a surprisingly wide variety of answers. These definitions of stakeholder participation ranged from: a process by which you ‘identify people in the broader sense who have an interest’ [I2]7 with the aim of involving stakeholders (active involvement and partnership), to conducting an empirical measurement of ‘perceptions in terms of causes and relationships’ [I7] with the aim of understanding how stakeholders are affecting the environment (informing and consultation), to ‘looking at power relationships between stakeholders and where they sit in a network’ [I9] with the aim of influencing key stakeholders (informing and manipulating). While these definitions correspond to different ways of analysing stakeholders, they also show the type of participation that would follow these analytical steps. Disregarding external limitations (e.g. financial constraints) and the contextual background (e.g. the institutional framework), these definitions equate to different levels of stakeholder involvement that are similar to specific steps of the classic ladder of participation defined by Arnstein (1969), which ranged from active involvement to non-participation (as noted in the brackets above).

The different definitions of how to analyse stakeholders (and the implication for participation) are important to consider because of the conceptual ambiguity that surrounds stakeholder participation in policy and the many purposes for which stakeholder participation is employed in practice. This demonstrates that the participatory process is dependent on the manager’s value orientation. In fact, it was stated by one interviewee that the ‘law on participation can only be guided by what the project manger wants stakeholder involvement or engagement to achieve’ [I5]. This supports the presumption made earlier with regard to participation being controlled by the manager, despite legal requirements for policy to be inclusive.

4.2. Case I. ICPDR and UNDP projects based on the TDA and SAP methodology 4.2.1. The project and institutional framework. ICPDR, in collaboration with the EU, UNDP and other Danube countries, has promoted an integrated approach to the management of the Danube river basin since 1998, when ICPDR was created. There has been a strong emphasis on stakeholder participation following ICPDR’s first stakeholder conference in 2005 (ICPDR, 2006). For the projects in this study, ICPDR and UNDP operationalised their own policy on stakeholder participation through the 7

Information in square brackets after a quotation refers to the coding given to interviews.

70

F. Aggestam / Water Policy 16 (2014) 62–78

TDA and SAP methodology, as required by the GEF. It comprised the main framework for the stakeholder analysis and public involvement (UNDP, 2002; Teng, 2006). The TDA and SAP framework furthermore prescribed steps for specific actions (e.g. policy, legal or institutional reforms) to reduce transboundary water problems for each project.

Despite the host organisation’s public emphasis on active involvement, PMs were granted the discretion to decide on the level of stakeholder involvement. Several interviewees noted similar statements, such as that ‘actively involving them [the stakeholders] as part of the decision-making process was only encouraged [by the organisation]’ [F]. This confirms that stakeholders were, for the most part, involved because of operational requirements, and not because of the benefits that stakeholder participation can generate. Key stakeholders were not even allowed to participate in some projects, to avoid conflicts and delays, and/or to facilitate consensus. Referring to the TDA and SAP methodology, it was stated that the ‘weights of the different [project] elements were decided by the manager, and the manager can shift funding from one section to another depending on [their] preferences’ [I5], demonstrating a financial discretion linked to the host organisations’ lack of control during project development. In effect, this implies that stakeholder participation was controlled through both design and restricted financing by the PM. 4.2.2. Rationale for stakeholder participation. Most of the PMs and expert consultants agreed that stakeholder participation was vital, especially to resolve disagreements on project objectives, something which applied to all projects. Most of the managers did, however, prefer a one-way communication process (e.g. input through consultation) rather than having to invest in a two-way process (e.g. partnership). Furthermore, several PMs did not know who the key stakeholders were during project conceptualisation. As a result, many key stakeholders had nothing to do with project implementation, nor were they involved at a stage when the project design could be changed. This is a problem that has been emphasised in the literature, namely that stakeholders are not involved early enough in the project cycle (Brody, 2003; Schusler et al., 2003). The interviewees’ most common argument for not involving stakeholders was that stakeholder input in decision-making was considered negative to project performance. These perceptions and/or assumptions were often supported by practical arguments, such as limited funding, difficulties in persuading governments, the host organisation’s lack of support for stakeholder participation, and the difficulty of finding a representative list of stakeholders. It was also often noted that stakeholders had a narrow and local view of the problem area.

In contrast to the PMs’ perspective, responses from stakeholders illustrate differences between how the PMs and stakeholders defined the problem area. Among stakeholders there was a prevailing disappointment, in that they were not allowed to influence decision-making, nor provide input at relevant stages of the project cycle. This suggests that many PMs only aimed to fulfil operational requirements, such as providing information and/or consultation. In effect, PMs (and the host organisations) preferred a literal interpretation of the WFD, which only requires that stakeholders are heard and considered. In practice, this meant that stakeholders were heard, but not considered.

While all projects involved some elements of participation, PMs avoided active stakeholder involvement (e.g. inclusive decision-making) during project conceptualisation and in the early stages of the project cycle. Several managers even expressed unease at going beyond the minimum operational requirements for stakeholder participation. Relinquishing control through stakeholder participation was opposed, suggesting an unease at the changing of power dynamics. In fact, most managers preferred neither to challenge their own perception of the problem area nor to engage stakeholders once the projects were up and running, leaving no alternatives to change project implementation. This suggests a

F. Aggestam / Water Policy 16 (2014) 62–78

71

prevailing technocratic perspective on project management that does not challenge the managers’ own value orientation.

4.2.3. Shaping participation policy implementation. From the results presented above, it is clear that the TDA and SAP, EU policy, and the operational requirements by GEF provided the official driving force for involving stakeholders. It was surprising to find that the project environment still allowed for significant operational variations as regards how stakeholder participation was implemented, despite guidelines provided through the TDA and SAP. In many cases the reluctance towards participation was because it challenged the managers’ perception of the problem area and/or their position of power. This caused them to take steps to limit the impact that stakeholders could have, by reducing the funding for stakeholder participation, which effectively limited the level of involvement (from inclusive to non-participatory). Alternatively, they controlled requirements for consensus-building by only inviting specific stakeholders who did not jeopardise the PMs’ vision for the project or create conflicts. Managers also designed the project so that the participatory process came at a late stage of the project cycle, making it impossible for stakeholders to influence the outcome and/or design of the project. These steps were common measures that PMs applied to maintain control and/or impose value orientations that were contradictory to those held by stakeholders. This is also linked to the institutional arrangement, since the host organisations did not impose policy directions, and the institutional framework allowed for discretion during policy implementation. All in all, serious concerns are raised for the legitimacy of the projects’ participatory processes.

4.3. Case II. IIASA research-based projects 4.3.1. The project and institutional framework. In 2006, IIASA implemented two research-oriented projects in the Tisza river basin. The first project collaborated with scientists and non-governmental organisation (NGO) activists to develop tools for communicating complex ideas and creating qualitative or quantitative models for stakeholders (Sendzimir et al., 2006, 2007; Krolikowska et al., 2007). The second project was summed up by one of its contributors as investigating situations ‘where you can accept one solution for different reasons’ [I8]. It aspired to explore discourse on flooding and why flooding constitutes a problem for some but not for others (Sendzimir et al., 2006, 2007).

Owing to the research objectives of the project, and the contextual background of being implemented by a research-driven organisation, IIASA associated different values with stakeholder participation as compared to UNDP and ICPDR. One interviewee noted that IIASA engaged stakeholders in discussions of paradigms and mental models. The key to IIASA’s approach was to respect differences – ‘not to come up with a consensus view … and not imposing a ‘right’ and ‘wrong” [I8] – in the participatory process. So, while the host organisation did not impose more control over the project design, the institutional framework (e.g. informal customs and norms) encouraged stakeholder participation and the inclusion of other value orientations.

4.3.2. Rationale for stakeholder participation. Stakeholder participation was central to the IIASA projects, which required active stakeholder involvement and commitment from the PM. These managerial considerations (different from Case I) support the assumption that stakeholder participation is not marginalised if the PM agrees with policy on stakeholder participation. In this case, it was even argued that managers (and stakeholders) should adapt their behavioural filter (or mental models), to see beyond their

72

F. Aggestam / Water Policy 16 (2014) 62–78

own value orientations. This highlights an important difference in that managers need to associate positive values with stakeholder participation for it to be successful. In this case, the involvement of stakeholders was seen as crucial rather than as a burden or barrier to project performance.

It is of further interest to note that the managers in this case still preferred to involve stakeholders after project conceptualisation. So even when a project focuses exclusively on participation, there are managers who are not open to challenging their value orientation (or perspectives) as regards to how a project should be designed. This reluctance was not linked to power dynamics (as in Case I) but rather the managers’ expertise. There is thus prevailing support for a technocratic approach, not for the whole project cycle but at least for how a project should be conceptualised. For instance, a prevailing argument for not involving stakeholders (in both cases) remains linked to costs and benefits, especially when discussing early stakeholder involvement. Finally, as noted by one interviewee, the IIASA projects also demonstrate that active stakeholder involvement might not always be realistic, considering the time and financial resources that need to be invested.

4.3.3. Shaping participation policy implementation. The trade-offs required to actively involve stakeholders are significant (e.g. time and financing). There is consequently a correlation between the managers’ willingness (or openness) to engage stakeholders and the arguments, such as the cost/benefits of stakeholder participation, being applied to marginalise stakeholder participation. This case further substantiates the impact that managers and organisations can have on policy implementation. For participation, the results highlight either that the manager or host organisation needs to value stakeholder participation, or that policy on participation needs to become stricter, to ensure that stakeholder input is integrated into the project design. It was, however, surprising to find that even those managers with a strong belief in the benefits that stakeholder participation generate are reluctant to relinquish control over how a project is conceptualised. Thus, when considering the similarities between the cases (e.g. same region, policy background and discretion provided to managers), it is clear that the managers’ framing of stakeholder participation (whether positive or negative) ultimately determines how stakeholder input is utilised.

4.4. Case III. Caspian Sea and Kura–Aras basin 4.4.1. The project and institutional framework. The organisational and institutional background for this case is similar to that of Case I; however, it differs in that the expert consultants were given more freedom in designing the participatory processes. The first project concerned a stakeholder analysis for the CEP. The analysis aimed to identify any conflicts among stakeholders that could constrain effective interventions, and explore how stakeholders prioritised environmental and social problems that had been identified by experts (CEP, 2002, 2006). The second project was implemented in the Kura–Aras basin by the same manager and expert consultants. It had a more specific objective, namely, to find solutions for generating a safe supply of drinking water.

4.4.2. Rationale for stakeholder participation. The CEP was launched with a stakeholder analysis as part of a TDA and SAP, but the analysis moved away from being an investigative tool to becoming a participatory process. It was noted that, after the regional analysis of the Caspian Sea, ‘the notion of stakeholders was expanded from the standard focus’ (Matthews, 2004, p.10), as defined by the TDA and SAP. The participatory process that developed out of the stakeholder analysis resulted in the

F. Aggestam / Water Policy 16 (2014) 62–78

73

CEP objectives being reprioritised. ‘Stakeholder groups rated some concerns much higher than experts, whereas others that the experts believed to be the most prominent were ranked far lower than expected’ (Matthews, 2004, p.10). This resulted in the manager changing the expert-based priorities in line with stakeholder input, which seems to have had a positive impact on project performance (e.g. in terms of acceptance by stakeholders). The participatory process was also conducted at the beginning of the project cycle, making it possible for the programme to accommodate new information.

Based on the frontline workers’ success with the CEP stakeholder analysis, the manager allowed more latitude for stakeholder participation in the Kura–Aras project. This resulted in the consultation being held with invited representatives, hand-picked based on personal experiences to represent a wide range of stakeholders. The number was kept low to facilitate a more detailed discussion. The recommendations that were generated through this approach were perceived as innovative by the PM and were incorporated into the project design (UNDP/GEF, 2005). These recommendations were mainly due to the different perspectives of the problem area (given by the stakeholders, government and NGOs involved) and were incorporated owing to the manager’s willingness to include new (and alternative) input. This reconfirms the assumption that it is only when the manager does not care about avoiding conflicts, or maintaining their position of power, that the participatory process can yield a truly positive outcome.

4.4.3. Shaping participation policy implementation. This case was included as a showcase for the positive impact that stakeholder participation can have on project performance. The results support the argument that involving stakeholders during project conceptualisation is beneficial for improving not only project design but also (as supported by previous research) the longevity of a project (Brody, 2003; Schusler et al., 2003). It was noted that ‘a great deal was won when the stakeholder analysis and consultation were done even before the project started, when there was flexibility to fully account for [stakeholders’] concerns’ [F]. The Kura–Aras project also demonstrates that the selective inclusion of stakeholders can have a positive impact on project performance (e.g. reduced risk for conflicts), but also raises concerns as to how legitimate (and representative) the selection process actually is, and over its susceptibility to personal biases. The results do, however, reiterate that the willingness to change personal value orientations is essential to ensure that stakeholder input is integrated into the project design. Furthermore, results from both projects show that relying explicitly on expert-driven input only is not sufficient to come to a socially accepted definition of a problem area (Raymond et al., 2010).

4.5. How did the managers control stakeholder participation?

One manager defined a prevailing practice for stakeholder participation, stating that ‘it has become best practice to create an opportunity and something which the public can see us do, even if it is only data collection wrapped up as participation’ [F]. This provides a straightforward example of how managers avoid active involvement by relabelling activities that are part of the traditional technocratic approach to project design. By relabelling activities, the PM and host organisation seemingly comply with operational requirements set out in policy, circumvent active stakeholder involvement and maintain the status quo. As demonstrated by Case I, this is linked to an inherent reluctance among PMs to incorporate input that goes against their own expert-based assessment of the problem area of a project. Most PMs did, however, fulfil policy and operational requirements by designing the project so that it involved stakeholders and responded to input, such as in Cases II and III.

74

F. Aggestam / Water Policy 16 (2014) 62–78

The prevalent negative value orientations associated with stakeholder participation in Case I (e.g. participation primarily perceived as a source of conflicts) meant that more than half of the managers were taking steps to enforce their value orientations on the project design. Many of these managers conducted a consultative process rather than active involvement. This signifies that they committed only to a oneway communication process and decided what to do with stakeholder input, which allowed them to collect input while restricting the stakeholders’ influence on the project design and/or activities. Another strategy was to selectively invite stakeholders in order to avoid conflicts. In fact, close to 70 per cent of the managers admitted to having identified important stakeholders that were not engaged in their project(s). Rather than involve a representative set of stakeholders, many managers invited stakeholders with a pre-existing relationship to the manager and/or the organisation. This strategy may be legitimised by its potential to reduce costs (as in Case III), but the selective inclusion was often carried out to minimise the risk of conflicts or to maintain control (as in Case I). The most common measure to control stakeholder influence was to place participatory elements at a late stage of the project cycle. Stakeholder participation was, as a matter of fact, most often carried out in Case I when the budget had already been earmarked (e.g. for expert recruitment) for projects. The end result was that, if any stakeholder came up with something new, it was nearly impossible to fit it into the project structure.

The principal motivational factors that influenced (or allowed) the above-noted decision behaviours were the organisational and policy framework, as well as the PMs’ personal values, sense of control and power. It was, for instance, noted that projects are expected to be inclusive and to build consensus, but the managers in Case I rarely received the organisational support required to take stronger action. This contributed to maintaining the status quo – a technocratic approach to project management. Taking another perspective, the policy framework on stakeholder participation and the lack of stricter regulations and control mechanisms may improve decision quality as they allow for flexibility in project management (Appelstrand, 2002).

However, for the managers this flexibility provided the motivation and discretion to introduce their own value orientations against stakeholder involvement during project conceptualisation. One interviewee stated that ‘the laws fail to define the relationship and obligation between the users [the stakeholders] and the PM’ [I12]. This kind of reasoning was often coupled with depicting stakeholder participation as a restriction on project performance, rather than as a process that promises positive implications. In these cases the managers’ personal values can also be linked to a prevailing unwillingness to share control and power. Nearly 55 per cent of them stated that no direct beneficiaries were allowed to participate in project management. Several managers did in fact note that stakeholder participation implies a changing power structure, which was perceived as negative (or as a threat) to the managers’ vision for the project.

To conclude, it should not be forgotten that stakeholders’ participation faces other barriers too, such as limited resources and stakeholder unwillingness. The results do not allow any conclusions on the relative weight of these factors; however, in all projects where the managers seemed to think that stakeholder participation limited their work, value orientations were found to influence the project design. In effect, when polices on participation are imposed without internal adaptation by the implementing organisation, PMs marginalise stakeholder participation based on their personal value orientations.

5. Discussion and conclusion This paper has relied on the inside views of PMs and expert consultants working for international organisations at the front line of policy implementation. The first objective was to analyse how managers

F. Aggestam / Water Policy 16 (2014) 62–78

75

frame stakeholder participation. As demonstrated by the empirical analysis, this cannot be separated from questions concerned with if (and how) the PMs’ value orientation affected the design of the participatory processes, and what role stakeholders were allowed to play in decision-making. Results from the case studies illustrate that, when PMs associate stakeholder participation with negative value orientations, they operationalise this by limiting the level of stakeholder involvement, such as only providing information or allowing consultation, or relegating the participatory process to a later stage in the project cycle. These types of measures allowed the PMs to comply with policy and operational requirements without having to change established practices. The impact of the managers’ value orientations is in line with previous studies on leadership and personality traits (Bruno & Lay, 2008; Moxley et al., 2012), which suggest a spectrum of management personalities, ranging from individualistic to cooperative and competitive. It is reasonable that the PMs’ personality and aspirations cause them to frame stakeholder participation with either negative or positive value orientations. These value orientations are, however, more similar to world-views that incorporate a broader range of values, which is why the concept of frames was applied from the onset. For example, ‘individualistic’ PMs are assumed to frame stakeholder participation as a threat to their control and power, which makes them adopt measures that circumvent this perceived threat (Wierzbicki et al., 2000). These personality traits and aspirations differ not only in their underlying values but also in their effect on how projects are managed, and how policies and problem areas are interpreted (Castelletti & Soncini-Sessa, 2007; Moxley et al., 2012).

Previous research about values among street-level bureaucrats also supports the findings that value orientations have a significant impact on project management (Lipsky, 1980; Hall & Davis, 2007; Aggestam, 2013). The diversity of value orientations regarding stakeholder participation is further reflected in the range of definitions that were provided for a stakeholder analysis. They show that PMs and front-line workers have quite different preferences for the participatory process in terms of impacts and outcomes.

The final objective of this paper was to address the effects of the institutional framework on both the PMs’ decision-making and stakeholder participation. The results demonstrate that the discretion provided by the host organisation charged with implementation and the lack of organisational support have an impact on how the managers’ value orientations are expressed. Effectively, the discretion accorded to the PMs defined whether they acted in accordance with their personal value orientation, the organisations’ values or policy rules. This corresponds to a fluid application of values that is dependent on contextual variables, such as institutional design and perspectives on stakeholder participation.

This is particularly apparent when organisations adapt to new policy demands that are ambiguous and leave room for interpretation. As the results have shown, the content of policy on stakeholder participation (e.g. guidelines) influences the interpretation and implementation. It may prescribe very concrete but different forms and procedures, which allows for conceptual ambiguities in its implementation (May & Winter, 2007). Consequently, if stakeholder participation had negative associations for managers, neither the institutional framework nor policy prevented them from operationalising these negative value orientations. This was made possible by the fact that the PMs could choose not to act on input from stakeholders while still complying with the funding agent’s operational requirements.

Essentially, and be it for more or less stakeholder involvement, the PMs impose their value orientations on the participatory process, if they can. This is not to say that most PMs were opposed to the integration of stakeholder participation, but rather that the organisations’ lack of control enabled them to operationalise personal value orientations during project implementation.

This lack of control and support from the host organisations stands out as a finding in a majority of the projects analysed in this paper. This lack created an environment in which the PMs had significant freedom

76

F. Aggestam / Water Policy 16 (2014) 62–78

in designing their project, even when clear steps were prescribed by policies on stakeholder participation. In most cases, the PM utilised this freedom to introduce their own value orientations. Institutional theorists assume organisations adjust to new policy demands – an external adaptation to stakeholder participation – but the street-level practices and the system of project governance remain unchanged (Lipsky, 1980; Schein, 1985). This is demonstrated by the lack of support given to PMs when implementing stakeholder participation and the prevalent reluctance to move away from a technocratic approach to project management. As a result, the policy-driven shift from a technocratic to a more collaborative and participatory management approach often tends to remain symbolic at the institutional level, as long as active stakeholder involvement is not incorporated in the organisation’s customs and/or corporate culture.

5.1. What are the implications for practice?

The most apparent way to improve the implementation of stakeholder participation policy is through more stringent regulations and guidelines, such as tighter control over budgets and implementation practice (control-and-command), not only by the organisation charged with implementation but also by EU and international policy. The problem could also be addressed through increased awareness-raising and education campaigns that confront the managers’ negative value orientations associated with stakeholder participation through, for example, reframing (Dewulf et al., 2004), value attunement (Hall & Davis, 2007), or alternative dispute resolution (Creighton et al., 1998). This would require a more structured and transparent strategy that addresses ambiguities in policy and support for stakeholder participation (Hare & Pahl-Wostl, 2002; Raymond et al., 2010). On the one hand, there are some dangers inherent in these suggestions, namely that strictly defined policies disengage decision-makers from responsibility, both in terms of project management and stakeholder participation. On the other hand, to be successful, stakeholder participation needs to be a two-way communication process that validates stakeholder input (Failing et al., 2004), and which should not be dependent on the PMs’ value orientation (Kaltenborn & Bjerke, 2002; Sadoff & Grey, 2002; Taut, 2008).

Practice could, furthermore, benefit from improved tools for assessing a project’s contextual background, such as improved procedures for ex ante evaluations. If funding agents were to evaluate projects more concretely in terms of the influence and impact stakeholders have on projects, integration may be improved. Tools could also be developed to assess projects prior to conceptualisation, so as to determine if, how and when stakeholders should be engaged. This could also be applied to assess the need for improved organisational control. For example, if there are no informal institutional customs that encourage participation, funding agents could choose to enforce more stringent regulations. This type of approach would allow for more flexibility, help prevent costly participatory processes in cases where they are implemented only to fulfil operational requirements, as well as prevent stakeholder frustration, which tends to increase barriers for project implementation and prevent stakeholders from getting engaged in similar participatory processes in the future.

To conclude, the results from this study point to open questions as regards the effectiveness of international and European policies on stakeholder participation, and highlight the importance of finding a more stringent compromise on what stakeholder participation means. Up until now, the vague definitions that are provided in many policies provide room for variations in policy implementation, both for better and for worse. This has to be taken into account when introducing policies that rely on the discretion of the implementers. There is, however, no single solution to fix this problem. Rather, there is a need for smart and flexible policy approaches to stakeholder participation and for policy

F. Aggestam / Water Policy 16 (2014) 62–78

77

approaches that proactively encourage projects to use stakeholder input, based on context-specific indicators, and which provide for flexible operational requirements at the project level.

References Aggestam, F. (2013). Wetland restoration and the involvement of stakeholders: an analysis based on value-perspectives.

Landscape Research, DOI: 10.1080/01426397.2013.819076.

Appelstrand, M. (2002). Participation and societal values: the challenge for lawmakers and policy practitioners. Forest Policy and Economics 4, 281–290.

Arnstein, S. R. (1969). A ladder of citizen participation. Journal of the American Planning Association 35, 216–224.

Brody, S. D. (2003). Measuring the effects of stakeholder participation on the quality of local plans based on the principles of collaborative ecosystems management. Journal of Planning and Education Research 22, 407–419.

Bruno, L. F. C. & Lay, E. G. E. (2008). Personal values and leadership effectiveness. Journal of Business Research 61, 678–683.

Buijs, A. E. (2009). Public support for river restoration. A mixed-method study into local residents’ support for and framing of river management and ecological restoration in the Dutch floodplains. Journal of Environmental Management 90, 2680–2689.

Castelletti, A. & Soncini-Sessa, R. (2007). Coupling real-time control and socio-economic issues in participatory river basin planning. Environmental Modelling & Software 22, 1114–1128.

CEP (2002). Transboundary Diagnostic Analysis for the Caspian Sea – Executive Summary and Environmental Quality Objectives. Volume 1. CEP, UNDP, Baku, Azerbaijan. http://projects.inweh.unu.edu/inweh/display.php?ID=1053 (accessed 10 August 2013).

CEP (2006). Communication and Public Participation Case Study for the Caspian Sea. CEP, Baku, Azerbaijan.

Creighton, J. L., Dunning, C. M., Priscoli, J. D. & Ayres, D. B. M. A. (1998). Public Involvement and Dispute Resolution: A Reader on the Second Decade of Experience at the Institute for Water Resources. Institute for Water Resources, US Army Corps of Engineers, Alexandria, VA, USA. Dewulf, A., Craps, M. & Dercon, G. (2004). How issues get framed and reframed when different communities meet: A multilevel analysis of a collaborative soil conservation initiative in the Ecuadorian Andes. Journal of Community and Applied Social Psychology 14, 177–192.

Failing, L., Horn, G. & Higgins, P. (2004). Using expert judgement and stakeholder values to evaluate adaptive management options. Ecology and Society 9, 13–30.

Freeman, E. R. (2010). Strategic Management: A Stakeholder Approach. Cambridge University Press, Cambridge, UK. GEF/C.7/6. Public Involvement in GEF Projects. Recommended Draft Council Decision. http://www.thegef.org/gef/sites/ thegef.org/files/documents/C.7.6.pdf (accessed 10 August 2013).

Hall, J. D. & Davis, R. A. (2007). Engaging multiple perspectives: A value-based decision-making model. Decision Support Systems 43, 1588–1604.

Hare, M. & Pahl-Wostl, C. (2002). Stakeholder categorisation in participatory integrated assessment processes. Integrated Assessment 3, 50–62.

Hill, H. C. (2003). Understanding Implementation: Street-level bureaucrats’ resources for reform. Journal of Public Administration Research and Theory 13, 265–282.

ICPDR (2006). ICPDR Response to the Stakeholder Consultations Process 2005. ICPDR, Vienna, Austria. http://www.icpdr. org/main/icpdr/icpdr-stakeholder-conference-june-2005 (accessed 10 August 2013).

ICPDR (2009). ICPDR Strategy for the Public Participation Process 2009: Danube River Basin Management Plan incl. JPM. ICPDR, Vienna, Austria.

Kaltenborn, B. P. & Bjerke, T. (2002). Associations between environmental value orientations and landscape preferences.

Landscape and Urban Planning 59, 1–11.

Krolikowska, K., Kronenberg, J., Maliszewska, K., Sendzimir, J., Magnuszewski, P., Dunajski, A. & Slodka, A. (2007).

Role-playing simulation as a communication tool in community dialogue – Karkonosze Mountains case study. Simulation & Gaming 38, 195–210.

78

F. Aggestam / Water Policy 16 (2014) 62–78

Linnerooth-Bayer, J. & Vári, A. (2006). Floods and fairness in Hungary. In: Clumsy Solutions for a Complex World: Governance, Politics and Plural Perceptions. Verweij, M. & Thompson, M. (eds). Palgrave Macmillan, New York.

Lipsky, M. (1980). Street-Level Bureaucracy – Dilemmas of the individual in Public Services. Russell Sage Foundation, New York.

Matthews, M. M. (2004). Stakeholder Inclusion in Caspian Basin Natural Resource Management. Woodrow Wilson International Center for Scholar, Occasional Paper #288, Washington, DC. May, P. J. & Winter, S. C. (2007). Politicians, managers, and street-level bureaucrats: influences on policy implementation.

Journal of Public Administration Research and Theory 19, 453–476.

Meglino, B. M. & Ravlin, E. C. (1998). Individual values in organisations: concepts, controversies, and research. Journal of Management Information Systems 24, 351–389.

Moxley, J. H., Anders Ericsson, K., Charness, N. & Krampe, R. T. (2012). The role of intuition and deliberative thinking in experts’ superior tactical decision-making. Cognition 124, 72–78.

Raymond, C. M., Fazey, I., Reed, M. S., Stringer, L. C., Robinson, G. M. & Evely, A. C. (2010). Integrating local and scientific knowledge for environmental management. Journal of Environmental Management 91, 1766–1777.

Reed, M. S. (2008). Stakeholder participation for environmental management: a literature review. Biological Conservation 141, 2417–2431.

Sabatier, P. A. (ed.) (2007). Theories of the Policy Process. Westview Press, Boulder, CO, USA. Sadoff, W. C. & Grey, D. (2002). Beyond the river: the benefits of cooperation on international rivers. Water Policy 4, 389–403.

Sandström, A. (2011). Navigating a complex policy system—explaining local divergences in Swedish fish stocking policy.

Marine Policy 35, 419–425.

Schein, E. H. (1985). Organisational Culture and Leadership. Jossey-Bass, San Francisco, CA, USA. Schusler, T., Decker, D. & Pfeffer, M. (2003). Social learning for collaborative natural resource management. Society & Natural Resources 16, 309–326.

Sendzimir, J., Balogh, P. & Vári, A. (2006). Modelling biocomplexity in the Tisza river basin within a participatory adaptive framework. iEMSs Conference 2006, Osnabrück, Germany.

Sendzimir, J., Magnuszewski, P., Bauknecht, D. & Vári, A. (2007). Anticipatory modeling of biocomplexity in the Tisza River Basin: first steps to establish a participatory adaptive framework. Environmental Modelling & Software 22, 599–609.

Stern, C. P., Dietz, T., Abel, T., Guagnano, A. G. & Kalof, L. (1999). A value-belief-norm theory of support for social movements: the case of environmentalism. Human Ecology Review 6, 81–97.

Taut, S. (2008). What have we learned about stakeholder involvement in program evaluation? Studies of Educational Evaluation 34, 224–230.

Teng, S.-K. (2006). Practitioner Guidelines for Preparation of Transboundary Diagnostic Analysis (TDA) and Strategic Action Programme (SAP) in East Asian Seas Region. Southeast Asia Regional Learning Center (SEA-RLC), Bangkok, Thailand.

UNDP (2002). The GEF IW TDA/SAP Process Notes on a Proposed Best Practice Approach. http://waterwiki.net/images/a/ae/ TDA_SAP_bestpractice_for_distribution.doc (accessed 10 August 2013).

UNDP/GEF (2005). Reducing Transboundary Degradation in the Kura-Aras Basin. UNDP Project Document. http://projects. moe.gov.ge/uploads/Project_Document_ENG.pdf (accessed 10 August 2013).

Urwin, K. & Jordan, A. (2008). Does public policy support or undermine climate change adaptation? Exploring policy interplay across different scales of governance. Global Environmental Change 18, 180–191.

Vári, A. (2001). Flood Risk Management in the Upper Tisza Region: Results of Stakeholder Interviews. IIASA, Laxenburg, Austria.

Vári, A., Linnerooth-Bayer, J. & Ferencz, Z. (2003). Stakeholder views on flood risk management in Hungary’s Upper Tisza Basin. Risk Analysis 23, 537–627.

Vugteveen, P., Lenders, H. J. R., Devilee, L. A. J., Leuven, S. E. W. R., Van Der Veeren, J. H. M. R., Wiering, A. M. & Hendriks, A. J. (2010). Stakeholder value orientations in water management. Society & Natural Resources: An International Journal of the Commons 23, 805–821.

Wierzbicki, A. P., Makowski, M. & Wessels, J. (2000). Model-Based Decision Support Methodology with Environmental Applications. Kluwer Academic, Boston, MA, USA. Yanow, D. (1996). How Does a Policy Mean? Georgetown University Press, Washington, DC. Received 3 August 2012; accepted in revised form 24 June 2013. Available online 29 August 2013

Water Policy 16 (2014) 79–101

Decoupling dependence on natural water: reflexivity in the regulation and allocation of water in Israel Michael Gilmont Department of Geography, King’s College London, Strand, WC2R 2LS, UK E-mail: mgilmont@gmail.com

Abstract Since 1999, Israel has secured a decoupling between the total water supplied and water derived from the natural environment through surface and groundwater. Between 2007 and 2010, this decoupling enabled a cut of over 20% in natural water use, while maintaining overall supply volumes despite drought. By 2009–2010, natural water supply had been reduced to the levels of the early 1960s. As seawater desalination contributed less than 10% to the total supply up to 2009, decoupling was achieved primarily through cross-sector efficiency and the recycling of urban effluent. This study uses the case of Israel to exemplify the challenges faced by semi-arid economies in achieving a gap between the upward trend in total national water use and local water withdrawn from natural resources. Decoupling will be shown to comprise two types. One type occurs when an economy ceases to be water self-sufficient. Another type is encountered when an economy has the capacity to remedy its overexploitation of natural water. It will be shown that the first decoupling is non-contentious and non-politicised.

The investments and reforms needed to bring about the second decoupling appear very demanding economically and very politically stressful.

Keywords: Decoupling; Israel water allocation; Natural water; Reflexivity; Water governance

1. Introduction This study uses the case of Israel to exemplify the concept of two types of decoupling of the supply of local natural water abstracted from the environment (herein ‘natural water’) to meet two national imperatives. The first imperative is the continued achievement of national food security in the face of rapidly increasing water demands for food production. The second is the restoration of the environmental sustainability of natural water after it has been over-exploited. Decoupling results when a gap opens up between the upward trend in water demand and the rate at which regional natural water can be further developed for productive use. One type of decoupling begins when an economy can no longer support self-sufficient food production. Another type is encountered when an economy begins to address its doi: 10.2166/wp.2013.171 © IWA Publishing 2014

80

M. Gilmont / Water Policy 16 (2014) 79–101

over-abstraction of natural water and develops alternative supplies that are not derived from the freshwater environment. These alternative supplies can include mobilisation of recycled water and/or desalination, as a means to both enhance supply and simultaneously reduce environmental abstraction.

The two types of decoupling will be called Economic Decoupling (Type One), and Natural Water Decoupling (Type Two). The first is long-practised and understood, and involves exporting – via international trade – the water footprint and other externalities of the food commodities required. This remedy is achieved by importing virtual water (Allan, 2001). Adopting Type One decoupling requires the abandonment of food self-sufficiency policies which attempt to produce nearly all food domestically.

Type One decoupling is a politically easy, relatively low-risk, and politically silent solution, common in economic development and diversification of an economy. Type Two decoupling, involving relatively small volumes of local natural water as opposed to virtual water, is associated with very challenging politics. Its installation requires very robust and effective institutional and technical capacities to put in place domestic and agricultural water tariffs and other regulatory measures to reduce water allocated to all sectors.

The study explores the concept of Type Two decoupling, and its relationship to Type One decoupling in the case of Israel, with potential application to other semi-arid economies. Section 2 presents the water resources background to decoupling and the terminology used. Section 3 develops a new conceptual model identifying the two types of decoupling in managing water. Section 4 introduces the case study of Israel and the data used to illustrate the decoupling process there. Section 5 analyses Israel’s experience of decoupling, and the political challenges it faced. Section 6 discusses the results of the analysis and explores the scope for further development of the model. Despite the political difficulties encountered in Israel, it is argued that Type Two decoupling has been effective. Politically sustainable changes have been implemented that have initiated a reflexive trajectory in the use of Israel’s natural water resources.

2. Background to decoupling and definitions The broad ideas and trends identified and developed in this study align with the 2011 United Nations Environment Programme (UNEP) report Decoupling Natural Resource Use and Environmental Impact from Economic Growth (UNEP, 2011). The term decoupling has been in use since at least the early 1990s to describe the separation of resource use and economic growth (OECD, 2001). For water, Smith et al. (2011) conceptualised and identified a number of forms of decoupling. They did not, however, draw a distinction between the decoupling of the rate of water use from the rate of economic growth, and the second decoupling of the rate of natural water consumption from the total use of natural water available to an economy. Rather they assumed reductions in the natural water component of total water to be a component of economic decoupling. Such a perspective fails to acknowledge the unique conditions in cases where natural water decreases through being substituted by other forms of water resources. The conceptualisation presented below highlights the clear distinctions between the various forms of economic and trade-based decoupling (Type One), and a second decoupling (Type Two) which addresses the competition for natural water between economic uses and ecosystem services of water in the environment. While Type One requires the import of virtual water, Type Two decoupling involves the reallocation of internal natural water use.

Type One decoupling has been long identified and discussed and has an established place in the literature (Allan, 2000, 2001, 2003). Type Two decoupling is, in contrast, an emerging phenomenon.

M. Gilmont / Water Policy 16 (2014) 79–101

81

A central feature of this second decoupling process is that the emergence of a disconnect between the growth in the total water requirements of an economy and the growth in extraction of water from the environment. Type Two decoupling is associated with an increased proportion of the total water supply being provided from new sources. It can be defined as ‘an increase of total water supply relative to water supplied from the environment’. In the case of Israel, the new water comes from recycled effluent and seawater desalination.

The outcome of Type Two decoupling is an example of an environmentally reflexive water policy. A reflexive water policy is one that responds to the changing priorities of society and its political processes, and reallocates water accordingly. In the case of water, policy is responding to the belief that the environmental services of water should be recognised because of their contribution to supply reliability and other intrinsic values (Allan, 2003). Turton (1999) conceptualised ‘a reflexive water demand curve … [as] a curve that moves from a condition of “water deficit to the sustainable level of supply”’ (Turton, 1999: 14). This behaviour is in contrast to previous unsustainable diversions of natural water whereby demand outstrips supply (Turton & Ohlsson, 1999). While the actual reallocation of water to the environment is one component of reflexivity, establishing sustainable demand on the resource is arguably a prerequisite to any attempt to rehabilitate natural water resources. Reflexivity in this study should therefore be understood as a move towards stabilising the natural resource base at reliable levels of exploitation below a historic unsustainable peak abstraction. It will be shown that a range of institutional reforms plus the effective adoption of new technologies associated with the Type Two decoupling have impressively mitigated the political risks associated with reducing the abstraction of natural water in Israel.

The analysis in this study is concerned with resource use and internal politics on a national level. The purpose of the study is to highlight some generic processes that impact some semi-arid political economies that have already experienced or are still experiencing rapid population increases and economic diversification and related increases in the demand for water. The study focuses on total volumes of water supplied and used across sectors. It is this national total which is constrained by local water availability and these constraints can be addressed through the two types of decoupling.

This study disaggregates the components of water supply as follows:

• Natural water – fresh water derived from environmental sources of surface and groundwater, most of which is delivered at potable standards, but including some brackish water.

• Effluents – predominantly domestic sewage treated to a standard only used to irrigate certain crops excluding those food crops that come into direct contact with irrigation water.

• Desalinated water – seawater treated to a potable level through reverse osmosis technology.

The changes in relative contribution of different types of water give rise to Type Two decoupling.

3. A conceptual model for water decoupling The 2011 UNEP report (UNEP, 2011) and the subsequent water-specific 2012 UNEP report (UNEP, 2012) identify two broad categories of decoupling – resource decoupling and impact decoupling. Resource decoupling is concerned with the decoupling of economic growth and resource use. Impact decoupling is related to resource use and the impact of such use on water ecosystems. The UNEP model is a general

82

M. Gilmont / Water Policy 16 (2014) 79–101

decoupling model designed to be applicable across natural resources. It fails to incorporate the unique trends of substitution of natural water with other types of water, and the transfer of environmental externalities (including energy) inherent in substitution of natural for alternative water supplies.

This study proposes a water-focused conceptual model of decoupling, based on natural water as the baseline environmental resource. From the evidence presented from Israel, decoupling is understood to be a sequenced incremental process, first economic, and secondly environmental. It has political, technical and economic dimensions. The case study demonstrates that environmental decoupling does not immediately lead to an improvement of the environmental conditions, but can be regarded as a prerequisite for subsequent environmental improvement. Therefore there is a lag between decoupling of the resource and environmental impact. This study provides a basis for a water-specific understanding of the decoupling processes and its general relevance to other semi-arid economies.

A proposed conceptual model in this study is a development of a five paradigm descriptive model theorised by Allan (2001). This basic model, illustrated in Figure 1, posits an era of water resource development (the Hydraulic Mission) characterised by rising water demand, being followed sequentially by three different approaches reflecting a succession of reflexive policies. It is important to note that the Hydraulic Mission development era of water ran in tandem with significant levels of socio-economic development. These economic development processes create the economic and technological capacity to facilitate decoupling.

The conceptual model of decoupling is shown in Figure 2. The end of the Hydraulic Mission in the 1980s was preceded by Type One economic decoupling. This decoupling enabled growth in water needs to be met without expanding national water use at the same rate of demand. Type One decoupling enables a region to overcome the limits of its water resources by exporting the consequences of further water-intensive production, including the negative environmental externalities of water use. The most

Fig. 1. Transition through five water paradigms (adapted from Allan (2003), Figure 1 and Allan (2001), Figure 8.5).

M. Gilmont / Water Policy 16 (2014) 79–101

83

Fig. 2. Conceptual model of decoupling of natural water.

important element of Type One decoupling is that food self-sufficiency policies have to be abandoned, even if they remain as aspirations in national mindsets and the media. Type One decoupling enables populations to grow beyond the level for which there is a secure internal supply of water. For economies which are not per capita water scarce, Type One decoupling can be defined as being an economic decoupling, where economic growth and national development cease to be related to levels of primary production. Gleick (2002) has identified the phenomenon in the United States, a trend which is broadly comparable to the UNEP’s Resource Decoupling (UNEP, 2011). Versions of Type One decoupling have occurred across industrialised neo-liberal economies during the 20th century, and are well-documented in the literature (Allan, 2001; EUEA, 2010; UNEP, 2011).

With Type One decoupling in place, the development of water resources does not need to keep pace with population increases and economic growth. Intrinsic economic, technical and political momentum in an advanced economy brings about continuing growth in water use for a period. It is arguably only through this continued growth in the use of water resources that environmental limits have been exposed (Gleick & Palaniappan, 2010). In the case of the most advanced economies in semi-arid regions, the eventual levelling off of natural water use normally occurs at an environmentally unsustainable level, as exemplified by Israel, the Murray–Darling Basin in Australia, and Southern California (Allan, 2001). Elsewhere in the world, including non-arid economies, changes in abstraction patterns may be induced without breeching nationally sustainable levels, as a consequence of the availability of new technologies such as recycling and marginal water use, and the progressive cost-effectiveness of water and other resource allocation and management options compared to natural water abstraction (Gleick et al., 2009). Even where water scarcity is not a national challenge, particular regions or catchments may experience scarcity due to particular activities and require action to redress over-abstraction. This is evident in the United Kingdom, where areas in East Anglia are identified as being water scarce due to high concentrations of irrigation (DEFRA 2011).

84

M. Gilmont / Water Policy 16 (2014) 79–101

4. Israel as a case study for decoupling The high quality datasets on water supply and consumption that exists for Israel enables both types of decoupling to be exemplified. The annual reports of the Israeli Central Bureau of Statistics (CBS) provide data up to 2010 on the water supplied by source (listed as ‘production’), and consumed by sector (listed as ‘consumption’). This study uses the term ‘supply’ which incorporates both abstraction from ground and surface water, and production of recycled and desalinated water. While provisional data exist for 2011 and early 2012, these figures are not necessarily consistent with the annual CBS data which is released approximately 18–20 months after a year’s end. As such, this analysis does not go beyond the 2010 official data limit.

Over the past 60 years, Israel has demonstrated both types of decoupling. Figure 3 shows data for Israel in the mode of the conceptual model proposed Figure 2. Israel executed a Type One decoupling from the early 1970s, with water requirements growing to exceed the water available within the national borders. The emergent gap between the estimated population water demand and national water supply was addressed by resorting to virtual water imports.

Fig. 3. Schematic diagram of Type One and Type Two decoupling in water demand from local natural water supply in Israel.

Note: The ‘Population Water Demand (extrapolated from 1961–65 water use)’ is based on the per capita water use of Israel from 1961 to 1965, extrapolated based on annual World Bank population figures (World Bank, 2011). It assumes linear growth with population, and does not include a factor for agricultural efficiency. The correlation between population growth rate and overall water growth up to 1972 observable in this figure corresponds with the period up to the early 1970s under attempted food self-sufficiency. This was also the period when the highest per capita internal natural water consumption was recorded at about 600 m3 cap 1 yr 1. In the absence of robust virtual water data, the extrapolation of ‘population water demand’, holding per capita total water (including water embedded in food) constant, is a proxy for the likely natural water requirement had Type One decoupling not been embraced.

M. Gilmont / Water Policy 16 (2014) 79–101

85

Figure 3 shows that Type Two decoupling emerged in the second half of the 1980s. The good intent of the initiation of Type Two decoupling faltered in the early 1990s. Abstraction increased throughout the 1990s. The trend of increased Type Two decoupling resumed in the first decade of the 21st century.

For the years 2009 and 2010, the natural water supply was reduced to between 1,300 and 1,350 million cubic meters (MCM). This range is the same as the early years of the 1960s. Type Two decoupling made it possible for the total water supply to average around 2,000 MCM for much of the period from 2000, even though natural water was reduced.

Figure 4 presents a more detailed view of supply trends from 1958. It illustrates the impact of recycled and desalinated water in the two decades up to 2010. It also includes rainfall data to illustrate the relationship between water availability and policy change. The impact of the extreme 1992 rainfall event on allocative policy is evident.

The rainfall dataset included is for a single point-gauge in Nablus. The data are presented to provide an indication of annual recharge. Israel is reliant on three main sources of water: the coastal and mountain aquifers, and Lake Kinneret (Lake Tiberius/Sea of Galilee). Despite the geographical distribution of these resources, there is a strong correlation between the annual rainfall totals at the Nablus gauge and the total rainfall-derived recharge across all storages (annual recharge data from Weinberger et al.

(2012)). The R 2 for a linear regression of total recharge against Nablus rainfall is 0.87, making the Nablus point data an effective proxy for total recharge. The rain-gauge data are used as they provide a longer time-series than is available from the recharge series, which is only complete from 1975.

Fig. 4. Israel water resource development 1958–2010. Rainfall trends for the region indicated by the long time series at Nablus (presented 1958–2008). Data sources: Water supply – State of Israel (2002b, 2004, 2005, 2008, 2011, 2012a); rainfall 1958– 2008 – Kislev (2010), rainfall 2009 – Palestinian Authority (2010).

86

M. Gilmont / Water Policy 16 (2014) 79–101

5. The politics of Israel’s water development and decoupling: from water mobilisation to reallocation The following section analyses how Israel has achieved its decouplings. Achieving both types of decoupling requires much more than the implementation of economic and technical instruments. Enacting these instruments is a complex political process involving negotiation between multiple interests, necessitating changes in behavioural priorities and ideologies. In addition to the numerical and published data used in this study, the analysis also draws upon interviews in 2006 and 2010 with a number of individuals with first-hand experience of achieving decoupling. This first-hand evidence highlights the importance of political factors in shaping the necessity for, the pace and makeup of change, in particular for Type Two decoupling. Without the negotiation process, it is unlikely that a decoupling policy would be politically sustainable, that is there would be insufficient support, or too much opposition from particular quarters, to successfully implement an ambitious policy. This is seen in the delays in the adoption of desalination until 2001, despite the technology being economically and technically feasible from the mid-1990s. The pace of adoption was slowed by the Ministry of Finance, preventing the faster implementation that was being advocated by some elements in the Water Commissioner’s office.

In achieving Type Two decoupling, Israel has demonstrated its ability to broaden the range of feasible options within the policy space; it has simultaneously embraced the reduced diversion of water from the environment and the increased supply of water from non-environmental sources. Such behaviour is in contrast to the policies based on the supply-dominated mindset which characterised the earlier water history of Israel.

5.1. The Hydraulic Mission –1948–ca.1970 – a self-sufficiency (and security) goal The Hydraulic Mission stage of Israel’s water development following its establishment in 1948 is well covered in the literature, including Feitelson & Fischhendler (2009), Nativ, (2004) and Tal (2002). The main aim of this stage was to change the geographical distribution of water to match available land and population centres. To meet these goals, a pre-state agricultural water development and supply cooperative, Mekorot, was absorbed into state apparatus, and a new planning organisation, TAHAL, was created. The first major water mobilisation project was the 100 MCM/year Yarkon–Negev Pipeline, which itself later formed part of the National Water Carrier. The carrier was completed in 1964. The impact of its 400 MCM annual capacity (Mekorot, 2009) is shown in the growth in abstraction from 1964 in Figure 4. The carrier conveyed water from the upper River Jordan to the centre and south of the country, using Lake Kinneret as a storage facility. An original plan would have seen water pumped directly from the Jordan headwaters, rather than pumping up from Lake Kinneret, 211 m below sea level. However this plan was abandoned when the United Nations prevented Israel from diverting the Jordan in 1953 (Kantor, 2001).

During the 1950s, the legal framework was based on inherited Mandate and Ottoman water law, with modifications concerning wells, water measurement and drainage (Laster & Livney, 2009). The 1959 Water Law combined all previous water legislation, reaffirmed water as property of the state, and created mechanisms for public accountability. The law also established the position of the Water Commissioner responsible for determining annual availability and allocation of water. The Water Commission was initially part of the Ministry of Agriculture and this arrangement inevitably caused recurring tensions

M. Gilmont / Water Policy 16 (2014) 79–101

87

with the Minister (Feitelson & Fischhendler, 2007). This tension is regarded as having hindered attempts to introduce cuts in water during drought years, as analysed by Feitelson et al. (2007). Despite the political challenges of the structure established in 1959, the legal framework led to the comprehensive monitoring of water. Such monitoring has enabled the control over water resources that has enabled effective decoupling, as well as the recording of data to analyse the trends.

5.2. Type One decoupling and Hydraulic Mission (growth goal) Water demands generated by economic development and population growth led to a shift in emphasis towards getting higher returns from water. The ambition for total food self-sufficiency was abandoned in the early 1970s (Katz, 1994). Certain foods continued to be produced domestically, but food imports were recognised as being crucial to meet both the needs of the growing population and of the developing economy. A policy of importing many staple foods allowed water to be directed towards higher value crops as well as to the very much higher value uses outside the agricultural sector. The shift of water away from agriculture is shown in Figure 5.

The evidence for the change in the sourcing of agricultural produce is presented in Figure 6. The demand for food by the growing population appears to have been mainly met by domestic production up to the early 1970s. In 1967, domestic production accounted for 70% of major agricultural products.

Fig. 5. Absolute use of all types of water by sector. Note the decrease in agricultural water use would be more pronounced if effluent were excluded from the total.

88

M. Gilmont / Water Policy 16 (2014) 79–101

Fig. 6. Food balance of Israel (‘Twenty most important food and agricultural commodities’ for domestic and imported commodities). Data sources: Agricultural supply – FAO (2012a, b); population – World Bank (2011).

From 1972, the domestic contribution declined steadily, comprising just below 50% by 2000. The commitment to importing grain in particular can be regarded as one of the most significant water decisions taken by Israel (Arlosoroff, 2006, personal communication), and resulted in Israel’s food security no longer being dependent on internal water resources (Shuval, 2006, personal communication). Wheat and maize comprised the largest agricultural imports by tonnage in 2010 (FAO, 2012a). Dairy products are the main domestic agricultural products, but they are also reliant on the imported grains, and therefore on virtual water, for animal feed. The shift to imported water in food is therefore greater than indicated in agricultural data that is focused on production quantities alone.

The policies of the early 1970s can be regarded as the cornerstone of Type One decoupling. Further structural adjustment included pioneering initiatives to increase output per unit of water. Between the 1970s and early 1990s, output per unit water doubled (Arlosoroff, 1997a). The trend in total water supplied in Figure 4 is generally stable throughout the 1970s. Water supplied to agriculture was stable throughout the 1970s, and then declined after a brief rise during the early 1980s. Despite decreases in agricultural water supply, agricultural output of major crops has gradually increased over time.

Figure 7 illustrates these increases in productivity, along with increases relative to water inputs.

Beginning in the early 1970s, further measures were undertaken to ensure existing water supplies were developed to their full potential. In 1964, the abstraction of groundwater was 861 MCM, and surface water 531 MCM. In 1985, groundwater abstraction was 1,213 MCM and surface water 865 MCM, an increase over 21 years of 42 and 63%, respectively. One of the impacts of this continued development of ground and surface water abstraction was significant seawater intrusion into the coastal aquifer,

M. Gilmont / Water Policy 16 (2014) 79–101

89

Fig. 7. Relative agriculture output (excluding livestock) and water input 1968–2010. Water data as for Figure 4; agricultural production data series derived from State of Israel (2012c).

and the lowering of the regulatory minimum level of Lake Kinneret (State of Israel, 2002a). These trends of unsustainable resource abstraction resulted in both risks to water quality and the decline in the environmental health of Israel’s surface-water environment.

5.3. Emergence of environmentally reflexivity From the 1970s to the 1990s, the deteriorating environmental conditions and greater impact of environmental voices marks the emergence of environmental concern onto the social and political agenda (Tal, 2002). However, concern over the declining health of particular environmental resources (for example Lake Kinneret) did not have an impact on the overall state water policy of supply and allocation. It is therefore difficult to argue that an environmentally inspired water policy paradigm, as suggested in Figure 1, existed in Israel. Instead, a bifurcation in the mainstream water and environment policy occurred. This approach embraced certain specific environmental issues or mitigated impacts of abstraction, rather than changing the Hydraulic Mission approach that had determined the way water was allocated and managed.

From 1969, early signs of the declining environmental health of Lake Kinneret were evident as a consequence of upstream interventions. These interventions included water abstraction, the draining of upstream swampland and the diversion of saline springs. The impacts were evident in lake mineral content (see for example Nishri et al. (1999)) and changes in flow behaviour in the lake. The Lake Kinneret

90

M. Gilmont / Water Policy 16 (2014) 79–101

Administration was formed to manage the condition of the lake. The Kinneret Administration was a nonstatutory government-affiliated entity acting as a forum for coordination of the protection of the lake by statutory bodies (Laster & Livney, 2009). The creation of the Ministry of Environment in 1988 further enhanced environmental stewardship. The ministry assumed responsibility for the environmental elements of the 1959 Water Law. It appears that 1988 marks the entry of the environment into mainstream policy space and the organisational structure, rather than the policy bifurcation that had previously characterised early environmental action on water.

The early 1980s appears to mark an important moment in the assessment and allocation of available resources. A linear regression of a lag-1, 2-year moving average (MA) of Nablus’s rainfall against annual water allocation reveals no correlation between rainfall inputs and change in allocations from 1959 to 1981. This trend signifies that during the period, water inflows did not appear to significantly influence allocation policy. From 1982 to 2010, however, there is a moderately positively correlated trend with an R-squared of 0.6. This metric excludes outlier years of 1988, 1991 and 1995; 1995 is heavily influenced by 2 years of declining rainfall relative to the atypical peak in 1992, while 1988 and 1991 experienced severe cuts as a consequence of the failure of brinkmanship behaviour in preceding years. Removing these three outliers serves to unmask the otherwise noticeable shift in behaviour from the pre-1981 period (see Figure 8). No such easily identifiable outliers exist in the pre-1981 period. The moderately correlated statistics show that, either by choice or by necessity due to reaching environmental limits, changes in environmental recharge starts to have a bearing on annual allocations of water for consumptive uses.

5.4. The foundations of Type Two decoupling – technology plus further allocative and productive efficiencies Along with the continued pursuit of productive and allocative water efficiency, the 1980s witnessed enhancement of the productive use of supplied water. These gains were achieved by developing largescale effluent use. The pace of this transition was increased by effluent water being priced at 50% of the average cost of natural water (OECD, 2010). Due to the generally stable nature of domestic effluent supply even in drought years, effluent is not vulnerable to the allocation cuts of natural water, except for the slight decrease in 2009 discussed later. These economic and reliability factors both incentivised use of effluent and reduced opposition to the substitution of natural water.

In 1987, the Shafdan water recycling plant opened south of Tel Aviv, with the first effluent being supplied to Mekorot in 1989. The impact of this new water on the national water supply is shown by the emergence of a separation between Total Water and Natural Water in Figure 4. Effluent use significantly increases the productive return from natural water, achieved by using the water twice at qualities appropriate to the use of the water.

The first large-scale mobilisation of effluent marks the beginning of a new approach in Israel’s national water policy, identified here as Type Two decoupling. In subsequent years, sewage reclamation plants have been built across the country. By 2008 these plants were facilitating a peak reuse of just over 50% of all water supplied to the domestic sector1.

1

The commonly quoted level of recycled effluents in Israel is 70% (see for example Inbar (2007)). The discrepancy with the 50% quoted here for 2007 is because not all domestic supply is returned as sewage, for example evaporation from garden irrigation.

M. Gilmont / Water Policy 16 (2014) 79–101

91

Fig. 8. Linear regression of interannual variation in rainfall (Nablus gauge) and interannual changes in water allocation 1959– 1981 and 1982–2010, demonstrating the marked changes in allocation behaviour to increased sensitivity to inflows from 1982 onwards.

92

M. Gilmont / Water Policy 16 (2014) 79–101

5.5. Political dynamics of reform and the entry of economic rationalism As natural water became increasingly developed, sensitivity to resource availability featured more prominently in allocation policies, as indicated by the rainfall regression trends discussed above, and progressive ideas gradually permeated water policy space. During the 1985 drought, significant cuts to allocation were made, mainly to agriculture. When the rains returned, the cuts were quickly reversed.

This reversal in allocation cuts was characteristic of Israeli policy of the period. The relaxation echoed the response at the end of the previous drought in 1979–1980, when allocations were increased (Allan, 1995). During the 1980s, responses to rainfall variation involved a form of brinkmanship, designed to avoid politically difficult and unpopular cuts. This brinkmanship approach involved allocating water for the coming year based on likely assessments of recharge the following winter, as opposed to the available resources at the start of the water year.

The brinkmanship behaviour has been described as a trade-off between a certain loss of water to agriculture, with an uncertain loss to storage volumes and environmental condition; policymakers tended to favour uncertain losses over certain losses (Feitelson, 2005; Feitelson et al., 2007). A further perceived advantage of brinkmanship was that through drawing down storages, the policy ensured that maximum storage capacity existed for the anticipated future rains. The brinkmanship policy failed in 1989, when the anticipated rains did not materialise. This was a year in which one of the highest annual supply volumes was recorded, resulting in the continued lowering of storage levels (Feitelson, 2006b).

Inspired by the shock of the drought-driven allocation cuts of the 1980s, the Israeli bureaucracy took the first steps in expanding the range of water demand management tools beyond agricultural efficiency technologies. A 1986 report by the Finance Ministry highlighted the lack of full cost recovery in the water sector and the significant control of the agricultural lobby over the decision-making process (Feitelson, 2006a). While the report did not have an immediate policy impact, it heralded an era of privileging economic principles in water management in which various types of demand management exerted considerable influence on the water sector during the 1990s.

The weaknesses of the brinkmanship approach were further exposed in 1991, when very significant cutbacks of 25% (489 MCM) were made, including a 21% cut in water use by agriculture. There was significant political opposition from the Ministry of Agriculture and the agricultural community to this policy. The dismissal of the then Water Commissioner has been strongly linked to the introduction of these cutbacks (Feitelson et al., 2007; Tal, 2002). Following an atypical rainfall spike in 1992 and against a background of opposition to the 1991 cuts, reductions in water allocations were reversed.

Despite the 1992 rainfall peak being a one-off, allocations to agriculture continued to increase up to 1999, demonstrating the political difficulties of water reallocation, and therefore permanent reductions in supply. While the reductions of 1990–91 were not sustained, the policy was the first instance of agricultural water allocations being reduced to historically low levels in reaction to drought, as seen in Figure 5. The policy also questioned the sustainability of past brinkmanship practices in the context of drought response. The action demonstrated that cutting agricultural allocations was possible, though very politically challenging. From an agricultural perspective, the 1991 cuts, in combination with those of 1985, ended the certainties of water provision. The experience of the 1985–91 period can, therefore, be regarded as widening the political envelope of feasible drought response.

The 1990s have been characterised as marking the end of the incrementalism and reactivity of the 1970s and 1980s (Menahem, 1998), with a shift towards strategic planning and analysis. The powerful voice of economic principles began to dominate the policy space from the early 1990s, along with an

M. Gilmont / Water Policy 16 (2014) 79–101

93

alternative mode of thought in favour of technocratic solutions through desalination (Feitelson et al., 2007). Because of the cost of desalination, and of other management options available, namely demand management, technocrats and economists were often at loggerheads. It can be argued that the peak of economically based arguments was evident in the 1997 report of ‘The Public Commission on Water Sector Reform’ (The Arlosoroff Commission). The report advocated sustained demand management. Additionally, the Arlosoroff Commission argued for revisions to the structures and laws established in 1959. This was the first time a large-scale institutional revision had been formally put on the political agenda. As part of a demand management strategy, the Arlosoroff Commission recommended the continued implementation of a ‘block rate’ water pricing structure for domestic and agricultural use, with rates controlled by the Water Commission. This recommendation was disputed within the commission members (Arlosoroff, 2006, personal communication). The report argued for use of demand elasticity as a means of shifting water to higher value uses. The proposed structural changes would influence the respective roles of the public and private sectors, as a means to reduce the influence of political interests.

Importantly, the Arlosoroff Commission recommended that desalination be delayed for at least 10 years with forecasts showing that demand could be met through further efficiency improvements. Efficiency savings were estimated to cost $0.15–0.35/m3 in the urban sector and $0.04–0.05/m3 in the agricultural sector. This contrasted with the then costs of desalination of between $0.8 and $1/m3 (Arlosoroff, 1997b). This economics based argument prevented desalination from being adopted for most of the 1990s, despite the forceful arguments made in favour of desalination. The continued deployment of economic instruments, assisted by the Arlosoroff Commission’s recommendations, are regarded as having been very important in re-shaping Israel’s water demand patterns (Shevah & Kohen, 1997).

5.6. Commitment to Type Two decoupling and institutional reform Political factors are shown to be very influential in highlighting the necessity for the adoption of Type Two decoupling and its nature. Political influences have also shaped the need for and execution of institutional reform to further facilitate decoupling. During the early 1990s, further decoupling as a means of reducing pressure on natural resources was by no means inevitable. Evidence exists that consideration was given by officials in the Ministry of Finance in the early 1990s to the abandonment of agriculture, or at least to reducing it to those producers that could pay the full cost of water. A former Israeli Parliamentary researcher recounts committee evidence given by Ya’akov Tzur, the Minister of Agriculture 1992–96, describing a discussion with the then director general of the Ministry of Finance concerning agricultural subsidies: ‘The Director General at that time of the ministry of finance said … “We need the policy of Singapore.”

And then Ya’akov Tzur [Minister of Agriculture] … . said to this Director General “yes but Singapore doesn’t have any agriculture” and he said “OK that’s what I’m saying, if it’s not an economic deal to have agriculture,” … that was the position [of the Ministry of Finance] in the early ’90s’ (Israeli Parliamentary Researcher (retired), 2010, personal communication).

Had a policy of abandoning agriculture, or significantly reducing it been pursued in the 1990s, up to 1,000 MCM of water would have been made available for other uses, including allocations to the environment and to increased urban supplies, and as a possible contribution to the advancement of

94

M. Gilmont / Water Policy 16 (2014) 79–101

peace agreements. Such a reduction in agricultural activity would have been socially and politically highly challenging. However the discourse demonstrates that the perpetuation of a large agricultural sector, and with it a requirement for Type Two decoupling, was not inevitable. Rather it is an outcome mediated by politics.

Despite the increasingly progressive water policies, annual operations and their impact on natural water resources continued to be shaped by evolving politics. Between 1992 and 1998 there was a significant increase in annual supply, leading to one of the highest recorded levels of natural water supply in 1998 of 1978 MCM. The unpopularity of cuts meant that the next drought, from 1998 to 2001 was not immediately met with the cuts necessary to protect the natural resource. The 1998–2001 drought response was in contrast to the cuts adopted in 1991 (Feitelson et al., 2007). One of the outcomes of this policy was the historically low level in Lake Kinneret in December 2001 (State of Israel, 2009).

Such behaviour exhibits a compromise between protecting environmental resources, and keeping cuts to agriculture within a politically and socially acceptable range. Unlike the 1991 drought, the 1998 drought led to marked changes in the political dynamics and management options that were pursued in the first decade of the 21st century.

A further reappraisal of policy direction occurred from 1999, driven by the drought and continued environmental degradation, and a reduction in the estimated cost of desalination. Desalination reappeared on the agenda from 1999, with the Water Commissioner becoming an important advocate (State of Israel, 2002a). Previous attempts to explore desalination, and thus an early enhancement of Type Two decoupling, appear to have been frustrated by the Ministry of Finance. The then Director of Planning in the office of the Water Commission, attempted to initiate a pilot programme of desalination in the mid1990s. He found this attempt frustrated by the Ministry of Finance. He also noted a rejection of desalination by the senior water establishment at that time. He argues that it was only after the 1999 drought that there was enough political traction to overcome the resistance of the Finance Ministry (Dreizin, 2010, personal communication). The political interplay between the Ministry of Finance, eventually influenced by the political and resource ramifications of drought, appear to have determined the timing of desalination adoption. Bids tendered in 2001 for Israel’s first commercial-scale desalination plant showed the cost of desalinated water to be $0.5–0.6/m3, with the final price achieved through careful drawing up of the 25year Build–Operate–Transfer contracts (Kronenberg, 2002). The 2001 tender quotes for desalinated supply reduced, but did not invalidate, the power of the economic argument against desalination. The marginal cost of desalination was comparable to other options, including urban conservation. Increased recycling capacity from 2000 alleviated some of the impact of drought, and restored a degree of supply reliability for agriculture in advance of desalination supply.

A strategic ‘Transitional Master Plan’ was produced by the Ministry of National Infrastructure in 2002, for the period up to 2010. It outlined desalination as a key contributor to the future mix of water sources. Desalination was justified on the grounds of meeting the key principles of the plan, including the assertion that ‘water sources will be rehabilitated and their quality maintained’ requiring ‘a switch from a policy of brinkmanship to one in which future needs will be reliability fulfilled’ (State of Israel, 2002a: 4).

The trajectory of the Master Plan appeared to combine reflexive elements of reduced environmental pressure, with a continuation of the supply augmentation strategy of the Hydraulic Mission, albeit not involving natural water. There was a target of 500 MCM of desalination by 2010. Importantly from the perspective of Type Two decoupling, the role of desalination is seen as combining supply augmentation

M. Gilmont / Water Policy 16 (2014) 79–101

95

and the restoration of environmental storages through recharging aquifers, both through rainfall and with desalinated water when the capacity exists (Dreizin, 2006).

The health and reliability of the surface and groundwater sources was therefore embedded at the core of the 2002 plan. The simultaneous pursuit of rehabilitation and augmentation/reliability appear as key elements of Type Two decoupling and address both sides of a widening policy space that spanned both augmentation needs and an awareness of environmental risk. The political responses to the droughts of 1991 and 1998 recognised the unsustainable nature of both cuts to agriculture and the impact of overabstraction on environmental health, and the cost of trading off agriculture against the environment.

Further reflexive change occurred in 2004, with the amendment of the 1959 Water Law to recognise the environment as a user of water, thereby enabling the Water Commissioner to allocate water for environmental purposes. As of 2011 however, the official data do not record any environmental allocations having been made. The absence of an official allocation to the environment does not mean environmental needs have been ignored. From 1994, with the establishment of the River Restoration Administration, efforts were made to rehabilitate the heavily degraded groundwater-fed coastal rivers. A policy was adopted to introduce a high-level treated effluent to restore flow (Bar-Or, 2000). This effluent could then be recaptured and sent to agricultural users. Projects to implement such schemes have suffered delays, but are currently progressing (Rinat, 2012), with a project for the Yarkon River commencing construction in summer 2013 (Mekorot, 2012). Such use, however, means that a proportion of recycled water will be deployed three times, first as natural water supply, secondly as environmental water, and thirdly as effluent to agriculture.

Despite the present absence of a recognised environmental allocation, successful efforts made up to 2010 to reduce the use of natural water can be regarded as a de facto environmental allocation since reduced abstraction will allow groundwater recharge to occur. Such stabilisation is regarded as a crucial first step in Israel’s 2002 transitional Master Plan (State of Israel, 2002a). There were near historic low natural water inflows between 2007 and 2009; only 2001 marked a lower 3-year average than 2009, while 2008 ranked fourth, behind 1986 (data from Weinberger et al. (2012)). Despite these trends, levels in aquifers and the Sea of Galilee did not breach historic low levels (based on data from State of Israel (2009, 2012b)). This fact demonstrates that theoretically there was additional natural water that could have been mobilised if the political will existed. Through Type Two decoupling, the need to push the environment to absolute limits was avoided.

A further amendment to the Water Law was achieved in 2006. The reform was designed to change the centre of power from a number of government ministers to a new Water Authority, especially with respect to the politically sensitive setting of water tariffs. The Water Authority was created in 2007, with the Water Commissioner replaced by the Director of the Water Authority. The Director additionally chairs a Water Council formed of representatives of water stakeholders. The reform was designed to remove the political influences over water allocation and the concentration of power inherent in the old system. The change reflects ongoing attention from the 1990s to the need to depoliticise water allocation. These changes are the result of ‘a search for structures more appropriate for the attitudes, visions and values of day … [replacing those based on] ideas of half a century ago’ (Feitelson, 2006a: 203).

The power shift inherent in the 2006 reform made passing the legislation politically difficult as normal avenues for introducing such legislation were considered ineffective. It is likely there would have been opposition from those arms of government that would have to surrender control to the Water Authority.

As such, the amendment to the Water Law was passed under the ‘Arrangements Law’ whereby it was included as an appendage to the 2006 Finance Bill, and passed as part of that package. To have opposed the water element would have risked bringing down the entire Finance Bill with wider ramifications for the

96

M. Gilmont / Water Policy 16 (2014) 79–101

coalition government of the time. By using the Arrangements Law, a potentially politically volatile reform was manoeuvred through the parliamentary system.

In many respects, the emerging institutional and resource landscape follows many of the recommendations laid out by the 1995 Arlosoroff Commission. The main difference to the 1995 vision is the early commitment to desalination before alternative measures had been fully implemented. The first desalination plant of 100 MCM capacity was operational in 2006 with a further 35 MCM capacity in 2007 and another 100 MCM plant active by early 2010 (State of Israel, 2010b). Even after the commitment to desalination by water managers, the Ministry of Finance continued to resist the adoption. Former Water Commissioner Shimon Tal describes his attempts to incorporate contingencies for climatic extremes in desalination capacity planning being limited by the Ministry of Finance who only approved a budget for a lesser volume of desalination (Tal, 2010, personal communication). It therefore appears that even after the demonstration of technical and financial feasibility, the pace of implementation continues to be determined by political forces. Arguably, as will be shown in the next section, the continued politically induced delays to desalination during the 2000s necessitated emphasis on recycling to further decoupling and reflexivity in natural water use, in conjunction with further emphasis on efficiency.

5.7. Crisis management through furthering Type Two decoupling Despite the reforms, when low rainfall recurred from 2006, significant storage deficits remained, and attention again focused on the structure of the policy system. The government response was to initiate a National Investigation Committee on the water sector. The committee’s 2010 report identified a number of immediate and longer-term causes of the crisis, and argued for further actions. The Water Authority was criticised for its slow response to the water crisis; for failing effectively to implement price increases as a means to cover cost and manage demand; and failure to allocate sufficient resources to the development of a long-term Master Plan. There was also criticism of the Ministry of Finance for delaying desalination (State of Israel, 2010a). The committee argued that natural resources were insufficient on their own to meet water needs, and a series of augmentation options were re-evaluated. The issues highlighted in the report contributed to an approach that has clearly embraced increased Type Two decoupling.

The Committee placed strong emphasis on increased demand management until 500 MCM of desalination capacity is available. Increasing the use and monitoring of wastewater was recommended, along with greater urban water efficiency and reduced wastage. While acknowledging the need for mediumterm augmentation, the emphasis of the enquiry was on immediate and sustained reduction in natural water, including further institutional modifications, and technical and behavioural measures to promote demand management (State of Israel, 2010b).

5.8. Cementing Type Two decoupling Despite the need for further action identified by the 2010 committee, this study shows that the overall trend of water supply and consumption does appear to demonstrate a powerful and clear reflexive trend in natural water allocation and management, achieved via Type Two decoupling. A significant reduction of 376 MCM from 1,689 MCM was achieved in response to drought from 2007 to 2009. This 22% reduction led to an annual supply of 1,313 MCM of natural water in 2009, the lowest natural water supply since 1961. This trend is clearly observable in Figure 4. Due to decoupling, there was a reduced

M. Gilmont / Water Policy 16 (2014) 79–101

97

percentage impact of these cuts; reductions of the total supply across all sectors was only 16% (a 350 MCM reduction from 2,199 MCM), while for agriculture alone it was only 14% (a 170 MCM reduction from 1,186 MCM). The volumetric contribution of effluent was not significantly affected by the cuts, and the overall effluent and desalination supply increased by 26 MCM in 2007–2009. In 2009, however, effluent did experience a small 4% decline due to the 75 MCM reduction in domestic water use in that year, and the consequential reduced inflow to treatment plants. Domestic consumption was successfully reduced through politically contentious and publically unpopular price rises (Rinat, 2010). Political opposition to price rises meant that these instruments would have been difficult to implement without the independent Water Authority. The mitigation of allocation cuts to agriculture through effluent supply reduced the opposition to natural water cuts. Such opposition had been influential in shaping the drought response in 1991 and 1998.

6. Discussion, wider applicability and further research The reductions in the use of natural water in Israel use by 2009 to the levels of the early 1960s shows that successful Type Two decoupling has replaced the volumetric dependence realised during the Hydraulic Mission. It should be noted that even with decoupling, the sources developed during the period from the 1960s, including the water carrier, are still significant. Reductions in natural water are arguably not yet enough to facilitate a simultaneous rehabilitation of degraded sources through specific allocations and also maintain supply reliability. If the reductions are maintained, as is planned by the Water Authority Planning Department (OECD, 2011), such rehabilitation should be possible over time through rainfall and desalination-based recharge. The medium-term success of decoupling will be determined in part by the degree to which the rains of 2011–12 bring about a lasting rehabilitation of ground and surface storages. Political forces, namely the influence of the Ministry of Finance, have been shown to have influenced the timing, pace and nature of Type Two decoupling, including the emphasis on recycling and conservation as measures that should be thoroughly implemented first.

The data for 2009 is the last year that desalination contributed less than 10% of the total supply. For 2010, the plant at Hadera was operational, adding a further 100 MCM. By 2013 it is anticipated that both Sorek and the delayed Mekorot operated plant at Ashdod will be commissioned, taking capacity to over 500 MCM, with a target of 750 MCM by 2020 (State of Israel, 2010b). The evidence provided by the experience up to 2009 is, however, important because it demonstrates that Israel has been able to successfully negotiate the contentious allocation challenges of Type Two decoupling. This successful outcome has mainly been achieved through the substitution of natural water with treated effluents and significant cross-sector efficiency increases. It is argued here that the short-term impact of desalination up to 2009 appears to have been largely political rather than volumetric. Large cuts in natural water allocation from 1,803 MCM in 2004 to 1,313 MCM in 2009 (a 700 MCM reduction) appear to have been achieved through the promise that the pain of these cuts will be alleviated in the future by increased desalination capacity. Desalination only compensated for 125 MCM during the period of the cuts. The trend from 2009 appears to involve desalination facilitating future growth in water use rather than substituting for natural water (OECD, 2010).

Type Two decoupling achieved through desalination and recycling cannot be regarded as a perfect solution to Israel’s water supply problems. The technologies employed are not without their own challenges. In particular there are significant energy and carbon costs associated with both methods, as well

98

M. Gilmont / Water Policy 16 (2014) 79–101

as health concerns relating to the mineral content of desalinated water, for which various solutions are proposed (Penn et al., 2009; Avni et al., 2010). There are further un-investigated concerns about potential environmental and health impacts of effluent re-use including endocrine disruptors (OECD, 2010, Box 3.3). While demonstrating a degree of substitution of externalities which therefore question the effective application of the UNEP’s proposed Impact Decoupling for water, these emerging and debated issues are beyond the scope of this discussion.

The decoupling transitions experienced in Israel have general relevance. Every political economy is unique, and the very politicised transition of the installation of Type Two technologies and policy reforms will play out differently in other economies, as may the degree of political ease of Type One decoupling. Israeli outcomes should not be regarded as a predictor of how other political economies should or could react. However, there are numerous other economies in the Middle East and North Africa region which are facing very similar challenges to Israel. The circumstances include growing urban populations, growing economic diversity and increasing water consumption in cities often located close to large irrigation areas. Type Two decoupling is pertinent to future growth and socio-economic development scenarios where water policies include a wide range of water managing instruments. These include demand management across all sectors, the recycling of urban water, and the appropriate augmentation of supplies, including via desalination. Further research into both the rich data for Israeli decoupling and the applicability and emergence of comparable Type One and Type Two trends elsewhere is required to establish the full scope, sequencing and therefore generalisability of the concept.

7. Conclusions Two types of decoupling have been identified and conceptualised in a new water-specific model of decoupling. It has been shown that a semi-arid political economy can meet its multiple social and economic needs associated with reliable water supplies. It can at the same time recognise the need to maintain the health and sustainability of water resources, both for ecosystem services and their role in achieving supply reliability. Through a detailed evaluation of water management and policy in Israel, the two types of decoupling have been identified in national policies. The reforms have impacted irrigation policies and domestic water tariffs, as well as investment in recycling urban waste water and desalination.

Type One decoupling involves augmenting local water supplies by importing water-intensive food commodities to meet the demands driven by population increase and socio-economic development.

These imports result in a decoupling of the total levels of water required by an economy from the levels of natural water resources available.

In contrast, Type Two decoupling severs the link between internal water use and local availability of natural water resources. It has been shown that in Israel’s Type Two decoupling it has been possible to simultaneously achieve the often conflicting policy demands requiring the expansion of supply and the reduction of abstraction from environmental sources.

Because of the need to change consumption patterns by economic and regulatory measures, Type Two decoupling appears to be much more politically contentious and demanding than Type One decoupling. While Type One decoupling emerged over a few years in the early 1970s in Israel, Type Two, which first emerged in reports in the 1980s, took 15–20 years of attempts and political interactions before bearing fruit. However, a politically sustainable outcome is shown to be possible if the political

M. Gilmont / Water Policy 16 (2014) 79–101

99

processes are allowed to function, attitudinal changes allowed to develop, and sensible assessments of political and physical costs and benefits and impacts occur. A headline outcome of Type Two decoupling has been that while total supply has increased, natural water use is now at a level comparable with the early 1960s, with natural water reductions occurring simultaneously with total supply increases.

Acknowledgements The author would like to thank five anonymous Water Policy reviewers whose comments and suggestions enhanced this paper during the publication process. Thanks are also due to experts in the region and colleagues at King’s College London who provided valuable feedback during the preparation phase.

References Allan, J. A. (1995). The role of drought in determining the reserve water sector in Israel. Drought Network News (International Drought Information Center), 7(3), 21–23.

Allan, J. A. (2000). Contending environmental knowledge on water in the Middle East: Global, regional and national contexts.

In: Political Ecology: Science, Myth and Power. Stott, P., and Sullivan, S. (eds). Arnold, London, pp. 117–131.

Allan, J. A. (2001). The Middle East Water Questions. I.B. Tauris, London.

Allan, J. A. (2003). IWRM/IWRAM: a new sanctioned discourse? SOAS Water Issues Study Group Occasional Paper 50.

Available at: http://lwrg.org/water-management-and-economics.html?file=tl_files/pdf/IWRAM%20-%20a%20new%20sanctioned%20discourse.pdf [Accessed 28/8/2013].

Arlosoroff, S. (1997a). Israel – a case study of water resources management. EAWAG News. 43E, December 1997, 8–11.

Arlosoroff, S. (1997b). The public commission on the water sector reform (The general ideas underlying its recommendations).

International Water and Irrigation Review 5, 1–9.

Arlosoroff, S. (2006). Personal communication, September 2006.

Avni, N., Eben-Chaim, M. & Oren, G. (2010). The impact of large scale desalination on water quality: an operational perspective. ASCE Conference Proceedings, 371, 290.

Bar-Or, Y. (2000). Restoration of the rivers in Israel’s coastal plain. Water, Air and Soil Pollution 123, 311–321.

DEFRA (2011). Water for Life. Presented to parliament by the Secretary of State for Environment, Food and Rural Affairs by Command of Her Majesty, December 2011.

Dreizin, Y. (2006). Ashkelon seawater desalination project – off-taker’s self costs, supplied water costs, total costs and benefits.

Desalination 190, 104–116.

Dreizin, Y. (2010). Personal communication, October 2010.

EUEA (European Environment Agency) (2010). State and Outlook 2010 – Country Assessments: Spain. Published 26 November 2010. Available at: http://www.eea.europa.eu/soer/countries/es/soertopic_view?topic=freshwater [Accessed 19/02/2012].

FAO (2012a). FAOSTAT. Agricultural Trade Domain ‘Top Imports – Israel’ (1961–2010). Available via: http://faostat.fao.org/ site/342/default.aspx [Accessed 20/09/2012].

FAO (2012b). FAOSTAT. Agricultural Production Domain ‘Top Production – Israel’ (1961–2010). Available via: http://faostat. fao.org/site/339/default.aspx [Accessed 20/09/2012].

Feitelson, E. (2005). Political economy of groundwater exploitation: the Israeli case. Water Resources Development 21(3), 413–423.

Feitelson, E. (2006a). A retreat from centralized water management? The Israeli case. In: A History of Water – Volume 2: The Political Economy of Water. Coopey, R. & Tvedt, T. (eds). I.B. Tauris, London.

Feitelson, E. (2006b). Impediments to the management of shared aquifers: a political economy perspective. Hydrogeology Journal 14, 319–329.

Feitelson, E. & Fischhendler, I. (2007). Politics and institutions for groundwater management in a systemwide context.

International Symposium on Groundwater Sustainability, Alicante, 24–27 January 2006. Available at: http://aguas.igme. es/igme/isgwas/Ponencias%20ISGWAS/22-Feitelson.pdf [Accessed 21/11/2008].

100

M. Gilmont / Water Policy 16 (2014) 79–101

Feitelson, E. & Fischhendler, I. (2009). Spaces of water governance: the case of Israel and its neighbours. Annals of the Association of American Geographers 99(4), 728–745.

Feitelson, E., Fischhendler, I. & Kay, P. (2007). Role of a central administrator in managing water resources: The case of the Israeli water commissioner. Water Resources Research 43(11), November 2007, W11415.

Former Israeli Parliamentary Researcher (retired) (2010). Personal communication, October 2010.

Gleick, P. (2002). Soft water paths. Nature 418, 25 July 2002.

Gleick, P. H., Cooley, H., Cohen, M., Morikawa, M., Morrison, J. & Palaniappan, M. (2009). Peak water. In: The World’s Water 2008–2009: The Biennial Report on the Worlds Freshwater Resources. Island Press, Washington, DC. Gleick, P. & Palaniappan, M. (2010). Peak water limits to freshwater withdrawal and use. PNAS 107(25), 11155–11162, June 22 2010.

Inbar, Y. (2007). New standards for treated wastewater reuse in Israel. In: Wastewater Reuse – Risk Assessment, DecisionMaking and Environmental Security. Zaidi, M. K. (ed.). Springer, Dordrecht, The Netherlands, pp. 291–296.

Kantor, S. (2001). The National Water Carrier (Ha’Movil Ha’Artsi) (online). Haifa University, Haifa. Available at: http:// research.haifa.ac.il/~eshkol/kantorb.html [Accessed 19/09/2012].

Katz, D. (1994). A Model for Evaluating Conflict Over Shared Water Resources: The Arab-Israeli Conflict as a Case Study.

MA Thesis, University of Washington.

Kislev, Y. (2010). Rainfall measured at Nablus, 1874–2008 (personal communication).

Kronenberg, G. (2002). The Ashkelon 100 MCM/year BOT project. Desalination 152, 103–112.

Laster, R. & Livney, D. (2009). Israel: the evolution of water law and policy. In: The Evolution of the Law and Politics of Water. Dellapenna, J.W. & Gupta, J. (eds). Springer, Dordrecht, The Netherlands, pp. 121–137.

Menahem, G. (1998). Policy paradigms, policy networks and water policy in Israel. Journal of Public Policy 18(3), 283–310.

Mekorot (2009). Mekorot: Israel’s National Water Company. Online company profile Available at: www.mekorot.co.il/Eng/ NewsEvents/catalogs/Mekorot_tadmit_eng7.pdf [Accessed 10/11/2009].

Mekorot (2012). Mekorot participates in the ‘Yarkon River Salvation’ Plan. Newsletter no.3, September 2012. Available at: http://www.mekorot.co.il/Eng/WaTech/Newsletter/Pages/September2012/YarkonRiverSalvationProject.aspx [Accessed 11/ 6/2013].

Nativ, R. (2004). Can the desert bloom? Lessons learned from the Israeli case. Groundwater 45(5), 651–657.

Nishri, A., Stiller, M., Rimmer, A., Geifman, Y. & Krom, M. (1999). Lake Kinneret (The Sea of Galilee): the effects of diversion of external salinity sources and the probably chemical composition of the internal salinity sources. Chemical Geology 158, 37–52.

OECD (2001). Decoupling environment from economic growth. OECD forum 14 May 2001: The transition to Sustainable Development: are we making progress in Decoupling Economic Growth from Environmental Degradation. Available at: http://www.oecdobserver.org/news/fullstory.php/aid/453/Decoupling_environment_from_economic_growth_.html [Accessed 21/01/2012].

OECD (2010). OECD Review of Agricultural Policies: Israel, 2010. OECD Publishing.

OECD (2011). OECD Environmental Performance Reviews: Israel 2011. Available at: http://www.water.gov.il/Hebrew/ ProfessionalInfoAndData/2012/03-From-OECD-Environmental-Performance-Reviews-Israel-2011-WATER.pdf [Accessed 12/11/2012].

Palestinian Authority (2010). Climatic Bulletin (Year 2009). Available at: www.pmd.ps/ar/climatebollity2009.pdf [Accessed 10/06/2012].

Penn, R., Birnhack, L., Adin, A. & Lahav, O. (2009). New desalinated drinking water regulations are met by an innovative post-treatment process for improved public health. Water Science & Technology: Water supply. 9.3, 225–231.

Rinat, Z. (2010). Uri Shani, head of the Water Authority, is another drought tax on the way? Haaretz, 13 December 2010.

Rinat, Z. (2012). New life returns to Israel’s ‘river of death’. Haaretz, 13 September 2012. Available at: http://www.haaretz. com/news/features/new-life-returns-to-israel-s-river-of-death-1.464633 [Accessed 11/6/2013].

Shevah, Y. & Kohen, G. (1997). Economic considerations for water used in irrigation in Israel. In: Water: Economics, Management and Demand. Kay, M., Franks, T. & Smith, L. (eds). E. F. Spon, London.

Shuval, H. (2006). Personal communication, September 2006.

Smith, M.H., Hargroves, K. & Desha, C. (2011). Decoupling economic growth from freshwater extraction. In: Cents and Sustainability: Securing our Common Future by Decoupling Economic Growth from Environmental Pressures. Smith, M., Hargroves, K. & Desha, C. (eds). Earthscan, London.

M. Gilmont / Water Policy 16 (2014) 79–101

101

State of Israel (2002a). Transitional Master Plan for Water Sector Development in the Period 2002-2010: Executive Summary (English). Available at: http://gwri-ic.technion.ac.il/pdf/wcom/master.pdf [Accessed 21/11/2008].

State of Israel (2002b). Water in Israel: Consumption and Production 2001. Ministry of National Infrastructures, Water Commission, Demand Management Division. Available at: http://gwri-ic.technion.ac.il/pdf/wcom/demand.pdf [Accessed 01/05/ 2009].

State of Israel (2004, 2005, 2008, 2011, 2012a). Central Bureau of Statistics Annual Agriculture Accounts. Water Production and Consumption by Source and Purpose Available through: www.cbs.gov.il [Accessed 09/09/2012].

State of Israel (2009). Perennial Data for Lake Kinneret. Israel Water Authority Commission (in Hebrew). Available at: http:// www.water.gov.il/Hebrew/ProfessionalInfoAndData/Data-Hidrologeime/DocLib/PerennialData-Kinneret09.pdf [Accessed 5/11/2011].

State of Israel (2010a). National Investigation Committee on the Subject of the Water Economy in Israel – Abstract (English).

Haifa. (Personal communication.) Full report available (in Hebrew) at: http://elyon1.court.gov.il/heb/mayim/doc/sofi.pdf [accessed 07/10/2010].

State of Israel (2010b). Seawater Desalination in Israel: Planning, Coping with Difficulties, and Economic Aspects of Longterm Risks. Israel Water Authority, October 2010. Available at: http://www.water.gov.il/Hebrew/Planning-and-Development/ Desalination/Documents/Desalination-in-Israel.pdf [Accessed 24/10/2011].

State of Israel (2012b). Water Level of Aquifers in Selected Drillings and Average Salinity. Available at: http://cbs.gov.il/ shnaton63/download/st01_07.xls. [Accessed 8/6/2013].

State of Israel (2012c). Agriculture in Israel: The Industry Account Price Indices of Output and Input 2010–2011. Table 2 – Output, Input and Net Domestic Product in Agriculture (Value and Indices of Price and Quantity). Available at: http://www. cbs.gov.il/publications11/1459/pdf/t02_2.pdf [Accessed 12/6/2013].

Tal, A. (2002). Pollution in a Promised Land. University of California Press, Berkley.

Tal, S. (2010). Personal communication, October 2010.

Turton, A. R. (1999). Water and social stability: the Southern African dilemma. Paper presented at the 49th Pugwash Conference ‘Confronting the Challenges of the 21st Century’. Working Group 5 The Environment, 7–13 September 1999, Rustenburg, South Africa.

Turton, A. R. & Ohlsson, L. (1999). Water scarcity and social stability: Towards a deeper understanding of the key concepts needed to mange water scarcity in developing countries. SOAS Occasional Papers Water Series. Available online at: http:// www.soas.ac.uk/water/publications/papers/file38360.pdf [Accessed 30/7/2012].

UNEP (2011). Decoupling natural resource use and environmental impacts from economic growth. A Report of the Working Group on Decoupling to the International Resource Panel. Available at: http://www.unep.org/resourcepanel/decoupling/files/ pdf/decoupling_report_english.pdf [Accessed 21/02/2012].

UNEP (2012). Measuring water use in a green economy. A Report of the Working Group on Water Efficiency to the International Resource Panel. Available at: http://www.unep.org/resourcepanel/Portals/24102/Measuring_Water.pdf [Accessed 01/09/2012].

Weinberger, G., Livshitz, Y., Givati, A., Zilberbrand, M., Tal, A., Weiss, M. & Zurieli, A. (2012). The Natural Water Resources between the Mediterranean Sea and the Jordan River. State of Israel, Jerusalem. Available at: http://www. water.gov.il/Hebrew/ProfessionalInfoAndData/Data-Hidrologeime/DocLib4/water_report-MEDITERRANEAN-SEA-ANDTHE-JORDAN.pdf [Accessed 21/03/2012].

World Bank (2011). Population, Total SP.POP.TOTL. Available at: http://api.worldbank.org/datafiles/SP.POP.TOTL_ Indicator_MetaData_en_EXCEL.xls [Accessed 23/12/2011].

Received 22 October 2012; accepted in revised form 25 June 2013. Available online 17 September 2013

Water Policy 16 (2014) 102–123

Comparison of small-scale providers’ and utility performance in urban water supply: the case of Maputo, Mozambique Jigar Bhatt Columbia University, 1172 Amsterdam Avenue, New York, NY 10027, USA. E-mail: jdb2171@columbia.edu

Abstract Following independence from colonial rule, African governments struggled to cope with the legacy of fragmented water services and new demands of peri-urban population growth. Privatization was presented as a panacea that would expand and improve water supply. Small-scale independent water providers (SSPs) were meanwhile often the only actors ensuring that services were available to the peri-urban poor. Nonetheless, they were ignored and even vilified in ‘pro-poor’ strategies of water supply reform. Recent studies have actually demonstrated the important role SSPs play in serving the poor in African cities, however, substantial knowledge gaps remain. This study of SSP activities in Maputo, Mozambique provides rigorous empirical evidence about the performance of fully private SSPs vis-à-vis a privatized utility at both the provider and household level. The findings belie long-held notions of informal water provision as inferior and inefficient and formal sector privatization as the preferred strategy for reaching the poor. Improving water supply in African cities requires an understanding of the specific advantages of provider-types and avoiding universal cures.

Keywords: Maputo; Privatization; Pro-poor; Small-scale providers; Water supply; Water utilities

1. Introduction Urban water and sanitation (W&S) infrastructure development has followed different trajectories in many parts of the developing world compared with much of Europe and the United States. The modern paradigm for urban service delivery in developing countries has been: (1) public and (2) centralized. The paradigm was largely adopted from industrialized countries during the colonial period (Mkandawire, 2001). However, application in developing countries often had an explicit goal of not achieving universal service, with colonial families enjoying the benefits of centralized public service delivery while ‘native’ families living outside the urban core in the ‘periphery’ received substandard public or, more often, whatever privately acquired services they could find (Mamdani, 1996; Elate, 2004). doi: 10.2166/wp.2013.083 © IWA Publishing 2014

J. Bhatt / Water Policy 16 (2014) 102–123

103

After many developing countries declared independence in the 1960s and 1970s, national and municipal governments began to grapple with the colonial legacy of these bifurcated public-service delivery systems. A substantial proportion, often a majority, of households in many cities of the developing world lacked access to piped water, improved sanitation and electricity. Moreover, the deep poverty in these unserved areas made user financing of the new trunk infrastructure nearly impossible. In many countries the institutional capacity to plan, implement and administer large-scale urban infrastructure vanished along with the exodus of the colonial leadership (Isaacman, 2005; Carrington & Lima, 1996).

Newly independent governments in developing countries, in partnership with international development organizations, strove to expand public-service delivery systems to unserved areas of the city.

Governments and donors envisioned a universal level of safe, reliable, piped water provision that mimicked the kind of infrastructure found in former colonial powers (Graham & Marvin, 2001). The planning paradigm underpinning most of their efforts was that of a centralized, publicly administered W&S utility. However, financial resources and fiscal policies could not match European W&S standards. Also, in several cases, years of centralized, top-down management contributed to a culture of complacency and conformity (Biswas & Tortajada, 2010). As a result a ‘downward spiral’ in W&S infrastructure ensued (Singh et al., 1993).

Over the next 20–30 years, the limitations of this model for developing countries became increasingly apparent. Revenues continued to present a challenge as sub-Saharan Africa became mired in a debt crisis. Coverage of basic W&S services in major cities of the developing world did not keep pace with population growth and urban migration. Utilities often failed to serve low-income populations living in informal settlements, resulting in major coverage gaps and inequity in service provision (Cross & Morel, 2005). Lack of political will for public utility reform led to poor management, substandard service quality for existing customers and financial instability. Mainstream economists and major international development agencies began to associate the centralized service provision model with a lack of accountability to customers, cost padding, service diversion, high prices (especially to the poor), corruption and inefficiency (Davis, 2005).

2. The private sector in developing countries’ urban water supply 2.1. Privatization of urban water supply Private sector participation (i.e. privatization) in government-run industries expanded rapidly to the developing world in the 1990s following its adoption in 1980s Britain. Privatization was encouraged and supported by international development agencies and embraced by many developing countries’ policymakers. Proponents argued that private sector participation (PSP) would cure many of the ills attributed to services delivered via a centralized bureaucracy. The level of the private sector’s responsibility can range from management, as in lease and concession agreements which involve running the system and sometimes responsibility for major capital investments, to construction as buildoperate-transfers where the private sector builds or rehabilitates state-owned infrastructure, to all out transfer of public assets to the private sector through divestiture (these are rare). Nonetheless, they all share similar objectives of increasing (i) infrastructure financing, (ii) economies of scale, (iii) efficiency and (iv) financial sustainability.

104

J. Bhatt / Water Policy 16 (2014) 102–123

Enormous capital needs for infrastructure have driven privatization in developing countries. Neither public utilities nor sub-national governments have access to the amount of financing necessary to support these large capital investments (Davis, 2005; Bayliss, 2003). This is especially true in very lowincome or post-conflict countries.

For many, privatization is synonymous with a single, large private-sector operator. Because investment in the water sector is considered to be ‘lumpy’, or large relative to the size of the market, and long-term, often requiring several decades for cost-recovery, a single provider model (i.e. natural monopoly) is considered most economical. It is feared that having multiple providers could result in wasteful duplication of infrastructure and poor coordination of investments, operations and regulation (Ehrhardt, 2003; Solo, 1998).

Efficiency gains are often touted as privatization’s greatest benefit. Many consider that private management’s responsiveness to shareholder goals results in an efficiency advantage for private firms (Shirley, 2002; Yeaple & Moskowitz, 1995). Greater efficiency in the W&S sector includes efficiency of investment, usually improved through lower non-revenue water rates and efficiency of operations and maintenance, usually improved by lowering the unit cost of providing water.

Private sector participation contract structures are also supposed to achieve financial sustainability.

Privatization advocates believe that political incentive combined with politicians’ close relationships with W&S utilities sometimes leads to short planning horizons that inhibit proper water asset and resources management. In contrast, the private sector, with its long-term lease and concession contracts (usually 15 and 30 years) has incentives to manage infrastructure for longer time horizons to recover capital investments and derive profits (Davis, 2005).

Privatization has had mixed results with respect to achieving service delivery goals. Introducing the private sector, particularly powerful multinational corporations, to poorly performing public utilities was not the panacea advocates hoped it would be. The involvement of the private sector can accelerate capital investment and improve efficiency. However, investment falls short of contractual targets and efficiency is usually gained at the expense of public employees and provision to the poor. Involving the private sector also pressures water suppliers to return profits for shareholders (Davis, 2005). Such politically unpopular measures are probably why the literature on privatization is largely critical of PSP’s inability to meet expectations for universal coverage, adequate service and financial sustainability, particularly in subSaharan Africa (Bayliss & Fine, 2007; Hall & Lobina, 2007; McDonald & Ruiters, 2005; Holland, 2005). Still, PSP, including through multinational corporations, has not been abandoned and is likely to continue to play a role in affecting water policy in the coming decades (Bakker, 2013).

2.2. Small-scale private providers of water In newly independent former colonies small-scale independent water providers (SSP) did not fit the ‘modern infrastructure ideal’ paradigm of municipal services delivery and therefore were either ignored in the planning process or vilified for providing inferior and high-priced services relative to the public municipal provider (Njiru, 2004; Kariuki & Schwartz, 2005). The inconsistency of this view and the idea that SSPs could represent the type of efficient provider that privatization advocates envisioned have rarely been pointed out (Kjellén & McGranahan, 2006; Opryszko et al., 2009).

By the end of the 1990s the first set of SSP case studies and reports began to emerge (Snell, 1998; Solo, 1999; Wegelin-Schuringa, 1999; Collignon & Vezina, 2000; Albu & Njiru, 2002). These largely descriptive studies found that SSPs are not marginal actors in the urban water sector. In some cases SSPs

J. Bhatt / Water Policy 16 (2014) 102–123

105

serve up to 50% of a city’s population, often the peri-urban poor, making them attractive to some for meeting the Millennium Development Goals (McGranahan et al., 2006). By 2004 there were calls to embrace SSPs for their innovativeness and customer focus and to partner them with formal, municipal water providers (Njiru, 2004.). However, little information is still known about SSPs’ service characteristics and operating environments.

Two field-based studies attempted to fill this knowledge gap (Kariuki & Schwartz, 2005; Baker, 2009). What they found was that SSPs are heterogeneous. Their water sources can be independent (e.g. borehole) or dependent on the municipal provider (e.g. purchase of water for resale). They deliver water through piped networks, point sources such as kiosks, or mobile means such as tanker-trucks or animal-drawn carts. The surveyed providers do not deal with sanitation or wastewater disposal – they service only one part of the water cycle. They charged high unit prices, but these were more a function of a challenging operating environment, such as expensive sources of informal finance, than pure profit maximization (Kariuki & Schwartz, 2005).

These studies concluded that more rigorous, non-anecdotal evidence on SSPs was needed and that data on SSP service quality are nearly non-existent (Kariuki & Schwartz, 2005; Baker, 2009). A recent literature review found that nearly 60% of SSP literature is based on reports by international organizations and that SSPs are not well documented in independent, peer-reviewed journals. Data gaps include risks and benefits of SSP operations, pricing and profit structures and effective business models (Opryszko et al., 2009).

It is still unclear whether SSPs experience setbacks to expanding coverage, reaching the poor affordably, staying financially sustainable and maintaining service levels as large concessionaires and lessees of municipal W&S networks often do and under what circumstances, if any, they can be viewed as an integral part of a W&S service delivery strategy. This study investigates whether or not existing negative characterizations of informal water supply are merited, especially when compared to preferred strategies of centralized water supply and large-scale, corporate led PSP. It asks the question: ‘What role can SSPs play in Maputo’s water sector and in urban sub-Saharan African water supply more broadly?’ Evidence on SSP operations vis-à-vis a privately managed municipal utility can shed light on the advantages and limitations of both privatized, single provider and SSP service delivery models.

3. Study site and operating environment In this paper Maputo refers to the metropolitan region made up by Maputo, the capital city and its urban neighbor Matola, which are supplied by the same water system and together make up the nation’s primary urban area. Modern day Maputo was originally established as Lourenço Marques in 1781. At first just a Southern African port on the Indian Ocean, Lisbon declared Maputo to be Mozambique’s capital in 1898 (Figure 1). Official statistics from the United Nations Department of Economic and Social Affairs (UNDESA) put the city’s population at 2 million in 2005, 2.7 million in 2010 and a forecasted 3 million by 2015 (UNDESA, 2009). This population occupies approximately 320 km2 1 ( Jenkins, 2000). In 2005, 9% of these residents (approximately 155,000) lived in the cement city, which covers approximately 110 km2 (Aguas de Moçambique, 2006; Jenkins, 2000). The remaining residents live in peri-urban ‘bairros’ with limited provision of basic services (Figure 2). At 6.3 persons 1

In 1999 the administrative boundaries of the city encompassed a total area of 675 km2.

106

J. Bhatt / Water Policy 16 (2014) 102–123

Fig. 1. Maputo and Mozambique in regional context. Source: Earth Resource Systems International.

Fig. 2. Greater Maputo’s administrative boundaries and water distribution networks. Source: Matsinhe et al. (2008).

J. Bhatt / Water Policy 16 (2014) 102–123

107

per household, Maputo’s peri-urban neighborhoods are some of Mozambique’s densest. Peri-urban residents are also relatively poor and work in informal sectors such as domestic services, street vending, local service provision and small-scale productive activities.

The main source of the water supply system for Maputo is the Umbeluzi River. The Umbeluzi river basin covers a total area of 5,460 km2. The current utility’s water supply system is solely based on intake from Pequenos Libombos dam. Maputo’s second major water source is groundwater. State investment in Maputo’s urban infrastructure nearly halted during the civil war that began in 1977 (the war officially ended in 1992). Aguas de Maputo, the public water utility at the time, lacked the necessary revenue to rehabilitate, upgrade, or expand its infrastructure. The central government also seemed reluctant to serve peri-urban areas (Pinsky, 1982). The public utility’s coverage and service levels suffered as a result.

Although service provision has improved since independence, it has not kept pace with population growth; annual rates of growth averaged over 4% since 1995. In 2007 the network still only reached half of Maputo’s residents; 37% obtained water from utility connections or standpipes and 13% retrieved water from a neighbor’s connection to the utility network (Aguas de Moçambique, 2006).

Since the current utility network does not reach all of Maputo’s inhabited areas, alternative sources, namely groundwater, have been in use for decades, particularly by peri-urban residents.

In 1998 the Mozambican government looked at external financing sources for utilities’ upgrading. The World Bank agreed to fund the National Water Development Projects to improve coverage and performance and created a new institutional architecture, dubbed the ‘Framework for Delegated Management’, to attract PSP for Mozambique’s urban water provision. Large scale PSP would become part of a triumvirate institutional design where: (i) a private operator manages the utility, (ii) a public investment agency, FIPAG (Fundo de Investimento e Património do Abastecimento de Água), takes over municipal water supply assets and investment duties, and (iii) an independent regulator, CRA (Conselho de Regulação do Abastecimento de Água), monitors the utility’s operations and mediates conflicts between the utility and FIPAG. The National Water Directorate (DNA), under the Ministry of Housing and Public Works, is responsible for promoting this framework’s implementation, mobilizing funds and preparing legal support when necessary. The municipal government plays a very minor role and has little sector influence in the face of these larger actors. It is still, however, the people’s representative and is responsible for voicing community concerns to the other institutions (Thompson, 2005).

Immediately following the transfer of municipal assets to FIPAG, a private operator was sought to take over utility operations. FIPAG entered into a 15-year lease contract in September 1999 with Aguas de Moçambique (AdM), a consortium of French, Portuguese and Mozambican companies, to manage potable water supply service in Maputo. In 2005 Aguas de Portugal, the international private partner in the delegated management framework, was the major shareholder of AdM (FIPAG, 2001).

Its lease officially expires in late 2014, however, in March, 2011 Aguas de Portugal sold its share in Aguas de Moçambique to FIPAG signaling, for the time being, the end to large-scale PSP in Maputo’s water supply.

When it took over operations in 2001, AdM demonstrated some progress with respect to coverage, service quality and bill collections. However, by 2005 progress had peaked and by 2007 its performance on the above factors had stagnated or worsened. In 2005 more than half of Maputo’s residents still had to rely on alternative water sources. AdM’s experience of large-scale PSP is similar to many subSaharan African utilities that underwent privatization in the 2000s.

Maputo’s SSPs, predominately in the form of what Kariuki & Schwartz (2005) call independent small network operators, are now an integral and dominant part of Maputo’s water service delivery. Like those

108

J. Bhatt / Water Policy 16 (2014) 102–123

also found in Philippines and Cambodia, they are fully private operators that draw water up from boreholes with an electric pump into storage tanks. They then distribute this stored water to standpipes next to the water source and nearby private connections via a gravity fed system (Figure 3). Those treating their water do so with common locally available chemicals (e.g. chlorine) at the point of storage.

In 2005, Maputo’s SSPs served between 175,000 and 210,000 peri-urban residents with standpipes that offered water during daylight hours and private connections that largely provided a 24 hour service.

They have existed in Maputo since as early as the 1970s but have multiplied significantly since the mid1990s. At the time of this study (2005/06) there were approximately 210 small water systems in Maputo.

Some reports have estimated that there has been a doubling in SSP activity, with more than 400 systems in operation today (Dardenne et al., 2009; Blanc et al., 2009). Thus, while Maputo’s water sector is still characterized by an ‘archipelago’ – multiple water supply sources and ownership structures interspersed among the utility network (Bakker, 2003) – it increasingly resembles a ‘dual system’ between the utility provider and SSPs (Moran & Batley, 2004).

4. Study methodology and data sources 4.1. Methodology The analysis is based on a comparison of each provider type’s summary statistics. A standardized set of indicators, found to be important to water sector policy and planning, were selected for comparison.

Fig. 3. Some of Maputo’s SSPs. Source: Author.

J. Bhatt / Water Policy 16 (2014) 102–123

109

Each provider’s system characteristics and performance are then summarized, juxtaposed and the difference between the values calculated. The data sources for each system are different and therefore parametric tests of significance (‘t-tests’) could not be calculated; means and other summary statistics must be compared in absolute terms. For a list of formulas used in the provider analysis, please refer to the Appendix (available online at http://www.iwaponline.com/wp/016/083.pdf) The SSP sample is based on 38 providers’ data collected at a single point in time (January 2006). The AdM sample is based on official documents and reports from 2005 and 2006 and 72 months of performance data (January 2002 to December 2007) from the single utility provider. These were the only years for which AdM data were provided to the author. The use of a representative sample of SSPs provides a snapshot of broader SSP operations in peri-urban Maputo and helps control for variation over multiple providers. Meanwhile, the use of 6 years of AdM data represents the utility’s performance since privatization in 1999 and helps control for variation over time. Ideally one would analyze SSP performance over time as well, but that was not possible and is a limitation of the analysis. Another limitation is the significant amount of time that has lapsed since SSP data collection.

A similar analysis is conducted for water quality. In this case, however, the samples for each provider were collected together at a single point in time by a single entity, allowing for t-tests of difference in means between important measures of water quality. Because AdM conducts its own water quality tests, data from AdM’s official samples were included to ensure consistency of results. Such a triangulation process was not possible with the SSP water samples because water quality testing for SSPs is not fully institutionalized.

Data from households were used to corroborate findings from the provider surveys and analyses.

Household data came from a single instrument allowing for stratification according to customer type – SSP or AdM – and an analysis of difference of means for a number of demographic, economic and water supply conditions through t-tests. Since existing literature considers SSPs’ customers and services to be very different from those of utilities (i.e. worse), in all presented comparisons and tests, the absence of significant differences between the two client groups is considered a significant and meaningful finding.

4.2. Utility data sources AdM collects monthly and annual performance data on its water supply operations in the Maputo region for internal review. Datasets with monthly performance data were provided by AdM management for the period January 2002 to December 2007. In addition to this performance data, management provided access to copies of the lease contract, annual reports for 2005 and 2006 and fee and tariff schedules. Internal data were supplemented and triangulated with data from the World Bank’s International Benchmarking Network for Water and Sanitation Utilities (IB-NET). For analysis purposes, the monthly and annual performance data from AdM and IB-NET were averaged for the 6-year period wherever possible.

4.3. Provider survey The provider survey was informed by 2 months of qualitative fieldwork with 15 SSPs and senior managers, planners and engineers at DNA, FIPAG, AdM, CRA and the City of Maputo. The final survey was based on a stratified random sample of Maputo’s SSPs that benefited from an existing

110

J. Bhatt / Water Policy 16 (2014) 102–123

provider list (i.e. sample frame) of all Maputo SSPs compiled by FIPAG. The list of 190 providers, representing 210 systems across all peri-urban neighborhoods where SSPs are active2, was stratified based on providers’ self-reported household connection figures into ‘large systems’ (greater than 30 connections), ‘small systems’ (fewer than 30 connections) and ‘unknown’. A stratified random sample that mirrored the distribution of systems between these three categories (large, small, or unknown) was drawn. The study’s original target was 30 randomly chosen systems. Favorable field conditions and cooperative providers allowed for a total of 38 observations, resulting in an even sample of 19 large and small systems each. The survey was carried out in January 2006 and interviews were generally conducted in providers’ homes (also most often the location of their systems).

4.4. Household survey The household survey was preceded by 1 month of qualitative research with dozens of peri-urban residents and pilot data collection from 500 households in the neighborhood of Costa do Sol. The sample frame for this study was based on Maputo’s fourth district – an urban district under Maputo municipality made up of four peri-urban neighborhoods (Hulene ‘B’, Albazine, Mahotas, FPLM).

The fourth district is where both AdM and SSPs operate and there is more overlap between the two provider types than in the central city or further out peri-urban districts. This made it an ideal location to compare the two provider types. A comprehensive list of households was not available for the neighborhoods so the sample frame was developed by interviewers systematically walking the neighborhoods and numbering households. Households were then selected randomly from the lists for interviews.

The survey was carried out by Estamos, a local non-governmental organization (NGO), with financial and technical assistance from WaterAid, an international charity and the Massachusetts Institute of Technology. Local interviewers collected data in September 2005. The survey consisted of 950 households but our analysis sample includes those 403 households which meet the selection criteria for analysis: the household’s primary water source for domestic needs is either (1) a private piped connection in the home or yard, (2) local standpipe, or (3) a neighbor’s piped connection in the home or yard and this source is provided either by an SSP or AdM. Of these 403 households, 330 get their primary water supply from SSPs while 73 get their primary water supply from AdM. 4.5. Water quality survey Water quality data were collected by random sampling by the NGO Water and Sanitation for the Urban Poor (WSUP) in the peri-urban neighborhoods of Maxaquene A, B, C and D, Ferroviario A and Mavalane B (these neighborhoods overlap with the provider survey but not the household survey). Samples were collected in November, 2008. Eighty-three water sources were sampled; 51 sources belonged to AdM and 32 sources belonged to SSPs. The AdM sample was a combination of network distribution center and well water while all the SSP samples came from SSP deep-wells.

The tests measured for nitrates, nitrites, turbidity, chlorites, ammonia, fecal coliform bacteria, pH and total dissolved solids.

2

Because some providers run more than one system, the number of systems, 210, is higher than the number of providers on the list.

J. Bhatt / Water Policy 16 (2014) 102–123

111

5. Findings and discussion For all tables and analysis, instances where sample sizes (n) differ from the samples cited in Section 4 are due to a narrower selection criteria or missing data on the particular indicator or variable being analyzed. Where applicable, confidence intervals at the 95% confidence level are provided in parentheses below the mean and separated by semi-colons. All monetary figures are inflation adjusted and in 2005 US dollars unless specified otherwise. Cells marked with ‘N/A’ mean ‘not applicable’ or ‘data not available’. Differences are always differences in means between the SSP value and the AdM value (SSP minus AdM) unless specified otherwise.

5.1. System characteristics AdM’s performance is mixed relative to other major utilities in sub-Saharan Africa (Table 1). It scores low on major metrics of service delivery and efficiency. In 2006, AdM’s coverage level in Maputo was about half that for the average utility in southern and eastern Africa. While it produced significant quantities of water per capita, it also lost a great deal of that to leakage and other forms of non-revenue water (NRW). AdM’s NRW rates were almost twice that of its peers. Hours of service also lagged behind, with AdM providing service for only half the day whereas other utilities were much closer to a 24 hour service. Where the picture is more favorable is in unit operating costs, where AdM spends less than some of its peers to produce a cubic meter of water (in general, most African utilities spent more than they earned in 2006).

Most SSPs whose operations have been documented began operating around 2000. Maputo’s SSPs are an average size among their peers, with a typical provider having 50 household connections (non-household connections such as businesses are rare for SSPs). However, they provide more water to their connections than other SSPs and employ standpipes more frequently than their urban counterparts in Manila. Maputo SSPs are also more profitable and concerned with water quality than similar entrepreneurs in south-east Asia (Table 2).

For the remainder of this paper ‘SSPs’ refers to Maputo’s SSPs. SSPs have typically been in business for 6 years with entrepreneurs establishing their systems while in their early forties. Female entrepreneurs play an active role – on average they own or run a fifth of systems. The typical entrepreneur has 3 more years of formal education than the average Maputo adult. Selling water is not a SSP’s only source of income with the majority of entrepreneurs – up to three-quarters – having other regular Table 1. AdM performance compared with other utilities in sub-Saharan Africa (all figures are sample means).

Water coverage (%) Water produced (liters/person/day) NRW per connection (m3/conn/day) Unit operating costs (US$/m3 sold) Hours of service per day

AdM

n

Southern Africa

n

Eastern Africa

n

All Africa

34 332 1.1 0.63 12

41 41 41 41 41

77 222 0.60 0.79 21

43 43 43 43 43

66 124 0.55 0.40 18

134 134 134 134 134

65 145 0.57 0.59 17

Figures may differ from other tables owing to differing data sources. Figures are for 2006, the only year data were available from the Africa Utility Performance Assessment.

Source: Water Operators Partnerships (2009).

112

J. Bhatt / Water Policy 16 (2014) 102–123

Table 2. Maputo SSP characteristics compared to similar SSPs in south-east Asia.

Length of time in business (years) Number of household connections Percentage with standpipe as part of water system Water sales per connection per day (m3) Profit margin (%) Percentage having water quality regulated or tested

n

Maputo

n

Manila

n

Cambodia

37 38 38 35 34 38

6.3 75 68 0.65 55 61

34 16 35 11 29 75

6 54 22 0.46 1 46

75 75 75 75 75 75

8 249 0 0.17 9 27

All figures are sample means.

All SSPs in the table are fully private, small network operators. All data were collected in 2006. SSP data in Manila come from urban operators while data in Cambodia come from operators in small-towns and rural areas.

Sources: Data on Manila’s providers are drawn and adapted from Davis et al. (2006); data on Cambodian providers are drawn and adapted from Baker (2009).

employment. They are relatively wealthy as they use their own financing to fund the initial investment (92% self-financed their systems in-part or whole). This investment is used both to source water and develop and extend the network. Systems typically cost US$10,000, or US$200 per household connection (the sample average for a system cost was US$13,804).

A SSP typically serves 50 households through piped connections and dozens more through standpipes. As a group they serve only a small fraction of their cities’ populations compared with the utility. However, they are growing exponentially faster than the utility and serve large segments of residents currently without access to utility services. SSPs in Maputo seem to rival utilities in network density. This is all the more striking since many of Maputo’s SSPs deliver water in less dense, periurban expanses.

Most SSPs are not formally registered and operate in the informal economy. However, many pay some type of fee or tax, have a desire to formalize their activities and are ready to pay taxes to do so. SSPs are labor intensive. This is due to the full-time stand pipe operators that entrepreneurs employ to manage delivery and collection of payments. SSPs also employ meters more frequently than utilities. This strategic use of labor and aggressive metering aids cost recovery and financial sustainability. The notion that high staff-to-connection ratios are a determinant of inefficient supply may be misguided or only relevant for a centralized municipal system (Table 3).

5.2. System costs, prices and financial sustainability Electricity is by far a SSP’s most burdensome operating cost. SSPs use small, motorized electric pumps to draw water from the aquifer up into storage tanks. These pumps are very expensive to operate.

As a result, electricity can make up 50% of a provider’s recurring costs and is also responsible for unit price increases over time. Labor, services and spare parts (mostly electric pumps) are also substantial costs. Pumps break down frequently and repairing them is costly, as is finding a replacement. Most pumps have to be imported, placing high exchange rate and import tariff costs on a SSP. Utilities struggle most with high staff costs, which has led to criticisms of bloated bureaucracies in the privatization debates. They also have high repairs and maintenance costs (Table 4).

J. Bhatt / Water Policy 16 (2014) 102–123

113

Table 3. System and service characteristics of Maputo SSPs and AdM. n

SSPs

n

AdM (6-year average, 2002/07)

Difference

Water source Population served

N/A 210

N/A 6

Umbeluzi river 623,608

N/A 490,888

Coverage rate: % of service area Number of household connections

N/A 210

Deep wells Total: 132,720 (91,350; 174,300) Total: 10 Total: 15,750 (9,870; 21,840)

6 55

26 67,091

Total number of operating stand pipesa Connections with meters (%) Household connections per km of piped network Average annual growth rate in household connectionsb (%) Number of employees per 1000 connectionsc

210 38 31

Total: 231 (151; 313) 62 (48; 76) 146 (71; 221)

55 N/A N/A

36 82,841 (81,300; 84,383) 355 (339; 371) 50 108

124 12 38

20

214 (42; 385)

6

4

200

37

27 (17; 38)

53

7.6 (7.3; 7.8)

19.4

Figures are sample means unless stated otherwise. 95% confidence intervals in parentheses. a Several private operators standpipes use rubber hoses that extend from the water source rather than traditional standpipes with faucets. b SSP growth rates examined over the life of the SSP (typically 5–6 years) while utilities’ growth rates were examined over a 6-year period between early 2002 and late 2007. c These employee figures only include paid employees; SSP owners and their family members are not included.

Table 4. Recurrent cost profiles for Maputo SSPs and AdM.

Water (%) Staff (%) Electricity (%) Services (repairs etc.) (%) Spare parts, tools, etc. (%) Debt service (%) Fuel (%) Equipment (%)

n

SSPs

AdM (yr ¼ 2005)

Difference

AdM (yr ¼ 2006)

Difference

38 38 38 38 38 N/A 38 37

0 28 (21; 36) 50 (42; 58) 6 (2; 9) 10 (5; 15) N/A 1 (0; 2) 3 (1; 5)

1 20 10 15 1 3 1 2

1 8 (1; 16) 40 (32; 48) 9 ( 13; 6) 9 (4; 14) N/A 0 ( 1; 1) 1 (0; 4)

1 20 8 16 1 5 1 2

1 8 (1; 16) 42 (34; 50) 10 ( 14; 7) 9 (4; 14) N/A 0 ( 1; 1) 1 (0; 4)

Average percentage of total monthly recurrent operating costs spent by provider on indicated expense category is given. 95% confidence intervals are in parentheses.

SSP administrative connection and unit costs from private connections are much higher than those of the utility. However, findings from the household survey below show that the disparity between connection costs is somewhat reduced once labor and material costs are accounted for. Also, because of the utility’s increasing block tariff, SSPs’ and AdM’s prices per cubic meter are comparable after a household reaches 10 m3 of water consumption. This is important because the average household connected to both SSPs and AdM use at least 15 m3 per month. An important observation for pro-poor service delivery is that SSPs charge the same per jerrican (a commonly used 20 l container) as AdM does (Table 5).

114

J. Bhatt / Water Policy 16 (2014) 102–123

Table 5. Connection and service fees for Maputo SSPs & AdM.

a

Administrative household connection fee (US$) Service fee per m3 for connected households: 1st block, 1–10 m3 (US$) Service fee per m3 for connected households: 2nd block, 11–20 m3 (US$) Service fee per m3 for connected households: 3rd block, 21–30 m3 (US$) Service fee per 20 l container of water (US$) Typical monthly bill for metered household connection using 15 m3 (US$) SSP typical fixed monthly fee for unlimited monthly water supply (US$)

n

SSPs

AdM (yr ¼ 2005)

Difference

33 28

Mean: 34 (25; 43) Mean: 0.90 (0.84; 0.97)

18 0.30

16 (7; 25) 0.60 (0.54; 0.67)

28

Mean: 0.90 (0.84; 0.96)

0.72

0.18 (0.12; 0.24)

28

Mean: 0.89 (0.83; 0.96)

0.76

0.13 (0.07; 0.20)

26 28

Mean: 0.02 (0.2; 0.2) 12.45

0.02 6.60

0 (0; 0) 5.85

20

10.38

N/A

N/A

95% confidence intervals in parentheses, where applicable. a This cost reflects the administrative fee only and excludes other significant costs, sometimes borne by the household, such as labor and materials.

Maputo’s SSPs demonstrate allocative efficiency through the market mechanism by providing water to those willing to pay. They (about 25%) also offer a range of discounts relevant to their peri-urban neighbors’ needs, such as to the elderly who live alone, ‘visibly poor’ and widows, using local information to tailor allocation while continuing to meet customer demand. Some provide free jerricans while others provide lower fixed monthly fees. Fifty-three percent of providers offer household clients a fixed monthly fee option that is typically US$10.38 for an unlimited supply of water.

SSPs are financially sustainable, more so in fact than the utility provider (Table 6). In Maputo, SSP working ratios are less than 0.5 and they spend less than half as much as the utility to produce a cubic meter of water. There are several caveats to this finding. First, Maputo’s SSPs do not pay for land, water, or regulation and their chemical treatment costs vary widely. However, in a static model that took water and chemical treatment costs into account, SSPs still remain profitable with an average working ratio of Table 6. Financial sustainability and profitability of Maputo SSPs and AdM.

3

Total production cost per m of water sold (US$) Revenue per m3 of water sold (US$) Profit margin (%) Working ratio Percentage with customer base in payment arrears a Collection ratio (%)

n

SSPs

n

AdM (6 year average, 2002/07)

Difference

34

0.30 (0.22; 0.37)

6

0.68

0.38 ( 0.46; 0.31)

34 34 34 38

0.71 (0.59; 0.84) 55 (37; 63) 0.45 (0.36; 0.53) 58 (41; 74)

6 6 6 N/A

0.55 27 1.27 N/A

0.16 (0.04; 0.29) 82 (64; 90) 0.82 ( 0.91; 0.74) N/A

N/A

N/A

N/A

71

N/A

All figures are sample means. 95% confidence intervals in parentheses where applicable. a A customer base in arrears is one where 25% or more of the provider’s customers pay late.

J. Bhatt / Water Policy 16 (2014) 102–123

115

0.6 (Bhatt, 2006). Nonetheless, unregulated water extraction by SSPs poses a challenge to sustainable service delivery going forward.

Second, AdM did not have to ‘pay’ for its land either. Its concession contract involved the management of infrastructure that remained in FIPAG’s hands. In 2005, AdM paid US$420,000 to lease the system and US$396,000 to its regulator, CRA. This amounted to 4% of its monthly operating costs, or US$0.04 per m3 of water supplied. Its water costs together amounted to no more than 1% of monthly operating costs in 2006, or less than US$0.01 per m3 of water supplied. In short, AdM paid 5% of its monthly costs in usage fees including leasing, regulation and water usage3.

Third, AdM receives heavy technical, legal and financial support from the government of Mozambique, the World Bank and other international agencies that SSPs are not officially eligible for. It is unclear whether this harms or helps AdM, but because most of the assistance is related to infrastructure investments, it should theoretically have a neutral or positive effect on AdM’s operating position (Le Blanc, 2008).

SSPs as a group, on average, demonstrate a strong operating efficiency that can withstand the additional expenses related to water extraction, treatment and regulation. Larger SSPs also can experience economies of scale – those that supply more water per month experience lower unit costs (Figure 4). Thus a trait thought to belong to natural monopolies and the central provider is experienced by the informal SSP. SSPs’ production and supply efficiencies are higher, leading to higher revenues.

But their unit prices to household connections are also higher. Both help to explain their comparatively stronger profit margin and working ratio.

5.3. System service performance In Maputo, SSPs provide service levels that are at least equal to AdM’s (Table 7). This is true of actual performance, as demonstrated by the data below and perceived performance, as subjectively

Fig. 4. Water supply and unit operating costs among Maputo’s SSPs (n ¼ 34 providers). Source: The provider survey. Graph produced in Stata v.10.0.

3

Payment terms taken from the Maputo Revised Lease Contract.

116

J. Bhatt / Water Policy 16 (2014) 102–123

Table 7. Service performance indicators for Maputo SSPs and AdM.

Household connections: Number of hours of service/day Standpipes: number of hours of service/day Volume of water sold per household connection/day (m3) Per capita volume sold to connected households (lpcd)a Volume of water sold per standpipe per day (m3) Percent unaccounted-for waterb Number of days in past year could not provide water

n

SSPs

n

AdM (6 year average, 2002/07)

Difference

38 25 35 35 25 27 38

19 (17; 21) 12 (12; 12) 0.65 (0.48; 0.83) 103 (76; 131) 4.9 (3.1; 6.7) 7 (4; 9) 2 (1; 3)

63 N/A 53 53 53 55 60

13 (13; 13) N/A 0.73 (0.65; 0.80) 115 (104; 126) 3.9 (3.2; 4.7) 58 (56; 60) 0.5 (0; 0.5)

6 N/A 0.12 12 1.0 51 1.5

All figures are sample means. 95% confidence intervals in parentheses where applicable. a Using an average household size of 6.3 in Maputo. lpcd stands for liters per capita per day. b Unaccounted for water for SSPs is self-reported while utility figures are calculated. Unaccounted for water is the difference between the amount of water produced by the system and delivered to/consumed by customers, divided by the total amount produced.

stated by the majority of providers in interviews (Bhatt, 2006). It is unclear if adequate service levels are a result of SSPs’ highly localized service areas or competition between SSPs. It could be argued that the SSP-customer relationship results in a ‘short-route’ to accountability whereby customers can voice their preferences and concerns to providers or hold them accountable to a minimum service quality (Ahmad et al., 2005). At the same time, peri-urban customers also have the option to switch to another SSP nearby. While SSPs informally collaborate over service areas to prevent such competition and there exist formidable switching costs in terms of connection fees, neither of these results in the absolute monopoly conditions under a single, universal utility provider.

5.4. Water quality One of the major concerns regarding the rise of SSPs is water quality. Without a designated, official regulatory body monitoring quality, the fear is that SSPs are unaware of their water quality and have little incentive to bear the costs to ensure its safety. The concern is legitimate given the risks and consequences. Nonetheless, it is wrong to assume that SSPs are neither aware of water quality nor have incentive to ensure its safety. Over half of all SSPs, 63%, had their water tested and checked by the Ministry of Health of the Government of Mozambique (pointing to the polycentric nature of not only water supply but water quality monitoring). Most have their water tested twice per year. Those receiving warnings related to their water quality usually respond immediately to rectify the situation, which often involves treating their water with chemicals or cleaning their water tanks. Among the providers that were not being monitored, four self-tested their water, often sending samples to a laboratory or testing it on site.

To understand water quality between SSP and AdM sources better, we analyzed 83 water samples taken by WSUP in November, 2008 (Table 8). The SSP samples test providers’ deep-wells, while the AdM sources test AdM’s distribution network as well as their wells in peri-urban areas. The one significant difference between the nine standard water quality measures is on nitrate levels. The SSP sources contained 126 mg of nitrates/liter more, on average, than AdM sources. This finding was significant at the 99% confidence level. This higher level is of concern and most likely due to a

J. Bhatt / Water Policy 16 (2014) 102–123

117

Table 8. Water quality parameters for Maputo SSP and AdM water sources.

WSUP 2008

AdM 2008

n

SSPs

n

AdM network

Difference

n

AdM network

pH Conductivity Turbidity Nitrates (mg l 1) (NO3)

32 32 32 32

7.3 (7.1; 7.5) 926.9 (836; 1,017) 0.81 (0.30; 1.33) 282.6 (242.5; 322.7)

51 51 51 51

7.4 (7.3; 7.6) 919.9 (863.0; 976.8) 0.99 (0.42; 1.57) 157.1 (117.8; 196.4)

322 322 322 N/A

7.41 744.8 2.2 N/A

Nitrites (mg l 1) (NO2) Chlorites (mg l 1) (Cl) Ammonia (mg l 1) (NH3) Total dissolved solids (mg l 1) (Ca2CO3) Fecal coliform (# of col/100 ml)

32 32 32 32

0.039 (0.025; 0.053) 135.2 (121.6; 148.9) 0.14 (0.04; 0.24) 180.8 (162.4; 199.3)

51 51 51 51

0.046 (0.023; 0.068) 147.5 (136.2; 158.8) 0.13 (0.07; 0.20) 192.9 (182.2; 203.7)

0.1 7.0 0.18 125.5*** ( p ,0.0001) 0.007 12.3 0.01 12.1

N/A 63 N/A 322

N/A 135.8 N/A 156.1

32

26.1 (12.3; 39.9)

51

13.2 (4.7; 21.7)

12.9

306

0

All figures are sample means. 95% confidence intervals in parentheses.

P-scores are provided in parentheses below differences in means. Statistically significant differences in means at the 99% level are represented by three asterisks (***), 95% level by two asterisks (**) and 90% level by one asterisk (*).

Source: Water and Sanitation for the Urban Poor (WSUP) and Aguas de Moçambique (AdM).

combination of the SSP sampling and operating area. The SSP sample consisted entirely of well water and nitrates can easily seep into well water especially in peri-urban areas where pit-latrine use is high. It should also be noted that both SSPs and AdM were above the 50 mg l 1 nitrates level recommended by the World Health Organization.

We should be careful not to draw definitive conclusions from these water quality findings. Where and when samples are taken is important because water quality conditions can change quickly. For example, one study that examined groundwater potential and water quality at 35 SSP wells found acceptable nitrate levels but unacceptable total coliform levels in 14 of the wells (four had unacceptable fecal coliform levels). That study concluded that overall, ‘The results of the groundwater quality assessment suggest that as for today, the quality of water tapped by SSIPs generally conforms to guidelines standards’ (Matsinhe et al., 2008: 417–419). Despite this sanguine claim, the study calls for regular, vigilant monitoring as abstraction continues to stress groundwater resources and this study recommends the same.

5.5. Household characteristics SSP customers look demographically similar to utility customers in areas where both provider types operate; there were no statistically significant differences in household demographics between the two samples. We are therefore comparing similar groups of customers (Table 9).

We do find, however, that SSP clients in Maputo are materially better off than utility clients (Table 10). The results are statistically significant for having electricity in the home and owning a radio. Based on correlation analysis by the Mozambican Ministry of Planning and Finance, increased radio ownership is a proxy for higher welfare (National Directorate of Planning and Budget, 2004).

118

J. Bhatt / Water Policy 16 (2014) 102–123

Table 9. Peri-urban households’ demographic characteristics by customer type.

Household size (persons) Number of children under 12 years old Percent female headed household Percent having living spouse Household head years of education completed Number of years living in Maputo

n

SSPs

n

AdM

Difference

330 330 300 328 316 284

6.4 (6.0; 6.7) 2.1 (1.9; 2.3) 26 (21; 30) 64 (58; 69) 5.9 (5.6; 6.3) 5.4 (4.3; 64)

73 73 68 70 62 59

6.2 (5.5; 6.8) 2.0 (1.7; 2.4) 24 (13; 34) 64 (53; 76) 5.3 (4.7; 6.0) 7.8 (5.1; 10.6)

0.2 0.1 2 0 0.5 2.5

All figures are sample means. 95% confidence intervals in parentheses.

P-scores are provided in parentheses below differences in means. Statistically significant differences in means at the 99% level are represented by three asterisks (***), 95% level by two asterisks (**) and 90% level by one asterisk (*).

Table 10. Peri-urban households’ socio-economic characteristics by customer type.

Percentage renting home Percentage households using improved sanitation source Percentage households with electricity in the home Number of radios in household Number of bicycles in household

n

SSPs

n

AdM

Difference

327 328 322 328 328

2 (1; 4) 59 (53; 64) 38 (31; 45) 0.9 (0.8; 0.9) 0.1 (0.1; 0.2)

67 73 68 68 67

4 (0; 10) 51 (39; 62) 16 (4; 28) 0.7 (0.6; 0.8) 0.1 (0; 0.2)

2 8 22*** ( p , 0.001) 0.2** ( p , 0.02) 0

All figures are sample means. 95% confidence intervals in parentheses.

P-scores are provided in parentheses below differences in means. Statistically significant differences in means at the 99% level are represented by three asterisks (***), 95% level by two asterisks (**) and 90% level by one asterisk (*).

This finding is not entirely surprising. These households are located in District 4, one of the few areas of the city where AdM and SSP networks overlap and look alike demographically. For these otherwise similar households to be able to afford the SSPs’ higher connection fees and per unit costs they would have to be somewhat wealthier.

Using the household survey findings we can triangulate to corroborate or refute our findings from the provider survey. On the whole, the findings from the household survey corroborate our findings from the provider survey regarding service quality and affordability.

5.6. Service quality Clients rate similar levels of satisfaction with service regardless of the provider. No statistically significant results were found between the two samples; it appears that SSPs and utilities are comparable in their service quality with respect to hours of service, water pressure, taste, color, odor and perceived safety. Also, both SSP and AdM clients consume similar amounts of water per day (Table 11).

While statistical significance may be absent, the findings are significant because most have considered and many still consider SSPs as providing a substandard level of service. The literature has cited their small size, informal operating environment, duplication of investments and other characteristics as hindering efficiency, thereby preventing adequate service delivery. For Maputo’s SSPs, such characterizations would be overstated.

J. Bhatt / Water Policy 16 (2014) 102–123

119

Table 11. Peri-urban households’ perspectives of water provider’s service quality.

Number of hours of service per day Percentage rating water pressure as good Percentage stating water has good/normal taste Percentage stating water is clean/colorless Percentage stating water is odorless Percentage treating water before drinking or cooking Water consumption household connections and standpipes (lpcd) Water consumption standpipes (lpcd)

n

SSPs

n

AdM

Difference

318 325 329 330 327 326

16.6 (15.7; 17.5) 78 (73; 82) 97 (95; 99) 100 (100; 100) 96 (94; 98) 10 (7; 13)

73 73 73 73 73 71

15.3 (13.4; 17.1) 73 (62; 83) 97 (93; 100) 100 (100; 100) 93 (87; 99) 7 (1; 13)

208

35 (32; 38)

62

36 (28; 44)

1

109

32 (29; 35)

54

34 (26; 42)

2

1.3 5 0 0 3 3

All figures are sample means. 95% confidence intervals in parentheses. lpcd stands for liters per capita per day.

P-scores are provided in parentheses below differences in means. Statistically significant differences in means at the 99% level are represented by three asterisks (***), 95% level by two asterisks (**) and 90% level by one asterisk (*).

Table 12. Peri-urban households’ reports of monetary costs of water.

Years with a household water connection Connection cost (US$)a Percentage paid connection costs in installments Cost per 20 l container (USD$)

n

SSPs

n

AdM

Difference

158 137 135 80

2.8 (2.4; 3.2) 70 (58; 79) 4 (1; 8) 0.02 (0.02; 0.02)

11 9 10 41

4.7 (1.2; 8.3) 59 (34; 84) 20 (0; 50) 0.02 (0.02; 0.02)

1.9*** (p , 0.01) 11 16 0

All figures are sample means. 95% confidence intervals in parentheses.

P-scores are provided in parentheses below differences in means. Statistically significant differences in means at the 99% level are represented by three asterisks (***), 95% level by two asterisks (**) and 90% level by one asterisk (*). a Includes all connection costs, including administrative fees, materials and labor.

5.7. Affordability Owing to limited sample size and missing data among our utility clients, it was difficult to determine statistical significance of the monetary costs of water. Nonetheless, it appears from the descriptive data that SSP clients pay slightly more on average to obtain a private connection and more often have to provide these funds up front. This is likely because SSPs use connection charges as a primary means of paying off costly loans for infrastructure investments or recouping savings used to finance system construction. While cubic meter charges were not calculated in the household survey, we do find that SSPs and AdM charge identical rates for 20 l of water at their standpipes (Table 12).

6. Conclusion This study has sought to address a major evidence gap in the literature on SSP operations by providing rigorous, representative evidence on a particular type of SSP – Maputo’s independent network operators – which has become the dominant source of water supply in peri-urban Maputo. It did so in an innovative way by comparing SSP operations against the operations of a central municipal

120

J. Bhatt / Water Policy 16 (2014) 102–123

provider that recently underwent large-scale PSP. By comparing these two models of service provision, one decentralized and fully private, the other central and partially privatized, the analysis informed if and how SSP operations accomplished sector goals.

What we can conclude with confidence is that Maputo’s SSPs have the potential to contribute successfully to sector goals. While they charge higher household connection fees and tariffs, as expected, they also serve an equity function by making standpipe water available to peri-urban residents at the same price as the utility. They are far from inefficient. Their revenues are at least twice their costs, making them financially sustainable. This is owed in part to those higher tariffs but also judicious employment of metering and human resources. Despite using what many engineers would consider ‘substandard’ infrastructure and technology, they are able to provide a quality of service on par with the utility provider. In the end, SSP customers consume as much water as utility customers.

One concern is water quality. On the whole, the characteristic that SSPs are unconcerned with quality does not apply in Maputo. However, there are indications that water quality could become a problem going forward. SSPs would be willing partners rather than adversaries in a strategy to ensure safe drinking water for their customers. After all, their highly localized operations make them all the more susceptible to repercussions if any of their customers were harmed as a result of their services.

It is important to keep this study’s findings in context. First, both the enthusiasm for PSP and rapid rise of SSPs are in part the result of international development agencies’ and policy makers’ broader disappointment and backlash against municipal services in the 1980s and into the 1990s. We do not know how AdM would have fared in the absence of privatization or if instead its public operations were strengthened and supported. While AdM’s experience with PSP was overall rather disappointing, the utility has of late been trying to increase access and improve its operations in peri-urban areas and has done so with some success. Notable accomplishments include new storage and distribution centers in District 4 and Matola and network expansion to the poor in Costa do Sol. Second, SSPs are heterogeneous and each type has its own operating profile and operating environment. The small-network operator in Maputo is a particular type of SSP, one that is able to contribute significantly to coverage goals and achieve cost savings through economies of scale. This is more difficult, for example, for kiosk operators such as those found in Nairobi, many of which operate precariously with a thin profit margin (Baker, 2009). Also, Maputo’s SSPs rely on groundwater while another city’s SSPs may rely on surface or utility water, raising different issues for sustainability.

Even among Maputo SSPs there are important differences. The average and median figures in this study, while useful for comparison between the small providers and utility, hide important power struggles within the informal water supply economy itself. Larger SSPs dominate the informal market and have the most to gain from any formalization process with FIPAG. These same providers dominate the SSP association, which threatened to shut down SSP water services during negotiations with a weakened municipality over a licensing scheme. Any successful ‘public–private’ partnership should take the unique functioning of the informal economy, SSPs’ social and labor relations and technological innovations into account or risk undoing some of the progress SSPs have already made.

Lastly, operating environments in growing sub-Saharan African cities are changing rapidly. In Maputo there are twice as many SSPs now compared to when this study was conducted. As competition in periurban areas among SSPs increases, profit rates and service characteristics will be likely to change. Thus, the answer is not to elevate and glorify one system type over another, but to cultivate a synergy of systems where each type is aligned with sector objectives to help build on its unique advantages. Research going

J. Bhatt / Water Policy 16 (2014) 102–123

121

forward should focus on the level of political and market power each provider possesses, what they do with it and how this impacts the most vulnerable consumers.

Acknowledgements The author owes a debt of gratitude to the advice and direction of Dr Jenna Davis. The author would also like to thank Ella Lazarte, Laura Bloomfield and Valentina Zuin and three anonymous reviewers for their feedback and assistance. The study would not have been possible without the collaboration and cooperation of Frederico Martins of Aguas de Moçambique (formerly of FIPAG), Manuel Alvarinho at CRA and the many entrepreneurs who opened their doors to the interviewers. Excellent data collection was the result of efforts by Stelio Jeremias Lhachuaio, Ernesto Jalane, Celeste Cazamulo, Pedro Pimentel, Beto, Macamo and the staff at WaterAid Maputo and Estamos. Financial and institutional support from the MIT Department of Urban Studies and Planning and World Bank Water and Sanitation Program-Africa is gratefully acknowledged. Any views and errors are those of the author alone.

References Aguas de Moçambique (AdM) (2006). Data and Statistics. Operations Department, Maputo.

Ahmad, J., Devarajan, S., Khemani, S. & Shah, S. (2005). Decentralization and Service Delivery. World Bank Publications, Washington, DC. Albu, M. & Njiru, C. (2002). The role of small-scale independent water providers in urban areas. Waterlines 20(3), 14–16.

Baker, J. (2009). Opportunities and Constraints for Small Scale Private Service Providers in Electricity and Water Supply: Evidence from Bangladesh, Cambodia, Kenya and the Philippines. World Bank Public-Private Infrastructure Advisory Facility, Washington, DC. Bakker, K. (2003). Archipelagos and networks: urbanization and water privatization in the South. The Geographical Journal 169(4), 328–341.

Bakker, K. (2013). Neoliberal versus postneoliberal water: geographies of privatization and resistance. Annals of the Association of American Geographers 103(2), 253–260.

Bayliss, K. (2003). Utility privatization in Sub-Saharan Africa: a case study of water. The Journal of Modern African Studies 41(04), 507–531.

Bayliss, K. & Fine, B. (2007). Debating the Provision of Basic Utilities in Sub Saharan Africa: A Response to Nellis. United Nations International Poverty Centre One Pager, Brasilia, Brazil, p. 32.

Bhatt, J. D. (2006). The Strategic Use of Small Scale Water Providers: An Analysis of Private Sector Participation in PeriUrban Maputo. Massachusetts Institute of Technology, Cambridge, MA. Biswas, A. & Tortajada, C. (2010). Water supply of Phnom Penh: an example of good governance. International Journal of Water Resources Development 26(2), 157–172.

Blanc, A., Cavé, J. & Chaponnière, E. (2009). Les Petits opérateurs privés de la distribution d’eau à Maputo : d’un problème à une solution? Agence Française de Développement, Regards Croises, Paris, France, p. 85. Available at: http://www.afd.fr/ webdav/site/afd/shared/PUBLICATIONS/RECHERCHE/Scientifiques/Documents-de-travail/085-document-travail.pdf.

Carrington, W. J. & de Lima, P. J. F. (1996). The impact of 1970s repatriates from Africa on the Portuguese labor market.

Industrial and Labor Relations Review 49(2), 330–347.

Collignon, B. & Vezina, M. (2000). Independent Water and Sanitation Providers in African Cities: Full Report of a TenCountry Study. World Bank Water and Sanitation Program, Washington, DC. Cross, P. & Morel, A. (2005). Pro-poor Strategies for Urban Water Supply and Sanitation Services Delivery for Africa. World Bank Water and Sanitation Program. Washington, DC.

122

J. Bhatt / Water Policy 16 (2014) 102–123

Dardenne, B., Razafinjato, G. & Dardenne, L. (2009). Le modèle Technique et Commercial des Petits Opérateurs de Réseaux Privés d’Eau à Maputo. Agence Française de Développement, Paris.

Davis, J. (2005). Private-sector participation in the water and sanitation sector. Annual Review of Environment and Resources, 30(1), 145–183.

Davis, J., Lazarte, E. & Lasmarias, N. (2006). Small-scale Private Providers of Water in Manila: a Profile of Costs, Revenues and Prices. World Bank Water and Sanitation Program, Washington, DC. Ehrhardt, D. (2003). Impact of market structure on service options for the poor. In: Infrastructure for Poor People: Public Policy for Private Provision. Penelope, J. B. & Timothy, I. (eds). World Bank, Washington, DC. Elate, S. (2004). A continuous urban history: European colonial urban images and the development of African cities. In: Seventh International Conference on Urban History, Athens, Greece.

FIPAG (2001). Maputo Revised Lease Contract. Fundo de Investimento e Património do Abastecimento de Água, Maputo.

Graham, E. S. & Marvin, S. (2001). Splintering Urbanism. Routledge, London.

Hall, D. & Lobina, E. (2007). Profitability and the poor: corporate strategies, innovation and sustainability. Geoforum 38(5), 772–785.

Holland, A. C. S. (2005). The Water Business: Corporations Versus People. Zed Books, London.

Isaacman, A. (2005). Displaced people, displaced energy and displaced memories: the case of Cahora Bassa, 1970–2004.

The International Journal of African Historical Studies 38(2), 201–238.

Jenkins, P. (2000). City profile: Maputo. Cities 17(3), 207–218.

Kariuki, M. & Schwartz, J. (2005). Small-Scale Private Service Providers of Water Supply and Electricity: A Review of Incidence, Structure, Pricing and Operating Characteristics. Social Science Research Network Scholarly Paper No. ID 822386. Rochester, NY. Kjellén, M. & McGranahan, G. (2006). Informal Water Vendors and the Urban Poor. International Institute for Environment and Development (IIED), London.

Le Blanc, D. (2008). A Framework for Analyzing Tariffs and Subsidies in Water Provision to Urban Households in Developing Countries. United Nations Department of Economic and Social Affairs, Working Paper No. 63, New York.

Mamdani, M. (1996). Citizen and Subject: Contemporary Africa and the Legacy of Late Colonializm. Princeton University Press, Princeton.

Matsinhe, N., Juizo, D., Rietveld, L. C. & Persson, K. (2008). Water services with independent providers n peri-urban Maputo: Challenges and opportunities for long-term development. Water SA 34(3), 411–420.

McDonald, D. & Ruiters, G. (2005). The Age of Commodity: Water Privatization in Southern Africa. Earthscan, London.

McGranahan, G., Njiru, C., Albu, M., Smith, M. & Mitlin, D. (2006). How Small Water Enterprises can Contribute to the Millennium Development Goals: Evidence from Dar es Salaam, Nairobi, Khartoum and Accra. Loughborough University, Leicestershire, UK. Mkandawire, T. (2001). Thinking about developmental states in Africa. Cambridge Journal of Economics 25(3), 289–313.

Moran, D. & Batley, R. (2004). Literature Review of Non-State Provision of Basic Services. Paper commissioned by DFID from Governance Resource Centre, University of Birmingham, Birmingham, UK. National Directorate of Planning and Budget (2004). Poverty and Wellbeing in Mozambique: The Second National Assessment.

Ministry of Finance and Planning, Maputo.

Njiru, C. (2004). Utility-small water enterprise partnerships: serving informal urban settlements in Africa. Water Policy 6(5), 443–452.

Opryszko, M., Huang, H., Soderlund, K. & Schwab, K. (2009). Data gaps in evidence-based research on small water enterprises in developing countries. Journal of Water and Health 7(4), 609–622.

Pinsky, B. (1982). The Urban Problematic in Mozambique: Initial Post-independence Responses, 1975–80. Centre for Urban and Community Studies, University of Toronto, Toronto.

Shirley, M. (2002). Thirsting for Efficiency: The Economics and Politics of Urban Water System Reform. Elsevier, Oxford.

Singh, B., Ramasubban, R., Bhatia, R., Briscoe, J., Griffin, C. C. & Kim, C. (1993). Rural water supply in Kerala, India: how to emerge from a low-level equilibrium trap. Water Resources Research 29(7), 1931–1942.

Snell, S. (1998). Water and Sanitation Services for the Urban Poor, Small-Scale Providers: typology & profiles. World Bank Water and Sanitation Program, Washington, DC. Solo, T. M. (1998). Competition in Water and Sanitation: The Role of Small-Scale Entrepreneurs. Public Policy for the Private Sector, Note No. 165. World Bank, Washington, DC.

J. Bhatt / Water Policy 16 (2014) 102–123

123

Solo, T. M. (1999). Small-scale entrepreneurs in the urban water and sanitation market. Environment and Urbanization 11(1), 117–132.

Thompson, G. (2005). Adapting Regulation to the Needs of the Poor: Mozambique Case Study. Consultoria em Desenvolvimento Social Lda, Maputo.

United Nations Department of Economic and Social Affairs (2009). World Population Prospects: The 2009 Revision and World Urbanization Prospects. United Nations Department of Economic and Social Affairs Population Division, New York.

Water Operators Partnerships (2009). Africa Utility Performance Assessment. Final Report, Available at http://www.wsp.org/ sites/wsp.org/files/publications/WOP_Report.pdf.

Wegelin-Schuringa, M. (1999). The SSIP Model and Community-Based Providers of Water and Sanitation Services. United Nations Development Program Conference on Public-Private Partnerships for the urban environment, Bonn.

Yeaple, S. & Moskowitz, W. (1995). The Literature on Privatization. The Federal Reserve Bank of New York Research Paper No. 9514, New York.

Received 21 May 2013; accepted in revised form 28 June 2013. Available online 3 October 2013

Water Policy 16 (2014) 124–143

Benefits of rainwater harvesting for gardening and implications for future policy in Namibia L. Woltersdorfa, A. Jokischb and T. Klugea a

b

Institute for Social-Ecological Research (ISOE), Hamburger Allee 45, 60486 Frankfurt/Main, Germany Corresponding author. Institute for Water Supply and Groundwater Protection, Wastewater Technology, Waste Management, Industrial Material Cycles, Environmental and Spatial Planning (IWAR), Darmstadt University of Technology, Franziska-Braun-Str. 7, 64287 Darmstadt, Germany. E-mail: a.jokisch@iwar.tu-darmstadt.de

Abstract Rainwater harvesting to irrigate small-scale gardens enhances food self-sufficiency to overcome rural poverty.

So far rainwater harvesting is not encouraged by the Namibian National Water Supply and Sanitation Policy nor supported financially by the Namibian government. This study proposes two rainwater harvesting facilities to irrigate gardens; one collects rain from household roofs with tank storage, the second collects rain on a pond roof with pond storage. The aim of this paper is to assess the benefits of rainwater harvesting-based gardening and to propose policy and financing implications for the Namibian government. We investigate the benefits of rainwater harvesting through a literature review, a cost–benefit analysis, monitoring of project pilot plants and a comparison with the existing irrigation and drinking water infrastructure. The results indicate that rainwater harvesting offers numerous benefits in technological, economic, environmental and social terms. The facilities have a positive net present value under favourable circumstances. However, material investment costs pose a financing problem. We recommend that government fund the rainwater harvesting infrastructure and finance privately garden and operation and maintenance costs. Integrating these aspects into a national rainwater harvesting policy would create the conditions to achieve the benefits of an up-scale of rainwater harvesting based gardening in Namibia.

Keywords: Benefits; Central-northern Namibia; Cost–benefit analysis; Financing; Gardening; Rainwater Harvesting; Roof and ground catchments; Water policy; Water supply

1. Introduction Rainwater harvesting for the irrigation of household gardens buffers the dry season and droughts (van Steenbergen & Tuinhof, 2009). Rainwater harvesting consists of a wide range of technologies that can be divided into in situ and ex situ techniques to collect and store water (Barron, 2009). In situ rainwater harvesting are soil management strategies that enhance rainfall infiltration and reduce surface runoff, doi: 10.2166/wp.2013.061 © IWA Publishing 2014

L. Woltersdorf et al. / Water Policy 16 (2014) 124–143

125

such as terracing, pitting or conservation tillage practices. The rainwater capture area is within the field where the crop is grown and the soil serves as a capture and storage medium at the same time. Ex situ technologies have capture areas external to the point of storage, being a natural soil surface with limited infiltration capacity or an artificial surface with low or no infiltration capacity. Commonly used impermeable surfaces are represented by rooftops, roads, pavements and slopes. Storage systems are often wells, dams, ponds or cisterns (Barron, 2009).

Owing to increasing water scarcity worldwide, in recent decades rainwater harvesting has experienced rapid expansion in many countries around the world (Barron, 2009). Especially in semi-arid regions, governments have promoted rainwater harvesting to raise agricultural yields and bridge dry periods.

Examples include the Laikipia District in Kenya (Hatibu & Mahoo, 1999; Malesu et al., 2006), the Western Pare Lowlands in Tanzania (Senkondo et al., 2004), Rajasthan and Gujarat in north-western India (Agarwal et al., 2001) and the Gansu Province in north central China (Li et al., 2000; Barron, 2009).

These regions are characterised by a semi-arid climate with short rainy seasons, high annual potential evaporation, severe seasonal droughts and water shortages and low agricultural productivity. South Africa and the Indian state of Rajasthan have already integrated rainwater harvesting into their national water policy (DWAF, 2004; Mwenge Kahinda et al. 2007; UN-HABITAT & Government of Madhya Pradesh, 2007). A general precondition to make rainwater harvesting practically and economically feasible is an annual precipitation of at least 300 mm, unless other sources are extremely scarce (Worm & van Hattum, 2006).

Namibia is the driest country in sub-Saharan Africa. In central-northern Namibia annual rainfall ranges from 300–600 mm, 96% falling from November to April (Heyns, 1995; Kluge et al., 2008; Sturm et al., 2009). The area is characterised by a semi-arid climate with short rainy seasons, high precipitation variability, alternating droughts and floods, ephemeral river systems and brackish or saline groundwater (Heyns, 1995). Presently, most drinking water is abstracted from a reservoir, the Calueque Dam in Angola on the perennial Kunene River that is shared between Angola and Namibia, and transported through an extensive grid of canals and pipelines. Most settlements in the region have access to such supplies in sufficient quantity to serve their drinking water requirements (Heyns, 1995). However, there is no infrastructure to supply irrigation water to rural communities. Many poor households depend on rain-fed subsistence farming during the rainy season to secure their livelihoods (Republic of Namibia, 2006; Republic of Namibia, 2008b). In rural and remote areas the incidence of poverty is particularly pronounced with 38% of the population being poor (Republic of Namibia, 2008a). Agricultural yields are generally very low, leaving many households vulnerable to food insecurity and inadequate food supplies (Republic of Namibia, 2008b; Werner, 2011). A survey conducted by the Food and Agricultural Organization showed that many inhabitants, especially women, wish to extend their garden activities. However, the biggest limiting factor is the lack of sufficient and affordable water for irrigation. Thus most respondents stated that they need help to collect rain (Dima et al., 2002). The Namibian Third National Development Plan (NDP3) recognised the low and erratic rainfall and the poor soil quality of the region to be major impediments to a meaningful poverty reduction (Republic of Namibia, 2008b). While the Government of Namibia has responded by developing a comprehensive policy framework to promote household food security, insufficient attention is given to encourage micro- to small-scale local food production (Werner, 2011). Current policies and legislation encourage the use of alternative water sources (Republic of Namibia, 2008a). During the 1950s and 1960s several attempts have been made to harvest rain in uncovered pump storage dams in central-northern Namibia. However, owing to poor water quality, caused by evaporation, pollution and salinisation, the dams fell into

126

L. Woltersdorf et al. / Water Policy 16 (2014) 124–143

disrepair (Driessen & Jokisch, 2010). New investments in more appropriate rainwater harvesting infrastructure could have a broad range of benefits and give essential impetus for the regional economy and poverty reduction. However, in spite of its potential in Namibia, rainwater harvesting has so far not been considered in the current National Water Supply and Sanitation Policy (Republic of Namibia, 2008c) nor in the latest Water Act (Republic of Namibia, 2004).

The aim of this study is to assess the benefits of rainwater harvesting in irrigating small-scale gardens and to propose policy and financing implications for Namibia. In addition, the goal was to draw possible generalisations and broader implications for other regions. The study summarises the benefits identified in previous studies and presents research results of a cost–benefit analysis and first monitoring results for the most promising pilot rainwater harvesting facilities. The amount of investment and number of created jobs related to an up-scaling of rainwater harvesting is modelled and considered in relation to existing Namibian investment in irrigation and drinking water supply. Financing problems are revealed and a financing concept and policy implications are proposed to up-scale the technology in Namibia.

2. Pilot rainwater harvesting facilities in central-northern Namibia The project CuveWaters1 introduced three different options for ex-situ rainwater harvesting in centralnorthern Namibia. The pilot plants were built in the villages Epyeshona and Iipopo in the Oshana region and were conceived based on a preliminary literature research (Gould & Nissen-Petersen, 2006), a participatory demand-responsive approach with local communities (Deffner & Mazambani, 2010; Deffner et al., 2012; Zimmermann et al., 2012) and consultations with Namibian ministries and the Namibian Desert Research Foundation. During this process the pilot plants were adjusted to local needs and wishes in terms of size, combinations and materials. The three introduced facilities differ in terms of harvesting surface and storage media; the first consists of a corrugated iron roof (100 m2) and a tank (30 m3) either made of ferrocement, bricks or polyethylene. Such tanks can be used for single households and public buildings (schools, clinics, etc.) and are sufficient to irrigate up to 90 m2 of cultivated garden area. A second pilot facility collects rainwater from a concrete ground catchment (480 m2) and a greenhouse roof (160 m2) and stores the water in a covered underground ferrocement tank (120 m3) and a covered and sealed pond (80 m3). The stored water irrigates an outside garden (900 m2) and a greenhouse (160 m2) jointly used by six households. A third pilot facility collects rainwater from nearby ephemeral rivers, so-called Oshanas, at the height of the rainy season and stores the water in a covered ferrocement underground tank and a pond with a combined storage capacity of 400 m3. The stored water is sufficient to irrigate a 1,000 m2 outdoor garden area and a greenhouse of 176 m2, which are jointly managed by ten households. The Oshanas are difficult to use for permanent irrigation owing to high evaporation rates and therefore quick quality degradation and thus salinisation of the water.

This study assesses the two most promising rainwater harvesting facilities based on a preliminary assessment of the pilot plants ( Jokisch et al., 2011). The first is the ferrocement tank with roof catchment at household level as piloted in Epyeshona village (Figure 1). The second is the pond with roof catchment at community level which is an optimal combination of piloted facilities in Epyeshona based on project experience and costs (Figure 2; Table 1). The possible duration of irrigation of gardens 1

http://www.cuvewaters.net.

L. Woltersdorf et al. / Water Policy 16 (2014) 124–143

127

Fig. 1. Rainwater harvesting with roof catchment and ferrocement tank at household level.

Fig. 2. Rainwater harvesting with roof catchment and pond at community level.

Table 1. Proposed rainwater harvesting options in central-northern Namibia.

Rainwater harvesting option Tank with roof catchment household level Pond with roof catchment community level

Tank material

Catchment material

Catchment area (m2)

Storage volume (m3)

Ferrocement Dam liner

Corrugated iron Corrugated iron

100 285

30 80

with harvested rainwater depends on the irrigation technique, cropping pattern, garden area and the extent of the rainy season. Considering these factors, the stored water is sufficient for the irrigation of one or two additional annual growth seasons. In the project region, rainwater harvesting is meant to enhance the water supply for the productive irrigation of small-scale gardens and not to serve as a substitute for drinking water. However, in remote areas far from the existing pipeline grid harvested rainwater could also be treated and serve as drinking water.

3. Methodology 3.1. Benefits of rainwater harvesting for gardening The benefits of rainwater harvesting were assessed through a literature review and categorised in technological, economic, environmental and social terms. In this section, non-market benefits were listed in

128

L. Woltersdorf et al. / Water Policy 16 (2014) 124–143

a qualitative manner since placing monetary values on environmental and social non-market costs and benefits is extremely difficult, controversial and not always meaningful (Atkinson & Mourato, 2008).

3.2. Financial cost–benefit analysis In this section, we assessed the financial benefits of rainwater harvesting-based gardening in monetary terms. We carried out a financial cost–benefit analysis by identifying monetary costs and benefits of rainwater harvesting and gardening for a household or micro-entrepreneur. A cost–benefit analysis involves the identification of costs and benefits occurring over the economic life of a project (Gilpin, 2000; Pearce et al., 2006; Ward, 2012). The common method of reducing costs and benefits over the lifespan of a facility to a unique value is the net present value (NPV) method (LAWA, 2005; Pearce et al., 2006). Key steps are first to identify the costs and benefits of a project, second to quantify costs and benefits in monetary terms as far as possible and third to discount costs and benefits over the lifetime of the project with a selected discount rate. In purely economic terms, the production of a good is economically justified when the total benefits exceed the total costs (Gilpin, 2000). Benefits correspond to the value of gardening produce at market prices, while costs are equal to expenses. A medium discount rate of 5% was used over an estimated life span of 40 years for the ferrocement tank and 20 years for the pond. These values are based on experiences made by the responsible Kenyan rainwater harvesting consulting company ‘One World Consultants’ (Kariuki, 2012, personal communication) which has constructed more than 100 rainwater harvesting tanks and ponds in several countries in eastern and southern Africa. The NPV has been calculated as Equation (1) (LAWA, 2005): NPV ¼ I0 þ

T X

Rt †(1 þ i) t

(1)

t¼1

where NPV ¼ net present value, I ¼ investment, t ¼ time period from 0 to T, Rt ¼ inflow-outflow in period t, T ¼ time horizon (life span), and i ¼ discount factor. The costs of a rainwater harvesting and gardening facility include material investment costs, labour construction costs and operation and maintenance costs. Material costs include the tank, the pipes and gutters for the roof, the garden fences and the drip irrigation system, Operation and maintenance costs include annual materials costs for spare parts, seeds, fertilizer and pesticides. In a first step, we conducted a cost–benefit analysis including all these costs. In a second step, to show the potential for a more positive cost–benefit ratio, we included only labour, garden material, operation and maintenance costs and excluded the material costs for the rainwater harvesting facility. We calculated with material costs that occurred during the pilot construction phase, operation and maintenance costs were estimated based on local costs and first project experience during the pilot phase. Additionally, we estimated material costs without the conditions of the project.

Financial benefits, the revenue from gardening products at market prices, were modelled based on crop yields, local irrigation requirements, garden area and local market prices. Crop yields were taken as indicated by Price Waterhouse Coopers (2005). Possible garden areas to irrigate with the harvested rainwater were calculated with modelled local irrigation requirements. Specific crop water requirements were calculated with local climate data using the Food and Agricultural Organisation (FAO) software CROPWAT 8.0 (FAO, 1992). A drip irrigation system efficiency of 0.75 was used

L. Woltersdorf et al. / Water Policy 16 (2014) 124–143

129

assuming a conveyance efficiency of 0.85 and an application efficiency of 0.9, calculated according to Brouwer et al. (1989) In a preliminary assessment four garden variants considering the amount of annual harvested rainwater were modelled (Woltersdorf et al., 2013). The garden size was fitted so that the rainwater harvesting facilities are sufficient for full irrigation with a frequency of 3 out of 4 years, a probability level recommended as appropriate by the FAO (Savva & Frenken, 2002). This study presents a subsistence and a market garden variant. The market scenario contains tomatoes planted in the pilot village Epyeshona, modelled assuming one annual growth cycle with a market price monitored in the market of Epyeshona in 2011. The subsistence garden variant contains vegetables and fruits suitable for household consumption; the surplus can be sold at local markets, planted for two annual growth cycles (Woltersdorf, 2010). Owing to the lack of local market prices, prices were assumed as indicated by Price Waterhouse Coopers (2005) which presents wholesale market prices for imported horticulture products from January to December 2003 to Namibia in N$ per ton. The follow-up study from Price Waterhouse Coopers in 2008 (Price Waterhouse Coopers, 2008) was not used to owing to the unavailability of all data needed. In addition, the CuveWaters Project monitors local prices, which are, however, inconsistent so far owing to the lack of experience of the rainwater harvesting tank owners. In both garden variants yields and revenues are determined to be achieved in 3 out of 4 years, when the garden area can be fully irrigated. In 1 out of 4 years, owing to natural precipitation variability, precipitation is lower and not sufficient to irrigate the garden area fully and therefore yields and revenues will be lower; this is not presented in this study. Further detail regarding garden variants, precipitation probability analysis and calculation of irrigation requirements exceeds the scope of this paper and is provided by Woltersdorf et al. (2013).

3.3. Monitoring pilot plants The rainwater harvesting pilot plants were monitored in terms of maintenance effort and costs, water use and gardening input and output among other criteria. Most data were monitored by tank owners, while some data were also monitored by project team members. This paper presents the first monitoring results from the pilot village Epyeshona of yields of harvested vegetables and local market prices achieved in 2011. Monitored market prices were compared to market prices used in this study for the cost–benefit analysis indicated by Price Waterhouse Coopers (2005).

3.4. Comparison of rainwater harvesting facilities to the Namibian green scheme project The Namibian government plans to implement an ambitious agricultural project known as the green scheme project (Republic of Namibia, 2008d). In order to put our proposed rainwater harvesting and gardening infrastructure in the light of the local situation, rainwater harvesting and associated garden facilities were compared to the envisaged Namibian green scheme project. The emphasis is to estimate an order of magnitude and to put these different infrastructures in relation to each other rather than calculating accurate absolute numbers. Information about the green scheme was taken from the literature (Weidlich, 2007; Republic of Namibia, 2008d). Over the next 15 years the green scheme plans to create 11,750 full time equivalent jobs. This study calculated how many rainwater harvesting facilities and gardens would need to be constructed to create the same number of 11,750 jobs. Then the green scheme and the rainwater harvesting and garden facilities were compared in terms of required investment and investment per job. For comparability, only investment costs of labour and material were included

130

L. Woltersdorf et al. / Water Policy 16 (2014) 124–143

over a time span of 15 years. Assumptions and estimations are based on the pilot construction phase of the CuveWaters project. The same market and subsistence planting schemes as designed for gardens irrigated with harvested rainwater were transferred to the envisaged area of the green scheme. Owing to the lack of further data about the green scheme, such as operation and maintenance costs, a NPV calculation was not possible.

3.5. Financing of rainwater harvesting facilities Financing rainwater harvesting and garden infrastructure is the determining criterion for an up-scale of the technology. These costs were related to the household income in central-northern Namibia in order to determine the possibility of financing these infrastructures. We evaluated the possibility of financing rainwater harvesting facilities with microcredits and proposed financing possibilities.

4. Results 4.1. Benefits of rainwater harvesting for gardening Rainwater harvesting for gardening offers numerous benefits to local communities in technological, environmental, social and economic terms (Table 2). While rainwater harvesting-based gardening broadly effects and stimulated the regional economy, livelihood benefits extend far beyond material gain. Therefore the technology has the potential to become an important part of Namibia’s water infrastructure.

During the construction of the pilot tanks, a team of tank technicians, known in the region as the ‘Blue Team’, have been enabled to build new tanks and operate and maintain existing tanks, plan budgets, calculate costs and procure construction materials. The technicians proved their skills in constructing a privately financed tank in the absence of the CuveWaters staff. Tank users were trained in proper tank operation and maintenance, gardening and irrigation techniques. Trained technicians, tank users and farmers were highly committed owing to community involvement from the very beginning. The pilot rainwater harvesting facility soon became locally known as the ‘Epyeshona Green Village’ and represents a local success story receiving considerable attention from the media and people from surrounding villages.

Nevertheless the implementation of small-scale water infrastructure is associated with certain risks and challenges. In rural parts of Namibia, like in most other parts of rural Africa, the low level of education makes it extremely difficult to implement the necessary structures to run gardening ventures that aim to supply markets in the region. Training and education is also essential to counter the lack of knowledge of horticultural production in the region. An additional challenge for planning gardens irrigated with rainwater harvesting is the high rainfall variability in the region (UNEP, 2006).

4.2. Financial cost–benefit analysis 4.2.1. Cost. The costs of pilot rainwater harvesting facilities and estimated costs without project conditions are shown in Table 3.

L. Woltersdorf et al. / Water Policy 16 (2014) 124–143

131

Table 2. Benefits of rainwater harvesting.

Area

Benefits of rainwater harvesting and gardening

Source

Technology

•

Li et al. (2000)

• Economy

• • • • • • • • • •

Environment

• • • • • •

Social

• • • • • • •

Maintenance is easy, therefore the technology is also appropriate for remote rural areas Local water resources are used instead of inter-basin water transfer Broad spill-over effects for the regional economy (e.g. knowledge extension for rainwater harvesting and gardening) Job creation and income generation in poor rural and peri-urban communities through Tank and garden construction, maintenance Local market sale of crops Education of tank builders, gardeners, etc. improves own (career) prospective for future life Productive use of rainwater Higher crop yields Extended annual planting season of crops through irrigation into the dry season Possible to plant crops with a longer growth period and higher water requirements (i.e. tomatoes, cabbages) Additional annual harvest during the dry season achieves higher revenues Adaptation strategy to climate change and climatic variability Effective use of heavy rainfall events Bridge the dry season Higher crop growth security by bridging rainfall variations and dry periods during the rainy season Provides additional water supply reducing pressure and demand on surrounding surface water Contributes to the regeneration of landscapes by increasing biomass for food, fodder, fibre and wood for human consumption Improved food-security and availability particularly during the dry season Increased household and community self-sufficiency Improvement of living conditions for vulnerable or marginalised groups through a better diet and the possibility to engage in a productive activity Time saved for productive activities through availability of water near the house Improvement of children’s education and health conditions due to additional income Enables communities to adapt to droughts and declining availability of drinking water Creation of knowledge and capacity building

Agarwal et al. (2001); Senkondo et al. (2004); Yuan et al. (2003); Rockström et al. (2002)

Pandey et al. (2003); Barron (2009); Rockström et al. (2002); UNEP (2006); Ngigi et al. (2007); Jianbing et al. (2010); van Steenbergen & Tuinhof (2009); Barron (2009); Li et al. (2000); Machiwal et al. (2004); Barron (2009)

Wakefield et al. (2007); Wills et al. (2010); Swanwick (2009); van Averbeke (2007)

132

L. Woltersdorf et al. / Water Policy 16 (2014) 124–143

Table 3. Costs of rainwater harvesting and gardening facilities in central-northern Namibia.

Type of facility

Material (N$)a

Ferrocement tank (30 m3)

School (under specific project conditions) Household (under specific project conditions) Without specific project conditions, estimated down (see below) Market: 1 crop cycle/yr (52 m2) Subsistence: 2 crop cycles/yr (84 m2) Community Estimated down

Garden

Pond (80 m3)

Garden

Market: 1 crop cycle/yr (229 m2) Subsistence: 2 crop cycles/yr (148 m2)

13,592 Ferrocement, gutters, pipes 18,571

Labour construction (N$)b

Operation and maintenance per year (N$/yr)

5,500

100

none

200 (material) 560 (seeds, pesticides, fertilizer) 200 (material) 100 (seeds, pesticides)c 155

12,000

2,572 3,320

Fence, drip irrigation, pedal pump, tools, (shade net)

48,766 Timber, dam liner, 35,000 Corrugated iron sheet, gutters, pipes 6,615 Fence, drip irrigation, pedal pump, tools, (shade net) 4,808

8,100

none

400 (material) 1,550 (seeds, pesticides, fertilizer) 400 (material) 300 (seeds, pesticides,***)

Currency exchange rate: 1 N$ ¼ € 0.07625 (oanda.com, on 17 September 2013).

Labour costs are calculated with Namibian union labour tariffs of 100 N$/day. c Subsistence farmers are assumed to use goat manure as fertilizer. a

b

The major cost component of a rainwater harvesting and garden facility are the material costs of the rainwater harvesting facility. The market garden contains a shade net, while the subsistence garden does not, as it contains fruit trees for shade. Costs for garden material and operation and maintenance costs are relatively low, for example because only pedal pumps are used for pumping the stored water into the irrigation system. It has to be considered that during the pilot construction phase material costs were extraordinary high, as the project was forced to build during a specific and limited timeframe before holidays and during the rainy season. During this period market prices are higher and, because Oshanas were flooded, the sand had to be purchased. Therefore prices are not transferable to ‘without project’ conditions and costs are expected to decrease down to an estimated 12,000 N$ for the ferrocement tank and 35,000 N$ for the pond if construction takes place at a greater scale without project conditions (i.e. built by locals in the dry season with optimised material use). In addition, government bulk purchase of important raw materials such as wood, steel and cement might drop current monopoly prices considerably.

4.2.2. Benefit. The ferrocement tank achieves annual revenues from gardening of 5,053 N$ (457 €) (337 kg tomatoes) in the market garden variant and 1,143 N$ (103 €) (548 kg of fruit and vegetable) in the subsistence garden variant (Table 4). The pond achieves annual revenues from gardening of

Table 4. Revenues per year of gardening products from rainwater harvesting with different garden variants.

Crop type

Planting datea

Harvesting datea

Cultivated area (m2)

Priceb (N$/kg)

Gross irrigation requirementa (m)

5.0

1.07

5.0

86

91

4.0 3.1 1.4 4.0 5.2 23.0

5.47 0.96 6.13 2.91 1.54 0.92

4.0

15.0c

10.1 12.9 10.8 13.8 8.6 25.5 86.7 63.9

68 53 24 68 88 161 548 337

372 51 146 198 136 149 1,143 5,053

5.0

1.07

14.1

243

259

4.0 3.1 1.4 4.0 5.2 23.0

5.47 0.96 6.13 2.91 1.54 0.92

4.0

15.0c

28.5 36.5 30.6 39.2 24.4 1.5 174.9 174.1

193 151 67 193 249 345 1,442 918

1,055 145 413 562 385 318 3,138 13,774

Production (kg)

Revenue (N$)

L. Woltersdorf et al. / Water Policy 16 (2014) 124–143

Ferrocement household tank (30 m3) with roof catchment (100 m2) Subsistence Water 1 Jan 21 Mar 17 (worst case) melon Cucumber 1 Jul 13 Oct 17 Cabbage 1 Jan 14 Jun 17 Pepper 1 Dec 30 Mar 17 Tomato 15 Apr 2 Sep 17 Potato 1 Apr 24 Jul 17 Oranged 1 Jan 31 Dec 1 Sum/year 52 Market (best Tomatoes 1 Jan 4 Jun 84 case) Pond (80 m3) with roof catchment (200 m2) Subsistence Water 1 Jan 21 Mar 48.2 (worst case) melon Cucumber 1 Jul 13 Oct 48.2 Cabbage 1 Jan 14 Jun 48.2 Pepper 1 Dec 30 Mar 48.2 Tomato 15 Apr 2 Sep 48.2 Potato 1 Apr 24 Jul 48.2 Oranged 1 Jan 31 Dec 3 Sum/year 146 Market (best Tomatoes 1 Jan 4 Jun 229 case)

Local yield per areab(kg/m2)

a The planting and harvesting date has been determined based on Savva & Frenken (2002) with the growth season coinciding with the rainy season. The gross irrigation requirement has been calculated with Cropwat 8.0 based on local climate data from Ondangwa station; data: Namibian Weather Bureau and crop data for semi-arid regions Savva & Frenken (2002). The area is fitted with probability of tank failure determined to occur in 3 out of 4 years (Woltersdorf et al., 2013). b Data: Price Waterhouse Coopers (2005). c Data: project monitoring of market price in Epyeshona village in 2010. d The orange fruit tree is assumed to occupy 1 m2 on the ground and 6 m2 at the treetop.

133

134

L. Woltersdorf et al. / Water Policy 16 (2014) 124–143

13,774 N$ (1,247 €) (918 kg tomatoes) in the market garden variant and 3,138 N$ (284 €) (1,442 kg of fruit and vegetable) in the subsistence garden variant.

4.2.3. Cost–benefit. The NPV of both rainwater harvesting facilities is negative when assuming subsistence garden production and integrating the material costs of the rainwater harvesting facility (Table 5). Assuming a market garden production, both facilities have a positive NPV: the ferrocement tank of þ46,943 N$ (4,248 €) and the pond of þ95,711 N$ (8,662 €). Subsistence garden production can also have a positive NPV when excluding the material costs of the rainwater harvesting facility, while still including labour costs for rainwater harvesting facility construction, operation and maintenance (O&M) costs and garden material costs. In this case, the ferrocement tank has a NPV of þ6,997 N$ (633 €) and the pond of þ25,578 N$ (2,315 €). Further research results for the CuveWaters project clearly show that in remote villages of central-northern Namibia (e.g. more than 65 km distance from the pipeline scheme) the construction of an adequate number of rainwater harvesting tanks can be considerably cheaper than a connection to the pipeline scheme ( Jokisch et al., 2011).

4.3. First results from monitoring Pilot rainwater harvesting tanks and gardens were built in the village of Epyeshona in 2010; a drip irrigation infrastructure was added in 2011. The first harvest in 2010 included butternut, spinach and different varieties of pepper. Gardening products served household consumption and achieved good prices on local markets contributing to household income generation. Since February 2011, the farmers monitor the amount of harvest, income, amount of fertilizers and pesticides applied on the fields. The most popular crop so far is spinach, mainly because it can cope well with the poor soil conditions and grows fast. In 2012, individual household farmers earned up to 900 N$ per month from the sale of spinach. In the greenhouse, tomatoes performed best, as they can be harvested over a long period and generate the highest income. On local markets these tomatoes achieved a mean price of 13 N$/kg compared to 2.91 N$/kg indicated by Price Waterhouse Coopers (2005). In 2012 the farmers focused mainly on spinach and tomatoes based on their experiences in 2011 and stabilised their income from the individual gardens at around 900 N$ per month from the sale of spinach and tomatoes, but in parallel also produced certain other crops for their own consumption thus improving their own diet and health situation. Furthermore the daily water use decreased as a consequence of more experience and knowledge gained in 2011. So far Table 5. Cost–benefit analysis of two rainwater harvesting options in combination with gardening assuming best case costs (estimated in case of large-scale production), with discount rate of 5%.

NPV (N$) Rainwater harvesting option Ferrocement tank (30 m3), lifespan 40 years Pond (80 m3), lifespan 20 years

Including: material investment costs, labour construction costs, O&M costs

Including: labour construction costs, O&M costs, excluding: material investment cost

Subsistence (52 m2) Market (84 m2)

10,503 þ46,943

þ6,997 þ64,443

Subsistence (146 m2) Market (229 m2)

17,521 þ95,711

þ25,579 þ138,811

Garden variant

L. Woltersdorf et al. / Water Policy 16 (2014) 124–143

135

monitoring shows that revenues used in the worst case garden variant are (partly) underestimated, as real income is considerably higher than expected. This is mainly due to higher prices on local markets in central-northern Namibia compared to wholesale prices in the capital Windhoek, used in the Price Waterhouse Coopers (2005) study. Nonetheless, observed water use was higher than calculated and fluctuated considerably over the course of the season, mainly owing to the little experience of the users.

4.4. Comparison of rainwater harvesting facilities to existing water and irrigation infrastructure in Namibia In 2003 the Namibian government adopted its green scheme policy through the Ministry of Agriculture, Water and Forestry (MAWF). The green scheme’s objectives are to increase commercial large-scale irrigated crop production, import substitution, self-sufficiency, food security at national and household level and create jobs (Weidlich, 2007; MAWF, 2008; Republic of Namibia, 2008d).

The scheme aims to put an area of approximately 27,000 hectares under irrigation over a period of 15 years (Republic of Namibia, 2008d; Allgemeine Zeitung Namibia, 2011). Irrigated crops include maize, wheat, pearl millet (mahangu) and vegetables mainly along perennial rivers at Namibia’s borders (Weidlich, 2007). For project realisation, over the next 15 years the Namibian government aims to invest 3,311 million N$ and 7,430 million N$ are expected to be contributed by the private sector (Weidlich, 2007; Allgemeine Zeitung Namibia, 2007). According to government estimates the green scheme could create 10,000 permanent and 3,500 seasonal jobs. However funding is a permanent constraint and the major reason for slow progress (Weidlich, 2007). In 2007, Namibia had 9,000 ha under irrigation, including 3,000 ha under the green scheme (Weidlich, 2007). Table 6 puts the proposed rainwater harvesting technology in relation to the Namibian green scheme in terms of total investment, created jobs, irrigated area, estimated value of garden produce and investment per job.

We estimated the amount of rainwater harvesting facilities and gardens that can be constructed and cultivated when creating the same number of 11,750 full time equivalent jobs as planned under the

Table 6. Comparison of gardening with rainwater harvesting and the Namibian green scheme (over 15 years’ investment), assuming the creation of 11,750 full-time equivalent jobs per technology option.

Green scheme

Ferrocement rainwater harvesting tank

Rainwater harvesting pond

Amount of RWH facilities and gardens that can be constructed, creating circa 13,500 jobs over 15 years Total investment (million N$)

–

21,875

22,500

10,741 million N$

747 million N$ (626 million N$ material costs, 120 million N$ labour costs)

Irrigated area (ha) Estimated value of horticulture produce assuming same crop schemes (million N$/year) Investment per job (N$/job)

27,000 ha 3,150 million N$ to 5,264 million. N$ 914,145 N$/job

114 to 184 ha 25 million N$ to 111 million N$

1,119 million. N$ (936 million N$ material costs, 182 million. N$ labour costs) 328 to 515 ha 71 million N$ to 310 million N$

63,775 N$/job

95,312 N$/job

136

L. Woltersdorf et al. / Water Policy 16 (2014) 124–143

green scheme. We estimated that the construction of one ferrocement tank requires a team of one skilled and ten unskilled workers and takes 12 days. Calculating with 250 annual working days it finds that 770 tank builders (70 teams) can construct 21,875 ferrocement rainwater harvesting tanks over a time span of 15 years. A further 10,938 jobs are created in the gardening sector, assuming that workload and income from the gardening with the water from one ferrocement tank is equivalent to one half day job. When creating 11,736 new jobs, 22,500 ponds and gardens can be constructed and cultivated. For this, 486 pond builders (54 teams) with one skilled and eight unskilled workers per team take nine days to construct 1 pond, resulting in 1,500 ponds per year and 22,500 ponds over a period of 15 years. Further 11,250 jobs are created to cultivate gardens, assuming that the workload and income from the gardening with the water from one pond is equivalent to a half day job.

The construction of 21,875 ferrocement rainwater harvesting tanks requires an investment of 747 million N$ resulting in an investment per job of 63,775 N$. The construction of 22,500 rainwater harvesting ponds requires an investment of 1,119 million N$, resulting in an investment per job of 95,312 N$. The income generated with this number of rainwater harvesting tanks and gardens through the sale of horticulture products is 25–111 million N$ per year. These ponds and gardens generate an annual income of 71–310 million N$ (subsistence and market garden variant, respectively). In contrast, the green scheme is expected to require a considerably higher investment of 10,741 Mio N$ creating the same number of jobs but requiring a significantly higher investment per job of 914,145 N$. Assuming the same crop schemes for the green scheme would result in an annually generated income of 3,150– 5,264 million N$. However it has to be considered that in reality the green scheme is also producing maize and wheat so that the generated income will be considerably lower than estimated here.

In central-northern Namibia, investment costs for water infrastructure are extraordinary high, owing to the large water supply network (Heyns, 1995). In central-northern Namibia the sales price of grid water is currently around 8.3 N$/m3, but this price is heavily subsidised. In contrast, we estimate that the full cost recovery price including infrastructure investment costs is between 10 and 15 N$/m3. In comparison to this, the full cost recovery price of our proposed rainwater harvesting infrastructure (ferrocement rainwater harvesting tank) is 15 N$/m3. Therefore, the costs per square meter of harvested rainwater are not higher than the cost of grid water.

4.5. Financing of rainwater harvesting facilities Private financing of initial investment costs represents a problem for most micro-entrepreneurs and is the major limiting factor for the up-scaling of rainwater harvesting. Average annual household income in central-northern Namibia ranges from 26,788 N$ in the Oshikoto region to 45,708 N$ in the Oshana region (Republic of Namibia, 2006). In relation to this income level, tanks and gardens have high investment costs, while the maintenance costs of rainwater harvesting facilities are very low. Therefore, we propose other sources to finance infrastructure material investment costs.

The results of this study indicate that microcredits are not suitable to finance material costs of rainwater harvesting facilities. On the one hand, traditional microcredit loans are usually too small with too short repayment periods (up to 2 years) and are not compatible with the necessary medium- to long-term investment of over 6 years for investment sums of over 12,000 N$. On the other hand repayments for annual interest rates (currently 24–35% p.a. in Namibia) (Chitambo et al., 2006) exceed annual garden revenues of the subsistence garden. However, if only garden construction costs have to be financed through a microcredit, the credit can be easily repaid within a reasonable time. For instance, a

L. Woltersdorf et al. / Water Policy 16 (2014) 124–143

137

micro-entrepreneur could assume a microcredit for garden construction of 3,320 N$ (with tank) or 6,615 N$ (with pond) and then repay it within 1 or 2 years with the profits generated from gardening, having subtracted annual maintenance costs for rainwater harvesting and gardening (considering an interest rate of 24%). Annual tank and garden maintenance costs can be easily paid with annual revenues together constituting 21–52% of annual garden revenues in the case of the tank and 15–27% in the case of the pond. Based on these considerations, other sources of finance have to be identified to cover rainwater harvesting facility investment costs. Our suggestion for financing is summarised in Table 7.

5. Discussion 5.1. Proposed policy implications Owing to the inability of many micro-entrepreneurs and poor households to finance rainwater harvesting and garden infrastructure investment costs privately, other financing solutions have to be found.

International institutions such as the Organisation for Economic Co-operation and Development (OECD) and The World Bank are becoming aware of the financing issue and argue that it is unrealistic to base financial planning of water services on full cost recovery of investment costs (OECD, 2009; Banerjee et al., 2010). The OECD therefore adopted a pragmatic policy towards financing investment costs for water services by advocating the concept of sustainable cost recovery.

The concept of sustainable cost recovery entails securing and programming financial means from all sources available to the country in an appropriate combination. This includes tariffs to finance operation and maintenance costs as well as government (taxes) and donor (transfers) support to finance recurrent and investment costs. State support can be justified by the external public benefits from good water services as well as the need to make these services affordable to the poorest households (OECD, 2009).

This is also applicable for investments for agricultural water infrastructure.

Many countries wrap their subsidy element into ‘soft’ loans to utilities or local authorities, which has the advantage of preserving the incentive to make efficient use of the money. While recovering operation and maintenance costs or even investment costs from tariffs is an important economic principle in most circumstances, using tariffs to recover full costs of water services, including investment and Table 7. Financing a proposal for rainwater harvesting and gardening infrastructure.

Type of cost

Financing

Pay back

Material cost for rainwater harvesting infrastructure

Government-funded (with beneficiary contribution depending on poverty level) Micro-entrepreneur with microcredit Micro-entrepreneur with revenues from gardening

No pay back

Material cost for garden Annual maintenance cost for rainwater harvesting infrastructure and garden

Tank: 2 years (market), 22 years (subsistence)a Pond: 1 year (market), 2 years (subsistence)a O&M costs constitute: Tank: 21% (market) to 52% (subsistence) of annual revenues. Pond: 15% (market) to 27% (subsistence) of annual revenues

a Time to pay back microcredit considering the available profit after having subtracted operation and maintenance costs from annual revenues.

138

L. Woltersdorf et al. / Water Policy 16 (2014) 124–143

major rehabilitation, is unusual even in developed countries. In practice, in many countries the governments prefer to subsidise investment costs through taxation (OECD, 2009). Nonetheless, recovering the cost of providing service, at least for operation and maintenance, is a stated objective of water utilities around the world (Banerjee et al., 2010). Therefore, government could subsidise rainwater harvesting infrastructure investment costs. Beneficiary contribution of capital costs, for instance of 5–20% depending on beneficiary poverty level, could be considered in order to enhance ownership and sustainability.

Then, local tank owners and farmers can finance garden investment costs and maintenance of tanks and gardens by assuming a microcredit and repay it with market sale of gardening products. In doing so, the government could give incentives for value added production, local job creation, improvement and extension of water infrastructure and regional development. Therefore this study recommends statefunded rainwater harvesting material costs.

Besides these financial aspects, current Namibian policy is also an important precondition for the further development of rainwater harvesting-based gardening. The FAO recognises agricultural growth involving smallholders, especially women, to be most effective in reducing extreme poverty and hunger when it increases returns to labour and generates employment for the poor (FAO et al., 2012). Historically, smallholders have proved to be key players in meeting food demand. Today, smallholders face considerable challenges, such as limited accessibility to markets, credit, information and resources. Yet, smallholders are capable of meeting these challenges, although they need an appropriate enabling environment in order to do so. Providing improved rural infrastructure such as roads, markets, storage facilities and communication services will reduce transaction costs, enable farmers to reach markets, contribute to a better conservation of products and provide the possibility to add value to products by, for example, processing food. Interventions to ensure land tenure and property rights security will encourage smallholders to invest in land improvements. Provision of education in rural areas is essential if smallholders are to participate in markets (FAO et al., 2012). Currently, Namibia has an extensive policy framework to foster food security (Werner, 2011).

With regard to irrigation, however, current policies focus on large-scale commercial production and do not specifically target small-scale food producers at the local level. Therefore, despite the political intention of improving household food security, the majority of poor households in rural areas cultivating less than 20 ha does not directly benefit from current political programmes (Werner, 2011). The proposed rainwater harvesting infrastructure is explicitly not intended to replace the large-scale agriculture plans of the Namibian government. Instead, it is intended to complement it by also addressing small-scale agriculture and local market production. Thus, Namibian and international experts (e.g.

Dima et al., 2002; Werner, 2011) recommend a review of the current policy framework to provide more focus on micro- to small-scale food production (below 20 ha) and on appropriate technical support and advice in urban, peri-urban and rural areas. This policy review also needs to address institutional mandates and responsibilities in order to provide the appropriate regulation. Promotion of gardening and rainwater harvesting require a concerted campaign at all levels of government and the target population to explain the potential importance and benefits (Werner, 2011). To complement gardening activities, a working infrastructure will be necessary including extension and consulting services for gardening, plant protection and seed nurseries.

An example of financing rainwater harvesting infrastructures with grants is the South African policy ‘Financial Assistance to Resource Poor Irrigation Farmers’ of the South African Department of Water Affairs and Forestry (DWAF) published in 2004 as part of the National Water Act of 1998. South Africa and Namibia have a similar history of political and economic imbalance between different parts of the

L. Woltersdorf et al. / Water Policy 16 (2014) 124–143

139

population. The South African DWAF aims to promote social and economic development in the country through the use of water in an equitable way. It acknowledges micro-scale vegetable farming, where an estimated 150,000 farmers produce food for millions of people, to be an important sector of rural farming in South Africa. It complements the top-down managed large irrigation schemes that are one of the biggest success stories in agricultural development in the country. The act provides financial assistance for the development of irrigated agriculture by providing resource-poor farmers with grants and subsidies for water supply infrastructure and assistance for water management committees. The grant serves to construct rainwater harvesting storage tanks for resource-poor farmers in rural areas, to serve family food production and other productive uses. The grants provides annually 425,500 €, sufficient to build around 1,000 rainwater harvesting tanks per year (DWAF, 2004). Through this programme, the South African DWAF aims to contribute to the achievement of the UN Millennium Development Goals in South Africa by reducing the number of households suffering from food insecurity (DWAF, 2004).

A similar programme can be established in Namibia to reduce poverty especially in its northern regions which are also disadvantaged in terms of economic and agricultural development.

5.2. Transferability of results Our research and previous studies revealed a broad range of benefits of rainwater harvesting in technical, economic, environmental and social terms. Existing challenges can be handled by training and educating the local population. The results of the cost–benefit analysis in this study showed that rainwater harvesting is a profitable activity with a positive NPV over the lifespan of the infrastructure when planting crops that achieve high local market prices and excluding material costs for the rainwater harvesting facility, while including maintenance and operation costs and garden material. It has to be considered that the result of cost–benefit analyses depends on the choice and quality of data input and often, as in our case, only limited data (e.g. prices and yields of only 1 year from 2003) or data not specific for the model region (e.g. length of growth season for semi-arid regions) are available.

The validity of the results of our cost–benefit calculation has two aspects: the specific data used from the literature are reliable and therefore our specific results are also reliable. However, owing to inter-annual variability of, for example, market prices or agricultural yields, the literature data used from 1 year has limited significance. Therefore, the results of the cost–benefit calculation can be considered as preliminary. In the following years further field data should be collected in order to refine the cost–benefit calculation and obtain a more representative result for central-northern Namibia. In reality, the cost–benefit ratio depends on the actual lifespan of the facilities which we determined according to the extensive experience of a Kenyan rainwater harvesting consultant who constructed more than a hundred rainwater harvesting facilities in Kenya and Uganda in low-tech areas with comparable conditions to central-northern Namibia. In addition, irrigation requirements were modelled which also depend on climatic and crop data input.

First monitoring results showed that local market prices are higher than assumed (according to Price Waterhouse Coopers, 2005) in the worst garden variant and therefore revenues might be underestimated.

This indicates that the benefit might be closer to the best case garden variant with a positive cost–benefit ratio. In addition there is a high demand for agricultural products in local markets. Putting rainwater harvesting in the frame of current water and irrigation infrastructure in Namibia, the results of this study indicate that rainwater harvesting-based small-scale gardening has relative low investment costs per created job. Therefore, the invested funds in rainwater harvesting and small-scale gardens are very effective in creating new jobs.

140

L. Woltersdorf et al. / Water Policy 16 (2014) 124–143

The role of rainwater harvesting for the irrigation of small-scale gardens has not been sufficiently examined in Namibia. However, ground and roof rainwater harvesting is significant for regions with an annual precipitation of at least 300 mm (Gould & Nissen-Petersen, 2006), contrasting rainy and dry seasons and is suitable for rural as well as peri-urban areas. A constraint to an up-scaling in Namibia is the high material cost of steel mesh, cement and wood compared to the poor local income situation. In other African counties comparable rainwater harvesting tanks, that is, 30 m3 ferrocement tanks, have similar costs to Namibia. In Asia, material costs are 60–80% lower (Li et al., 2000; Agarwal et al., 2001; Cruddas, 2007; Kariuki, 2012, personal communication). The reasons are the unavailability of cement and clean graded river sand in some parts of Africa and a lack of sufficient water for construction in others. In addition, many parts of Africa have lower and seasonal rainfall and impervious roofs are smaller in number and size. In particular, compared to typical household incomes rainwater harvesting tanks are more expensive in Africa than in Asia.

Nevertheless, rainwater collection is becoming more widespread in Africa and in some parts rapid expansion has occurred in recent years, even though progress has been slower than in Southeast Asia (UNEP 2002). In Namibia, government subsidies are necessary to finance the water harvesting infrastructure. The advantage of these technologies is that they are low-tech, they can be constructed by local inhabitants themselves, they better integrate into the natural landscape and into social circumstances and necessary investments are significantly lower than for large-scale irrigation projects. The water buffering capacity of the rainwater harvesting facilities are a good adaptation to the increasing variability of precipitation caused by climate change. Therefore, rainwater harvesting for irrigating gardens has a great future.

6. Conclusion This study has shown that rainwater harvesting for the irrigation of small-scale gardens and the associated capacity development measures provide a wide range of benefits. Water harvesting and its productive use for horticulture is one key in reaching the poor in peri-urban and rural areas as the decentralised infrastructure provides them with direct access to means of production and allows them to improve their daily meals and their income in order to overcome poverty. In addition, rainwater harvesting is an effective adaptation strategy to climate change and climatic variability. Yet, the potential of rainwater harvesting in combination with gardening has not been developed in Namibia so far. To achieve broader benefits for the regional economy, investments in infrastructure and an adequate policy framework are needed. Owing to the high material costs in Namibia compared to low household incomes, subsidies are necessary to finance the water harvesting infrastructure. We recommend government funding of the rainwater harvesting infrastructure and private finance of garden and maintenance costs. The adoption of rainwater harvesting in Namibia’s water policy framework would improve water access for communities in rural areas. Then, rainwater harvesting is a valuable contribution to reach Namibia’s Vision 2030 and the Millennium Development Goals.

Acknowledgments The research project CuveWaters is funded by the German Federal Ministry for Education and Research (BMBF) and is a joint research project of the Institute for Social-Ecological Research ISOE and the University of Technology Darmstadt. The authors thank the inhabitants of the Epyeshona village for their continuous

L. Woltersdorf et al. / Water Policy 16 (2014) 124–143

141

willingness to answer our questions and for their overwhelming hospitality. We thank Piet Heyns, Alexandra Lux, Julia Röhrig, Marian Brenda and Martin Zimmermann for their comments to improve the paper.

References Agarwal, A., Narain, S. & Khurana, I. (2001). Making Water Everybody’s Business. Practice and Policy of Water Harvesting.

Centre for Science and Environment, New Delhi.

Allgemeine Zeitung Namibia (2007). Iyambo Verteidigt das Green Scheme. Allgemeine Zeitung Namibia, 03.06.2007. URL: http://www.az.com.na/politik/iyambo-verteidigt-das-green-scheme.18652.php (accessed 18 September 2013).

Allgemeine Zeitung Namibia (2011). Steuergeld gezielt einsetzen: Ökonom fordert Prioritäten bei staatlichen Großinvestitionen sowie Zielanalyse. Allgemeine Zeitung Namibia, URL: http://www.az.com.na/wirtschaft/steuergeldgezielt-einsetzen.128995.php (accessed 18 September 2013).

Atkinson, G. & Mourato, S. (2008). Environmental cost–benefit analysis. Annual Review of Environment and Resources 33, 317–344.

Banerjee, S., Foster, V., Ying, Y., Skilling, H. & Wodon, Q. (2010). Cost recovery, equity, and efficiency in water tariffs: evidence from African utilities. Policy Research Working Paper. The World Bank, Washington.

Barron, J. (2009). Rainwater Harvesting: A Lifeline for Human Well-Being. UNEP. Stockholm.

Brouwer, C., Prins, K. & Heilbloem, M. (1989). Irrigation Water Management: Irrigation Scheduling. Training manual no. 4. Food and Agricultural Organization. Rome, Italy.

Chitambo, G., Mathole, S. & Cloete, J. (2006). Study on Microfinance for Small, Medium and Micro Enterprises (SMEs) Development in Namibia. Study commissioned by the Joint Consultative Council (JCC) and executed by the Zimbabwe Association of Microfinance Institutions (ZAMFI), Joint Consultative Council, Windhoek.

Cruddas, P. (2007). An Investigation into the Potential to Reduce the Cost of Constructed Rainwater Harvesting Tanks in Uganda. Master Thesis. Cranfield University School of Applied Sciences. Cranfield.

Deffner, J., Kluge, T. & Müller, K. (2012). Pressure of urbanisation and a sustainable sanitation infrastructure: experiences with a research-driven planning method in Northern Namibia. GWF-Wasser Abwasser, International Issue 2012, p. 2–8.

Deffner, J. & Mazambani, C. (2010). Participatory empirical research on water and sanitation demand in central-Northern Namibia: a method for technology development with a user perspective. CuveWaters Papers, No. 7, Institute for Social-Ecological Research (ISOE), Frankfurt/Main, Germany.

Dima, S., Ogunmakun, A. & Nantanga, T. (2002). The Status of Urban and Peri-Urban Agriculture. Food and Agricultural Organization, Windhoek and Oshakati, Namibia.

Driessen, C. & Jokisch, A. (2010). Current Status of “Stengel Dams” in Central-Northern Namibia. Internship Report, Darmstadt University of Technology.

DWAF Department of Water Affairs and Forestry of South Africa (2004). Policy on Financial Assistance to Resource Poor Irrigation Farmers. In terms of sections 61 and 62 of the national Water Act, 1998. Pretoria, South Africa.

FAO (1992). CROPWAT: A computer program for irrigation planning and management. FAO Irrigation and Drainage Paper 46. Developed by: Martin Smith, Rome, Italy.

FAO, WFP & IFAD (2012). The State of Food Insecurity in the World 2012. Economic growth is necessary but not sufficient to accelerate reduction of hunger and malnutrition. Rome.

Gilpin, A. (2000). Environmental Economics. A critical Review. John Wiley and Sons, Chichester.

Gould, J. & Nissen-Petersen, E. (2006). Rainwater Catchment Systems for Domestic Supply. Intermediate Technology Publications. London.

Hatibu, N. & Mahoo, H. (1999). Rainwater Harvesting Technologies for Agricultural Production: A case for Dodoma, Tanzania. Animal Traction Network for Eastern and Southern Africa. Bulawayo, Zimbabwe.

Heyns, P. (1995). The Namibian Perspective on Regional Collaboration in the Joint Development of International Water Resources. Water Resource Development 11(4), 467–492.

Jianbing, Z., Changming, L. & Hongxing, Z. (2010). Cost–benefit analysis for urban rainwater harvesting in Beijing. Water International 35(2), 195–209.

Jokisch, A., Kariuki, I., Woltersdorf, L., Zimmermann, M., Brenda, M. & Polak, M. (2011). Rainwater Harvesting for Microscale Gardening within the frame of Project CuveWaters Planning, Construction and Operation of Pilot Plants 2009–2011.

Internal CuveWaters Working Paper.

142

L. Woltersdorf et al. / Water Policy 16 (2014) 124–143

Kariuki, I. (2012). Costs of Rainwater Harvesting Tanks in Kenya and in Other Parts of Africa. Personal communication. Oshakati, 14 January 2012.

Kluge, T., Liehr, S., Lux, A., Moser, P., Niemann, S., Umlauf, N. & Urban, W. (2008). IWRM concept for the Cuvelai Basin in northern Namibia. Physics and Chemistry of the Earth 33, 48–55.

LAWA (2005). Leitlinien zur Durchführung dynamischer Kostenvergleichsrechnungen (KVR-Leitlinien). Länderarbeitsgemeinschaft Wasser (LAWA). Kulturbuchverlag, Berlin.

Li, F., Cook, S., Geballe, G. T. & Burch, W. R. (2000). Rainwater harvesting agriculture: an integrated system for water management on rainfed land in China’s semiarid areas. an integrated system for water management rainfed land in China’s semiarid areas. Royal Swedish Academy of Science 29(8), 477–483.

Machiwal, D., Jha, M., Singh, P. K., Mahnot, S. C. & Gupta, A. (2004). Planning and design of cost-effective water harvesting structures for efficient utilization of scarce water resources in semi-arid regions of Rajasthan, India. Water Resources Management 18, 219–235.

Malesu, M., Sang, J., Odhiambo, O., Oduor, A. & Nyabenge, M. (2006). Rainwater Harvesting Innovations in Response to Water Scarcity: Runoff water harvesting in Lare, Kenya. Technical Manual. World Agroforestry Centre, Nairobi.

MAWF (2008). Green Scheme Policy. Namibian Ministry of Water Agriculture and Forestry, Windhoek.

Mwenge Kahinda, J. M., Taigbenu, A. & Boroto, J. (2007). Domestic rainwater harvesting to improve water supply in rural South Africa. Physics and Chemistry of the Earth, Parts A/B/C 32(15–18), 1050–1057.

Ngigi, S., Savenije, H. & Gichuki, F. (2007). Land use changes and hydrological impacts related to up-scaling of rainwater harvesting and management in upper Ewaso Ng’iro river basin, Kenya. Land Use Policy, 24, 129–140.

OECD (2009). Strategic Financial Planning for Water Supply and Sanitation. OECD, Paris.

Pandey, D. N., Gupta, A. & Anderson, D. (2003). Rainwater harvesting as an adaptation to climate change. Current Science 85 (1), 46–59.

Pearce, D., Atkinson, G. & Mourato, S. (2006). Cost–benefit Analysis and the Environment: Recent Developments. OECD, Paris.

Price Waterhouse Coopers (2005). Irrigation Development in Namibia: Green Scheme and Horticulture Initiative for Namibia – Cost–benefit Analysis. Price Waterhouse Coopers, Windhoek.

Price Waterhouse Coopers (2008). Horticultural Production and the Maximum Possible Import Substitution. Namibian Agronomic Board, Windhoek.

Republic of Namibia (2004). Water Resources Management Act 2004. Government Gazette of the Republic of Namibia, Windhoek.

Republic of Namibia (2006). Namibia Household Income and Expenditure Survey 2003/2004: Preliminary Report. Windhoek.

Republic of Namibia (2008a). Review of Poverty and Inequity in Namibia. Central Bureau of Statistics, Windhoek.

Republic of Namibia (2008b). Third National Development Plan 2007–2011: Volume 1. National Planning Commission Namibia, Windhoek.

Republic of Namibia (2008c). Water Supply and Sanitation Policy. Ministry of Agriculture, Water and Forestry Namibia, Windhoek.

Republic of Namibia (2008d). The Green Scheme: Policy Articulation and Project Implementation – Past, Present and Future?

Ministry of Agriculture, Water and Forestry Namibia, Windhoek.

Rockström, J., Barron, J. & Fox, P. (2002). Rainwater management for increased productivity among small-holder farmers in drought prone environments. Physics and Chemistry of the Earth, Parts A/B/C 27, 949–959.

Savva, A. & Frenken, K. (2002). Crop Water Requirements and Irrigation Scheduling. Irrigation Manual Module 11. Food and Agricultural Organization. Rome, Italy.

Senkondo, E., Msangi, A., Xavery, P., Lazaro, E. & Hatibu, N. (2004). Profitability of rainwater harvesting for agricultural production in selected semi-arid areas of Tanzania. Journal of Applied Irrigation Science 39, 65–81.

Sturm, M., Zimmermann, M., Schütz, K., Urban, W. & Hartung, H. (2009). Rainwater harvesting as an alternative water resource in rural sites in central northern Namibia. Physics and Chemistry of the Earth, Parts A/B/C 34, 776–785.

Swanwick, C. (2009). Society’s attitudes to and preferences for land and landscape. Land Use Policy, 26 S, S62–S75.

UNEP (2002). Rainwater Harvesting and Utilisation: An Environmentally Sound Approach for Sustainable Urban Water Management: An Introductory Guide for Decision-Makers. Examples of Rainwater Harvesting and Utilisation Around the World. http://www.unep.or.jp/ietc/publications/urban/urbanenv-2/9.asp (Accessed 17 June 2013).

UNEP (2006). Harvesting Rainfall a Key Climate Adaptation Opportunity for Africa. United Nations Environment Programme (UNEP). http://www.unep.org/Documents.Multilingual/Default.asp?ArticleID=5420&DocumentID=485&l=en.

L. Woltersdorf et al. / Water Policy 16 (2014) 124–143

143

UN-HABITAT & Government of Madhya Pradesh (2007). Measures for Ensuring Sustainability of Rainwater Harvesting.

Policy Paper. Nairobi. van Averbeke, W. (2007). Urban farming in the informal settlements of Atteridgeville, Pretoria, South Africa. Water SA 33, 337–342. van Steenbergen, F. & Tuinhof, A. (2009). Managing the Water Buffer for Development and Climate Change Adaptation: Groundwater Recharge, Retention, Reuse and Rainwater Storage. 3R organization, Denhaag.

Wakefield, S., Yeudall, F., Taron, C., Reynolds, J. & Skinner, A. (2007). Growing urban health: Community gardening in South-East Toronto. Health Promotion International 22, 92–101.

Ward, F. (2012). Cost–benefit and water resources policy: a survey. Water Policy 14(2), 250–280.

Weidlich, B. (2007). Green Scheme in Desperate Need of Funding. The Namibian, 13.03.2007. URL: http://www.namibian. com.na/indexx.php?archive_id=35671&page_type=archive_story_detail&page=4274 (accessed 9 September 2013).

Werner, W. (2011). Policy Framework for Small-Scale Gardening. CuveWaters Papers. CuveWaters, Frankfurt am Main.

Wills, J., Chinemana, F. & Rudolph, M. (2010). Growing or connecting? An urban food garden in Johannesburg. Health Promotion International 25, 33–41.

Woltersdorf, L. (2010). Sustainability of Rainwater Harvesting Systems Used for Gardening in the Context of Climate Change and IWRM. An example from the Cuvelai-Etosha Basin in Namibia. Master Thesis. University Frankfurt am Main. Frankfurt am Main.

Woltersdorf, L., Liehr, S. & Döll, P. (2013). Rainwater harvesting for small-holder horticulture in semi-arid regions: Design of garden variants and assessment of climate change impacts and adaptation. Paper Submitted to Water Resources Management.

Worm, J. & van Hattum, T. (2006). Rainwater harvesting for Domestic Use. Agromisa Foundation and CTA, Wageningen.

Yuan, T., Fengmin, L. & Puhai, L. (2003). Economic analysis of rainwater harvesting and irrigation methods, with an example from China. Agricultural Water Management 60, 217–226.

Zimmermann, M., Jokisch, A., Deffner, J., Brenda, M. & Urban, W. (2012). Stakeholder participation and capacity development during the implementation of rainwater harvesting pilot plants in Central Northern Namibia. Water Science & Technology 12(4), 540–548.

Received 19 April 2013; accepted in revised form 28 June 2013. Available online 14 October 2013

Water Policy 16 (2014) 144–156

Groundwater banking: opportunities and management challenges Robert G. Maliva Schlumberger Water Services, 1567 Hayley Lane, Suite 202, Fort Myers, FL 33907 E-mail: rmaliva@slb.com

Abstract Groundwater banking is the use of aquifers to store water to balance seasonal or longer-term variations in supply and demand. The large storage capacity provided by aquifers can be a valuable tool for conjunctive use of surface water and groundwater as well as other elements of integrated water resources management. Successful groundwater banking requires favorable hydrogeological conditions to efficiently recharge, store, and abstract large volumes of water. Additionally, groundwater banking is also highly dependent upon water management and operational policies to ensure that stored water is not abstracted by other users and that the water accounting system of the bank remains in balance. Accumulated credits to withdraw water should not exceed the capacity of an aquifer to safely produce the water at the design rate-of-return for the bank. System participants need to have confidence that credits issued for recharge can be safely recovered when needed. Groundwater banking systems can cause significant local adverse impacts to other aquifer users and sensitive environments during recovery periods.

Groundwater modeling is required to develop a sustainable management system that accounts for temporal and spatial variations in the impacts of both recharge and abstraction activities.

Keywords: Aquifer recharge; Groundwater banking; Integrated water resources management

1. Introduction Aquifer overdraft, whereby groundwater pumping rates exceed the aquifer recharge rate, is a growing threat to water resources and the sustainability of irrigated agriculture and potable water supplies in many areas of the world. Overdraft can be reduced or ended by increasing capture, reducing demands (water use), substitution by surface water or reclaimed water, conjunctive use of surface water and groundwater, and managed aquifer recharge (Harou & Lund, 2008). Conjunctive use of groundwater and surface water affords opportunities to optimize the use of both resources and can be an essential element of integrated water resources management (IWRM). The basic strategy is that surface water is used as the primary water supply when available, and groundwater is reserved for periods when surface water supplies are inadequate to meet demands. With respect to irrigated agriculture, groundwater doi: 10.2166/wp.2013.025 Š IWA Publishing 2014

R. G. Maliva / Water Policy 16 (2014) 144–156

145

pumping capacity reduces risks associated with reliance on surface water supplies and maximizes expected income (Bredehoeft & Young, 1983).

Groundwater banking is a type of managed aquifer recharge (Dillon, 2005) in which an aquifer is used to store water to balance seasonal or longer-term variations in supply and demand. The great storage volume available in aquifers is taken advantage of to store excess water for later use (typically surface water), which might otherwise not be put to beneficial use. Groundwater banking also includes regulatory storage-type aquifer recovery and storage (ASR) systems, in which injection of water into any aquifer confers a later right to withdraw the water (Maliva & Missimer, 2008, 2010). Groundwater banking systems typically store water in overdrawn freshwater aquifers. Implementation of groundwater banking has been concentrated to date in the semiarid and arid south-western United States (particularly parts of the states of California, Arizona, and Nevada), where there is a great need for water storage and large available alluvial aquifers. Groundwater banking, as the term is used herein, does not include ASR systems that locally store freshwater in brackish or saline aquifers through a displacement process.

The principal value of groundwater banking is that the storage of water in aquifers can help ensure regularity and continuity of water supplies and reduce both water scarcity and the costs associated with scarcity (Greydanus, 1978; Thomas, 1978; Pulido-Velazquez et al., 2004). Economic-engineering optimization modeling performed for southern California demonstrates that groundwater banking provides flexibility for better temporal regulation of flows and can facilitate water transfers that are needed to take economic advantage of conjunctive use (Pulido-Velazquez et al., 2004).

Groundwater storage banks, which involve the physical storage of water, are not the same as ‘water banks’ or ‘water transfer banks,’ which establish a procedure or process to facilitate the transfer of water between willing sellers and buyers (Miller, 2000). The overall goal of water banking is to facilitate the transfer of water from low-value to high-value uses by bringing buyers and sellers together (Frederick, 1995; Clifford et al., 2004). Water banks have assumed the roles of broker, clearing house, and market maker (Clifford et al., 2004). Brokers connect or solicit buyers and sellers to create sales. Clearing houses serve mainly as a repository for bid and offer information. Market makers attempt to ensure that there are equal buyers and sellers in the market, and they act to create and increase liquidity by ensuring trades occur even when counter parties are not available in the market.

Groundwater banking requires an accounting system to track the recharge and abstractions of stored water and may also include a market system to encourage the storage of water and make stored water available to users with the greatest needs (i.e., increase economic efficiency). Depositors would receive or earn ‘credits’ for the recharge of a given volume of water, which could later be cashed in, for water recovered from the groundwater bank. Banking systems could be designed that allow and facilitate the trading of credits. The water resources benefits of groundwater banking are compelling in terms of capturing and storing flows that may not otherwise be put to beneficial use and optimization of the overall management of water resources. However, the long-term viability of groundwater banking systems requires that the physical aspects of the system (i.e., water storage) and accounting system be congruent.

For a groundwater bank to be successful, the bank’s depositors must have assurance that water deposited in the bank will be available to them for withdrawal at a later time when needed (Sandoval-Solis et al., 2011).

The success of systems will ultimately depend upon the existence of regulatory frameworks and system-specific policies that ensure that credits for recharged water have clear, known, and protected values in terms of the ability of, and conditions placed upon, credit holders to later withdraw water.

The rights of credit holders also need to be reconciled with the rights of existing and future groundwater

146

R. G. Maliva / Water Policy 16 (2014) 144–156

users, along with environmental concerns. Groundwater banking systems thus present greater water management and policy challenges than are usually first appreciated. The banking analogy breaks down when one considers that deposits and withdrawals of money at different locations within a monetary bank have no adverse effect on the bank. This is not true for water banking systems, where excessive withdrawals at a location unsupported by the recharge of real water credits can cause significant adverse effects to other bank participants. Another significant issue is that over the long term, water credits can exceed the sustainable yield of the water bank. The objective of this paper is to address some key technical and regulatory issues that affect groundwater banking systems using the results of a theoretical groundwater modeling investigation of the impacts of system operations on water availability.

2. Groundwater storage concepts The underlying concept of groundwater banking is that an aquifer is essentially used as an underground storage tank. Recharge of a given volume of water results in a corresponding increase in the volume of water stored in an aquifer, which would be manifested by an increase in water levels (or heads). In the tank analogy, withdrawal of the same volume of water as recharged would result in a decrease in water levels back to pre-recharge (baseline) levels. However, the tank analogy is a gross simplification as aquifer water levels depend upon a number of other factors including groundwater pumping by other users, leakage, and natural recharge and discharge.

The essential feature of groundwater banking is that the recharge of water by either land application or injection wells increases the volume of water stored in an aquifer and that at least some of the water will be recoverable at a future date. Groundwater banking thus needs to be considered in the context of the water budget of an aquifer or groundwater basin, in which inputs of water minus outputs is equal to the change in storage. More specifically, RMAR þ RN þ L ¼ D þ Q þ DSV where, RMAR ¼ managed aquifer recharge, RN ¼ natural recharge, L ¼ net leakage into aquifer, D ¼ natural aquifer discharge, Q ¼ pumping abstractions, and ΔS ¼ change in storage (all units are volume). The change in the volume of water in storage is equal to the product of the change in head (Δh), aquifer area (A), and aquifer storativity (S), DSV ¼ DhAS integrated over the entire area of an aquifer.

A key point is that if there is no persistent increase in head, then there has been no increase in the volume of stored water. Recharged water may leak out of the system, be lost to discharge, or abstracted by other users who may not be participants in the groundwater banking system. Additionally, where the area of an aquifer is very large (e.g., a regional aquifer), the increase in head for a given volume of recharge becomes imperceptible.

Another fundamental aspect of the hydrology of groundwater banking systems is the difference between dynamic and static responses to recharge and recovery (abstractions). During recharge and recovery, water levels will correspondingly rise or fall in the vicinity of the wells or infiltration

R. G. Maliva / Water Policy 16 (2014) 144–156

147

basins. This dynamic change in heads quickly dissipates once recharge or recovery stops. Water levels return towards background levels. For groundwater banking systems, the key issue is the change in static water levels that persists after the dynamic response from recharge or recovery dissipates (Figure 1).

If water levels recover to pre-recharge levels (Δh ¼ 0), then no local change in storage has occurred. A consequence of the difference between the dynamic and static response is that the small or moderate rise in static water levels resulting from recharge will only partially offset the larger local dynamic drawdowns from later pumping. Where local hydrologic impacts (e.g., maintenance of spring or river flows) are a constraint on groundwater use, then the location and timing of both recharge and recovery are critical issues for the operation of groundwater banking systems.

Aquifer recharge can be performed by either injection or land application (surface spreading). For land application systems, such as infiltration basins, some applied water will not reach the water table, particularly where the water table is deep, significant lateral flow occurs in the vadose zone, and/or there is a high rate of evaporation relative to the infiltration rate. Actual aquifer recharge in land application systems, especially in arid climates, may be significantly less than the volume of water applied.

Water can also be banked by ‘in-lieu recharge,’ whereby credit is given for allocated groundwater that is not withdrawn. In terms of an aquifer water budget, the net effects of recharging a given volume of water or not withdrawing the same volume of water that would otherwise be used are similar. In-lieu recharge may be more effective than direct groundwater recharge because it does not have associated water losses, such as percolating water that does not reach the water table (Franson, 1989). In-lieu recharge may involve the substitution of surface water for groundwater or a reduction in actual water use (e.g., reduction in irrigated area or adoption of more water-efficient irrigation techniques).

Although the in-lieu recharge concept is technically sound, it can be subject to abuse if credits are issued for reductions in groundwater use that would normally have occurred anyway, and there are thus no real savings to an aquifer. Groundwater users often do not pump their full allocation each year because the water is not needed. For example, during a wetter than normal growing season, farmers will usually irrigate less. Allowing credits to be earned for reduced groundwater pumping in wet years,

Fig. 1. Conceptual diagram illustrating dynamic and static responses to injection and recovery. Large dynamic changes in water levels may occur near the sites of injection and recovery, which dissipate once injection or recovery is terminated. Injection results in net storage of water only if there is a persistent increase in heads (Δh) after the injection is terminated (track A).

Injection results in no net storage if water levels recover to static levels (track B).

148

R. G. Maliva / Water Policy 16 (2014) 144–156

which will later be used for actual groundwater pumping during drier years, will have the effect of increasing actual groundwater use. This will cause a negative impact on the aquifer water budget.

An important technical issue is the aquifer response to both managed aquifer recharge and recovery.

Where an aquifer is hydraulically connected to a surface-water body, increases in the water table elevation as the result of managed recharge may either increase the rate of discharge in gaining stream reaches or decrease the amount of induced recharge in losing stream reaches, both of which cause water to leave the basin and reduce the potential increase in aquifer storage (Purkey et al., 1998; Thomas, 2001; Contor, 2009). Managed aquifer recharge can cause stream flow to be larger than would otherwise occur. The opposite would occur during recovery, where the local lowering of the water table could increase the rate of infiltration, thereby reducing stream flow. The recharge and recovery rate of water in groundwater banking systems could also affect the rates of lateral and vertical groundwater flow into an aquifer. Therefore, a one-to-one correspondence may not occur between the volume of recharged or recovered water and the change in aquifer storage, which needs to be considered and incorporated into the groundwater bank accounting system.

3. Water accounting system The main objectives of the water bank accounting and regulatory system are to track water deposits and withdrawals and to control the amount, timing, and location of withdrawals by the participants in the system and other aquifer users. Where groundwater use is not closely regulated, the opportunity exists for unauthorized water users to free-ride and take advantage of the managed recharged water.

The water accounting system for groundwater banks serves to:

• protect the rights of bank members, and ensure that they receive benefits commensurate with their deposits;

• optimize the economic productivity of water by facilitating transfer of water credits; • ensure the sustainability of the system by preventing (or mitigating) adverse impacts associated with operation of the system.

Contor (2009) proposed that the double-entry accounting method be used, in which every transaction is recorded as both a debit entry and a credit entry in separate ledger accounts, and at all times the accounting system tracks both inventory in asset accounts and claims to inventory in ownership accounts. Deposit activities are actions that cause more water to be stored in an aquifer than would otherwise have been the case (Contor, 2009). With respect to the operation of a groundwater banking system, the recharge of a given volume of water by a system participant would be entered as a credit in the participant’s account and as a liability (i.e., water owed) in the groundwater bank’s account.

The basic requirement for a double-entry accounting, and the successful operation of a groundwater bank, is that both ledgers be balanced. The assets (i.e., rights to water) held in the participants’ accounts should be equal to (or less than) the volume of water that can be physically recovered from the system. If assets are greater than liabilities, then the groundwater bank is insolvent, although the insolvency may not be manifest until a drought occurs when multiple users attempt to withdraw water. As is the case for a financial bank, an unsound groundwater bank may continue to operate for a long time as long as

R. G. Maliva / Water Policy 16 (2014) 144–156

149

annual deposits are greater than withdrawals. The insolvency might be revealed when users try to withdraw a greater number of credits during a drought than the aquifer can safely provide.

Despite its importance, the capability of groundwater banks to provide accumulated credits has seldom been publicly addressed. Data are available for the Las Posas Basin (Southern California) groundwater bank, which provide an example of the types of issues that may arise (Maliva & Missimer, 2010). Fox Canyon Groundwater Management Agency (FCGMA), the regulatory agency that has jurisdiction over the project, noted that the historic accumulation of credits within the FCGMA has been steadily increasing, approaching 550,000 acre-feet (AF) (678 106 m3) in 2006 (Fox Canyon Groundwater Management Agency, 2007). The estimated total net credit balance in the East, West, and South Las Posas Basin at the end of calendar year 2006 was 116,002 AF (143 106 m3) compared to an annual abstraction of 27,234 AF (33.6 106 m3). The accumulated credits are over four times the annual abstraction rate. The volume of credits that are accumulating through the operation of the recharge system and in-lieu recharge greatly exceeds the amount of water that could be extracted during a short-time period (e.g., major drought).

The FCGMA (2007) noted that the use of a significant number of credits in a short period of time during a period with limited groundwater recharge represents a threat to the regional groundwater resource. It was noted that even a 5% use of the total amount of credits currently available would result in a net 24% increase in annual extraction, which could result in persistent depressions in groundwater elevations, land subsidence, and seawater intrusion (Fox Canyon Groundwater Management Agency, 2007). The US$150-million-dollar project is now recognized to be a failure that was ‘marred by insufficient research, poor judgment and hollow assurances’ (Blood & Spagat, 2013).

The accounting system can provide opportunities for increasing the efficiency of water use where the credits for stored water can be sold, purchased, or traded. For example, a party with excess surface water might use it to recharge an aquifer and later sell the accumulated credits to another party with a greater need for the water (Frederick, 1995; Contor, 2009). Groundwater systems can thus facilitate the reallocation of water to higher-value uses, provided that any regulatory obstacles to the practice can be overcome.

4. Regulatory framework Groundwater banking systems also require regulatory or other control over the groundwater basin, to enforce the design withdrawal rates of system participants and to exclude other aquifer users from initiating or increasing abstractions and thus taking the stored water. Under ideal circumstances, the groundwater bank owner or its participants are the sole groundwater users in the basin, and abstractions can be readily controlled. Otherwise, some mechanism must be in place to control the actions of the other users such as a regulatory restriction against additional abstractions. Total abstractions may be constrained during some periods by external considerations, for example, a regulatory requirement to maintain in-stream flows for ecosystem or species protection. In these situations, the decision needs to be made as to who has priority to the stored water. For example, do the rights of groundwater bank participants to stored water supersede the rights of more-senior existing groundwater users?

Purkey et al. (1998) and Thomas (2001) discuss the technical and regulatory issues associated with groundwater banking in the state of California, USA. Major regulatory and organizational challenges

150

R. G. Maliva / Water Policy 16 (2014) 144–156

may occur when multiple entities have jurisdiction over all or parts of a project. For example, in California, different entities may be involved including (Thomas, 2001):

• reservoir owner who is in charge of the storage and release system used to generate source water for groundwater banking;

• local groundwater management authority that would allow for the ‘rent’ of aquifer space for tempor• • • • •

ary storage; operator of infrastructure needed to move water from the reservoir to the groundwater bank and then to the point of use; end users who would pay for the new yield and generate revenue streams to compensate the reservoir owner and groundwater banker; regulatory agencies with jurisdiction over groundwater and surface-water use; regulatory agencies with jurisdiction over groundwater and surface-water quality; and regulatory agencies with jurisdiction over environmental-protection issues.

While many of the regulatory issues are state-specific, the general concepts are broadly applicable. A basic requirement for any water banking scheme is that some mechanism must be in place to prevent the stored water from being abstracted by other aquifer users, particularly those who are not participating in the system. The accounting and regulatory framework for groundwater banking systems can be as important as, and must rely on, a detailed knowledge of local hydrogeology in determining the longterm success of the systems. The systems need to be operated so that they provide commensurate benefits to justify construction and operation of the system.

5. Groundwater modeling The impacts of the operation of a groundwater banking system on aquifer water levels were simulated using a numerical model of a hypothetical groundwater basin using the MODFLOW code (McDonald & Harbaugh, 1988). The simulated basin has an area of 10 km by 20 km, saturated thickness of 100 m, horizontal and vertical hydraulic conductivity of 50 m/d, and specific yield of 0.1. The model cells are 100 m by 100 m, and the unconfined aquifer is simulated as being surrounded by no flow cells.

The groundwater banking system is simulated as centralized vertical alignment of ten wells with a spacing of 100 m and injection and abstractions rates of 50,000 m3/day.

A groundwater banking system operated for only seasonal recovery of water was simulated by injecting for four months, followed by two months of storage, four months of abstraction, and then two months of no activity to complete the annual cycle (Figure 2(a)). Water levels in the immediate vicinity of the groundwater banking system wells are dominated by transient dynamic responses. More distant (3 km) from the wellfield, the dynamic response becomes muted relative to the static response. Owing to the relatively low hydraulic diffusivity (transmissivity divided by storativity) of the simulated unconfined aquifer, local water levels do not completely recover back to static level after injection and abstractions during the 60-day storage or rest periods. The low hydraulic diffusivity also results in peaks and troughs in water levels occurring later and at increasing distances from the wellfield. The offset is approximately 30 days at a 3 km distance in the simulated scenario.

R. G. Maliva / Water Policy 16 (2014) 144–156

151

Fig. 2. Hydrographs from model of a hypothetical groundwater bank. (a) Simulation of equal seasonal injection and recovery.

Hydrograph from center of wellfield (solid line) reflects large local dynamic responses. Hydrograph for a well 3 km away from wellfield (dashed line) shows a much muted response to operation of the system. Water level highs and lows are offset by about 30 days. (b) Simulation of 8 years of seasonal recharge followed by recovery of all stored water over two seasons of drought.

The hydrograph from wellfield (solid line) shows a dynamic response from injection and very large local drawdowns during recovery. Distant (3 km) well hydrograph (dashed line) shows a rise in static water level during injection period and recovery back to near static levels after recovery. (c) Simulation of a groundwater banking system in an aquifer experiencing overdraft.

During the simulated 8 years of recharge, the recharged water balanced the overdraft as indicated by stable water levels in the distant (3 km) well hydrograph (dashed line). During the 2 years of simulated seasonal recovery of the recharged water, large dynamic drawdowns occur in the wellfield (solid line) and static water levels drop as a result of the previous overdraft, which was hitherto not evident.

152

R. G. Maliva / Water Policy 16 (2014) 144–156

Groundwater banking for long-term storage was simulated as 8 years of seasonal recharge, followed by seasonal abstraction at four times the injected rate for 2 years (i.e., complete recovery of the injected water).

This scenario would occur where excess water is banked during wet or normal years and then recovered at a high rate during drought periods. Seasonal recharge is simulated to continue thereafter. The hydrographs for the wellfield at the 3 km location reflect both the dynamic responses to pumping and an increase in static water levels of about 2 m (Figure 2(b)). The abstraction of the recovered water resulted in large simulated drawdowns near the wellfield and a lesser, abrupt drop in static water levels.

Aquifer overdraft was simulated by applying a negative recharge of 0.0001 m/d to the long-term storage scenario (Figure 2(c)). A key feature of the simulation results is that the net recharge from a groundwater banking system can mask the aquifer overdraft from other aquifer users. The managed recharge compensated for the overdraft and temporarily resulted in stable static water levels. However, once the abstractions start, large simulated dynamic drawdowns occur near the wellfield, and static water levels abruptly drop, revealing the effects of the historic overdraft on aquifer static water levels. The magnitude of the dynamic and static responses depends upon the volume of recharged water, recharge and abstraction rates, aquifer hydraulic properties, and aquifer area.

6. Discussion Successful operation of groundwater banking systems requires that most of the recharged water is recoverable by system participants when needed. A major threat to sustainable operation of groundwater banking systems is that accumulated credits may eventually exceed the volume of water that can be safely produced during a high-demand period. This challenge is exacerbated by the likelihood that withdrawals from the groundwater bank may be concentrated during drought periods when aquifer water levels may be relatively low and demand is high.

Local dynamic drawdowns during recovery can impact local sensitive environments and other aquifer users. Aquifer-wide increases in static water level will not offset the much larger dynamic drawdowns that may occur during local abstractions. In regional confined aquifers, recharge may not result in any material increases in local aquifer pressures that persist until the time of abstractions. For example, the San Antonio Water System (SAWS) Twin Oaks ASR creates a cone of depression in the Carrizo Aquifer during recovery that is similar to that of a purely extractive wellfield. Impacts during recovery were recognized and the Carrizo Aquifer Well Mitigation Program was implemented under an inter-local agreement even though mitigation was not required under Texas water law (Evergreen Underground Water Conservation District, 2006). The well mitigation process may include lowering of pumps, drilling of replacement wells, or connection to an existing water purveyor. Similarly, recovery of water from water banks in Kern County, Southern California, during a drought period, was reported to have adversely impacted local water users (Barringer, 2011).

Sustainable groundwater banking requires that accumulated credits be kept in line with the safe aquifer yield. The term ‘safe yield’ is referred to herein as the volume of water that can be abstracted from a groundwater banking system at a given time without causing unacceptable adverse impacts. Sustainable operation of the systems thus requires an accurate understanding of the aquifer or basin water budget through both monitoring and numerical modeling. It is critical to quantify the main water budget elements and determine how much water can safely be withdrawn during various time periods. Some parameters (RN, D, L) are difficult to measure directly and may have to be estimated through the

R. G. Maliva / Water Policy 16 (2014) 144–156

153

model calibration process. Groundwater abstractions by all users, managed recharge, and aquifer water levels need to be measured and stored (ideally in real time) in a central database.

Several solutions are available to prevent accumulated credits from exceeding the aquifer safe yield.

Where some of the recharged water is lost by leakage or other means, a discount could be applied to withdrawals (Contor, 2009, 2010). Less than 100% of the recharged water may be allowed to be recovered. Commonly, a 10% loss is assumed in California groundwater banking systems, which is subject to adjustment based on monitoring results (Thomas, 2001). Groundwater losses could be subtracted from the participants’ bank accounts proportional to the amount of water stored in each account (SandovalSolis et al., 2011).

The discount could be applied once, at the time of recharge, or there could be periodic (e.g., yearly) discounting of credits. When it is recognized that the accumulated credits exceed the safe yield of a system, all credits can be discounted (devalued) to bring the system back into balance. For example, a factor of 0.9 could be applied to all the credits to address a 10% imbalance in the system. Credits for earlier recharged water would undergo numerous devaluations over time, which creates a disincentive for long-term hoarding of water. Periodic discounting maintains incentives for new recharge intended for short- and intermediate-term use. However, when the discount applied to recharge credits is too high, then there can be little economic incentive to store water.

Credits can be given a finite life and expire after a specified time if not used. Short credit lives (e.g., 1 year) minimize the risk of an imbalance between accumulated credits and system safe yield, but reduce the value of the system to participants. An intermediate-length life (e.g., 5 to 20 years) would increase the drought-protection benefits while still reducing potential imbalances.

The amount of credits that can be recovered in a given year can be restricted to an annual safe yield volume. Inasmuch as the recoverable credits will be less than the accumulated credits, a system would be needed to allocate the recoverable credits among system participants. One option is to prorate the recoverable credits amongst all the credit holders based on the number of credits held. The drawback is that accumulated credits would have lesser value and more uncertainty as a drought-proofing tool than participants may have expected. Limiting annual abstractions would also not address the problem of a progressively increasing number of accumulated credits in the bank. From a physical perspective, if recharged water is never abstracted, it will be discharged from the banking system by natural groundwater flow. The physical system must be accurately understood and represented in a groundwater-flow model, so that administrative credit devaluation accurately represents natural credit devaluation.

Additional options are either to accept the adverse impacts as a cost of the system or to mitigate the impacts. The owner or operator of the system could be required (or volunteer) either to compensate aquifer users impacted by the banking system rules or to implement mitigation measures for the natural system. These decisions also require an accurate representation of the physical system in a groundwater-flow model. The timing and location of withdrawals also need to be considered for sustainable groundwater banking systems. Groundwater abstractions result in local aquifer drawdowns, which can have adverse impacts, such as reductions in stream and spring flows and wetland water levels, and the lowering of water levels in wells. Local adverse impacts can occur even though the system is neutral in terms of the overall aquifer water budget. For example, if maintenance of dryseason spring flows is a limiting factor, then wet-season recharge may not offset additional dryseason abstractions in the vicinity of the springs.

Contor (2009, 2010) proposed the use of surface-water–aquifer-response functions that would equalize the hydrologic values of recharge and abstraction with respect to time and location. Groundwater

154

R. G. Maliva / Water Policy 16 (2014) 144–156

modeling can be employed to assess the impacts of proposed recharge and subsequent withdrawals on surface-water bodies and the aquifer. Returning to the spring example, the credits granted for recharge would depend upon the degree to which the recharge is demonstrated through modeling to result in an increase in spring flow during the time periods of concern. Similarly, the number of credits required to abstract a given volume of water would depend upon the modeled impacts of the abstractions on spring flows during periods of concern. Depending upon local circumstances, a substantial discount may have to be applied to avoid adverse impacts if they are to occur during dry periods when surface-water bodies are most vulnerable. From an economic perspective, the use of aquifer-response functions turns water into a homogenous quantity because the withdrawal point and time no longer make a difference with respect to surface-water impacts (Contor, 2010). The lack of a link between the banking system and the limitations of the physical system can be a cause of system breakdown.

The management challenges of groundwater banking systems become even more complicated when there are numerous aquifer users and many of the users do not participate in the groundwater banking system. All aquifer users may benefit from the net recharge in terms of higher aquifer water levels, which may encourage additional groundwater use. Existing groundwater users may benefit from higher groundwater levels and thus lower pumping costs while water is stored (Davis et al., 2001).

For example, a cost-benefit analysis of the Las Vegas Valley Water District (LVVWD) artificial recharge system in southern Nevada demonstrated that the overall benefits of the artificial recharge system are greater than the costs. Operation of the system benefits all aquifer users by lowering energy costs, decreasing the need to deepen wells, lessening impacts from land subsidence, and providing additional water for the aquifer system (Katzer et al., 1998; Donovan et al., 2002). The nonmunicipal aquifer users were receiving free benefits from the system, as they were not paying towards the system operation. The solution adopted to this free-rider situation was to bill an annual groundwater management fee to well owners and groundwater permit holders in the Las Vegas basin to support the artificial recharge program and other groundwater management and protection programs.

7. Conclusions Where local hydrogeological conditions are favorable for the practice, groundwater banking can be a valuable tool for IWRM by providing a means for storing excess surface-water flows that might otherwise not be put to beneficial use. However, operation of groundwater banking systems requires careful planning in order for the systems to be sustainable in terms of being able to recover stored water when needed, while avoiding unacceptable adverse impacts to the environment, surface waters, and other aquifer users. The key challenge is developing an operational/administrative scheme that is based on detailed physical knowledge, and accurate numerical representation, of the aquifer or groundwater basin water budget. The scheme must adequately address the temporal and spatial differences in (and dynamic responses to) recharge and abstractions. Even though the operation of a groundwater bank may be neutral or beneficial in terms of the aquifer water budget (i.e., result in a net aquifer recharge), the operation of a system may still cause significant local adverse impacts. An important component of sustainable groundwater banking is an accurate groundwater model of the aquifer or basin that can be used to quantitatively evaluate the impacts of both recharge and abstractions, which are then used to refine the water accounting system of the bank. The main challenge to sustainable long-term groundwater banking systems is ensuring that credits accumulated to withdraw from the bank are accurately

R. G. Maliva / Water Policy 16 (2014) 144–156

155

discounted as necessary over time, to be consistent with the ability of the aquifer to provide the water without adverse impacts.

Acknowledgments I would like to thank Mike Geddis, Kathy Champagne-Baker, and three anonymous Water Policy reviewers for their thoughtful reviews.

References Barringer, F. (2011). Storing Water for a Dry Day Leads to Suits. New York Times, July 27, 2011.

Blood, M. R. & Spagat, E. (2013). Las Posas Basin Aquifer Failure Illustrates Risks of Underground Reservoirs. Associated Press, August 24, 2013.

Bredehoeft, J. D. & Young, R. A. (1983). Conjunctive use of groundwater and surface water for irrigated agriculture: risk aversion. Water Resources Research 19(5), 1111–1121.

Clifford, P., Landry, C. & Larsen-Hayden, A. (2004). Analysis of Water Banks in the Western States. Washington Department of Ecology. Publication No. 04–11–011, Olympia, WA. Contor, B. A. (2009). Groundwater banking in aquifers that interact with surface water: aquifer response functions and doubleentry accounting. Journal American Water Resources Association 45(6), 1465–1474.

Contor, B. A. (2010). Status of ground water banking in Idaho. Journal of Contemporary Water Research & Education 144(1), 29–36.

Davis, M. D., Lund, J. R. & Howitt, R. E. (2001). Banking groundwater to meet growing water demand in California. In: Bridging the Gap: Meeting the World’s Water and Environmental Resources Challenges, Proceedings of World Water and Environmental Resources Congress 2001, p. 13.

Dillon, P. (2005). Future management of aquifer recharge. Hydrogeology Journal 13(1), 313–316.

Donovan, D. J., Katzer, T., Brothers, K., Cole, E. & Johnson, M. (2002). Cost–benefit analysis of artificial recharge in Las Vegas Valley, Nevada. Journal Water Resources Planning and Management 128(5), 356–365.

Evergreen Underground Water Conservation District (2006). San Antonio Water System’s Twin Oaks Aquifer Storage & Recovery Program Status. PowerPoint presentation (January 16, 2006).

Fox Canyon Groundwater Management Agency (2007). Annual Report for Calendar Year 2006. Fox Canyon Groundwater Management Agency, Ventura, CA. Franson, J. W. (1989). Evaluating potential artificial recharge projects. In: Artificial Recharge: Proceedings of the International Symposium: Anaheim, California, August 22–27, 1988. Johnson, A. I. & Finlayson, D. J. (eds). American Society of Civil Engineers, Reston, VA, pp. 256–264.

Frederick, K. D. (1995). Adapting to climate impacts on the supply and demand of water. Climatic Change 37(1), 141–156.

Greydanus, H. W. (1978). Management aspects of cyclic storage of water in aquifer systems. Journal American Water Resources Association 14(2), 477–480.

Harou, J. J. & Lund, J. R. (2008). Ending groundwater overdraft in hydrologic-economic systems. Hydrogeology Journal 16(6), 1039–1055.

Katzer, T., Brothers, K., Cole, E., Donovan, D. & Johnson, M. (1998). A Cost-Benefit Analysis of Artificial Recharge in the Las Vegas Valley Ground-Water System, Clark County, Nevada. Report prepared for the Southern Nevada Water Authority Ground Water Management Program, p. 38.

Maliva, R. G. & Missimer, T. M. (2008). ASR, useful storage, and the myth of residual pressure. Ground Water 46(2), 171.

Maliva, R. G. & Missimer, T. M. (2010). Aquifer Storage and Recovery and Managed Aquifer Recharge using Wells: Planning, Hydrogeology, Design, and Operation. Schlumberger Water Services, Methods in Water Resources Evaluation Series No. 2, p. 578.

McDonald, M. G. & Harbaugh, A. W. (1988). MODFLOW. A Modular Three-Dimensional Finite-Difference Ground Water Flow Model. U.S. Geological Survey Techniques of Water-Resources Investigation Report 06-A1, p. 586.

156

R. G. Maliva / Water Policy 16 (2014) 144–156

Miller, K. A. (2000). Managing water supply variability: the use of water banks in the western United States. Drought: A Global Assessment, Volume II. Wilhite, D. A. (ed.). Routledge, London, pp. 70–86.

Pulido-Velazquez, M., Jenkins, M. W. & Lund, J. R. (2004). Economic values for conjunctive use and water banking in southern California. Water Resources Research 40, W03401.

Purkey, D. R., Thomas, G. A., Fullerton, D. K., Moench, M. & Axelrad, L. (1998). Feasibility Study of Maximal Program of Groundwater Banking. Natural Heritage Institute, San Francisco, CA, p. 77.

Sandoval-Solis, S., McKinney, D. C., Teasley, R. L. & Patino-Gomez, C. (2011). Groundwater banking in the Rio Grande Basin. Journal of Water Resource Planning and Management 137(1), 62–71.

Thomas, G. A. (2001). Designing Successful Groundwater Banking Programs in the Central Valley: Lessons from Experience.

The Natural Heritage Institute, Berkeley, CA, p. 112.

Thomas, H. E. (1978). Cyclic storage, where are you now? Ground Water 16(1), 12–17.

Received 31 January 2013; accepted in revised form 18 July 2013. Available online 30 October 2013

Water Policy 16 (2014) 157–167

A comparative study of single factor and multivariate statistical methods for surface water quality assessment Yanlong Konga, *, Zhonghe Panga, Chunyong Wub and Miao Jingc a

Key Laboratory of Engineering Geomechanics, Institute of Geology and Geophysics, Chinese Academy of Sciences, Beijing 100029, China *Corresponding author. E-mail: ylkong@mail.iggcas.ac.cn b China Nuclear Power Engineering Co. Ltd, Beijing 100840, China c Inner Mongolia Geological Engineering Co. Ltd, Hohhot 010010, China

Abstract Water quality assessment is an essential part of water resources management. Different methods, including the single factor (SF) method and multivariate analysis, employed in assessing water quality can give different results.

In this study, SF methods and multivariate analysis – including discriminant analysis (DA) and the distance function model (DFM) – were used to assess the water quality of Xinlicheng Reservoir in Northeast China, and the advantages and disadvantages for each method are demonstrated. Results from the SF method show water quality to be Class 4, while results from the DA and DFM methods show it to be Class 3, according to the Surface Water Standard in China. The concentration of total nitrogen is most polluted amongst all the evaluating factors (as a result of human activities), leading to a Class 4 result with the SF method. The alteration of an individual factor has little impact on the results of the DA and DFM methods, since they integrate the features of the all of the evaluating factors. Because anthropogenic pollution generally alters only a few evaluating factors, DA and the DFM are more useful for polluted water quality assessment than SF is. But DA and DFM methods should be used with sufficient consideration of their mathematical requirements to prevent misinterpretation of the results.

Keywords: Discriminant analysis; Distance function model; Single factor method; Water quality assessment; Xinlicheng Reservoir

1. Introduction Water quality is of serious concern to the health and state of disease of both humans and animals. Of all water’s different uses, water for drinking is the most relevant and key environmental factor for health (Doria, 2010; WHO, 2011). Assurance of drinking water safety is essential for the prevention and doi: 10.2166/wp.2013.042 © IWA Publishing 2014

158

Y. Kong et al. / Water Policy 16 (2014) 157–167

control of waterborne diseases (WHO, 2011). It is imperative to gain reliable information on water quality and to prevent drinking water from being polluted.

For effective pollution control and water quality management, it is often necessary to process a large amount of water quality data. The common methods for water quality assessment are the single factor (SF) method and multivariate statistical methods. In China, the SF method has been employed by decision-makers to assess surface water quality, as it is very simple (Yin & Xu, 2008; Zhang et al., 2011). There are various multivariate statistical techniques for water quality assessment, all of which shed light on the analyses of complex water quality databases, such as cluster analysis (CA), discriminant analysis (DA), principal component analysis (PCA)/factor analysis, source apportionment by multiple linear regression on absolute principal component scores, and the distance function model (DFM) (Dixon & Chiswell, 1996; Singh et al., 2005; Kong & Wu, 2007; Kazi et al., 2009; Koklu et al., 2010). The use of these methods has offered valuable information needed to develop appropriate strategies in effective management of water resources.

In recent years, the SF method and multivariate statistical methods have been applied to characterize surface water/groundwater quality, for the design of water quality monitoring programs, and for the identification of possible sources that influence water systems (Helena et al., 2000; Simeonov et al., 2003; Xu, 2005; Kong & Wu, 2007; Zhou et al., 2007; Kazi et al., 2009). The application of these methods has proven to be effective in extracting useful information from datasets; for instance, Zhou et al. (2007) used CA and DA to assess and reveal the complex temporal and spatial variations in the water quality of the watercourses in the Northwestern New Territories, Hong Kong. But studies on the methods themselves, including their advantages and disadvantages, are still lacking; Kong & Wu (2007), for example, developed a DFM, which is a revised CA method to assess water quality.

Although the DFM method seems to be useful, its limitations should be further examined. In this paper, we make a comparative study of three methods – SF, DA and DFM – to ascertain an acceptable usage of methods for water quality assessment. Section 2 describes the study area and data. Section 3 presents the principles of SF, DA and DFM, whilst Section 4 uses the three methods to assess the water quality of Xinlicheng Reservoir in Northeast China. Discussion on the usage of SF, DA and DFM is given in Section 5, with conclusions being given in Section 6.

2. Study area and data Data are taken from Kong & Wu (2007) who assessed the water quality of Xinlicheng Reservoir using DFM. The Xinlicheng Reservoir is located in the upper reaches of the Yitong River in Changchun City, Northeast China. The area of the reservoir is 1,970 km2 with a total storage capacity of 5.92 108 m3. The reservoir is used as a drinking water source and also for flood control.

The monthly water quality data (consisting of many parameters) were obtained in 2006 from three different monitoring sites located in the upper reaches, the centre and lower reaches of the reservoir, called Sites SU, SC and SL, respectively. Six parameters – dissolved oxygen (DO), permanganate index (PI), chemical oxygen demand (COD), ammonia-nitrogen (NHþ 4 -N), total phosphorus (TP) and total nitrogen (TN) – were chosen because they are the main components that affect the water quality in the Xinlicheng Reservoir (Kong & Wu, 2007; Yin et al., 2010). Annual averages of the monitoring results were used for the water quality assessment. The water quality data can be seen in Table 1.

Y. Kong et al. / Water Policy 16 (2014) 157–167

159

Table 1. Dataset for water quality assessment in the Xinlicheng Reservoir in 2006.

Site

DO (mg/L)

PI (mg/L)

COD (mg/L)

NHþ 4 -N (mg/L)

TP (mg/L)

TN (mg/L)

SU SC SL

10.68 10.87 10.65

4.53 4.75 4.88

17.73 17.93 18.14

0.25 0.24 0.23

0.043 0.048 0.050

1.29 1.36 1.41

SU: Upper reaches; SC: Central reservoir; SL: Lower reaches.

DO: Dissolved oxygen; PI: Permanganate index; COD: Chemical oxygen demand; NHþ 4 -N: Ammonia-nitrogen; TP: Total phosphorus; TN: Total nitrogen.

3. Methods 3.1. Single factor The SF method classifies water quality according to an individual factor, using the most polluted amongst all of the evaluating factors included in the drinking surface water standard. Therefore, it is very simple to assign a grade for a water sample using the SF method, without need of mathematical calculations.

3.2. Discriminant analysis DA is a method of analyzing dependence that is a special case of canonical correlation, and one of its objectives is to determine the significance of different variables which can allow the separation of two or more naturally occurring groups (William, 1966; Xu & Lou, 1990; Lattin et al., 2003; Zhou et al., 2007). It summarizes regularity for object classification by evaluating the data quality of several samples, establishing a discriminating formula as well as criterion, then discriminates and classifies unclassified samples. DA synthesizes each physical, chemical and biological parameter’s effect on water quality. So discriminant functions should be established to integrate all the parameters.

3.2.1. Establishing discriminant functions. Consider a set of observations (x) with known G classified groups and P factors. For each classified group, there are nj ( j ¼ 1, 2,⋯, G) samples. Given M discriminant functions (M ¼ min[G–1, p]): Yi ¼

p X

(L) v(L) i x

i ¼ 1, 2, M

(1)

L¼1

in which, x (L) is the factor L, Yi is the discriminant function i, and vi (L) (L ¼ 1, 2, … , p) is the coefficient L of discriminant function i. Taking the factor of sample k of class j into the discriminant function i, we get: (p) (p) (1) (2) (2) yijk ¼ v(1) i x jk þ vi x jk þ vi x jk

i ¼ 1, 2, , min [G 1, p], j ¼ 1, 2, , G, k ¼ 1, 2, nj

(2)

160

Y. Kong et al. / Water Policy 16 (2014) 157–167

For discriminant function i, the sum of squared deviations for samples of different classifications, Ei, is given by:

Ei ¼

G X

2 nj yij yi

(3)

j¼1

Fi, the sum of squared deviations for samples of the same classifications, is given by:

Fi ¼

nj G X X

2 yijk yij

(4)

j¼1 k¼1

It is necessary that, for yi, the value of Ei should be as large as possible. Meanwhile, Fi should be as small as possible, which means the parameter λi ¼ Ei/Fi should be as large as possible. The coefficient of yi (vi (L)) should therefore meet the equation: " # @ li @Ei @Fi ¼ Fi (L) Ei (L) =Fi2 ¼ 0 @n(L) @ni @ni

(5)

which, after some further calculation can be reduced to:

Bp li Wp Vi ¼ 0

(6)

in which: h i (p) (2) Vi ¼ n(1) n

n i i i 8 9 w11 w1p > > > > < = ....

Wp ¼ .. > > > > : ; w p1 w pp 8 9

b b > > 11 1p > > < = ....

Bp ¼ . . > > > > : ; b p1 b pp

(7)

Y. Kong et al. / Water Policy 16 (2014) 157–167

161

wst and bst are taken to represent the values at line s and column t of the matrix Wp and Bp, (1 s p; 1 t p), and are computed as: nj G X

X (s) (t) (t) x(s)

x

x x j j jk jk

wst ¼

j¼1 k¼1 G

X (s) (t) bst ¼ nj x(s) x(t) j x j x

(8)

j¼1

in which: j 1X x(s) nj k¼1 jk

n

x(s) j ¼

j G X 1X x(s) nj j¼1 k¼1 jk

n

x(s) ¼

(9)

Vi should meet the following equation: jBp li Wp Vi j ¼ 0

(10)

So the problem changes to become one of finding the solution of the eigenvalue λi and of the eigenvector Vi. On the basis of the principle that Ei/Fi should be as large as possible, we assign the corresponding eigenvector of the top M λi values in descending order to be the coefficient for the discriminant function. When the eigenvector is obtained, we can discriminate and classify unknown samples according to discriminating rules.

3.2.2. Discriminating rules. When a discriminant function is obtained, discriminating rules should be presented for classifications of the water samples. Given that there are G classified groups, the probability of each classified group is qj ( j ¼ 1, 2,⋯, G). If qs is the largest of those in qj, meaning that qs ¼ max(q1, q2, … , qG) (1 s G), the sample belongs to Class s.

Assuming a sample X ¼ (x (1), x (2), … , x ( p))0 in the P-dimensional space, a discriminant vector can be established such that Y ¼ (y1, y2, … , ym)0 . Given that Y ¼ V*X, V can be computed as: 2

(p) (2) v(1) 1 v1 :::v1

3

6 (1) (2) (p) 7 6 7 V ¼ 6 v1 v1 :::v1 7 4 ::: ::: ::: ::: 5 (2) (p) v(1) m vm :::vm

(11)

If we let pj ¼ nj/N, and use f (y/i) ( j ¼ 1, 2, … , G) to represent the conditional distribution density function, according to the Bayes principles, we get: pj f (y=j) P(j=y) ¼ PG k¼1 pk (y=k)

(j ¼ 1, 2, :::, G)

(12)

162

Y. Kong et al. / Water Policy 16 (2014) 157–167

The largest P(s/y) can then be identified as: P(s=y) ¼ max fP(j=y)g 1 j G

(13)

And thus the sample can be ascertained to be Class s.

3.3. Distance function model CA is an unsupervised pattern recognition method that divides a large group of cases into smaller groups or clusters of relatively similar cases that are dissimilar to other groups (William, 1966; Zhou et al., 2007). DFM is a revised CA method that distinguishes between samples using different distance calculating methods, such as Euclidean distance and absolute value distance (Otto, 1998; Kong & Wu, 2007). Vector distance can measure the degree of closeness between two vectors.

DFM works like a ruler which can measure the length of objects and the degree of closeness between samples.

3.3.1. Classification of evaluating indices. In general, evaluating indices/factors can be divided into benefit indices and cost indices. A benefit index is a forward index, which means the bigger the number is, the more effective it will be. By contrast, in a cost index, the smaller the number is, the more effective it will be, and so a cost index is also called a backward index.

3.3.2. Establishing a non-dimensional matrix and classified Based on the data and surface matrix.

water quality standards, an observed data matrix X ¼ xij m n and a classified standard matrix (y jt )n 5 can be calculated with five classifications of water quality. A non-dimensional observed data matrix can then be established as A ¼ (aij ) and a non-dimensional classified standard matrix as B ¼ (b jt ). All the indices should be converted to cost indices for the evaluation; and then we get: aij ¼

8 < min y jt =xij

(j is the order number of benefit index)

: xij = max y jt

(j is the order number of cost index)

8 < min y jt =y jt

(j is the order number of benefit index)

: y jt = max y jt

(j is the order number of cost index)

j

j

b jt ¼

j

j

(14)

(15)

where m is the sample’s number, n is the number of evaluating indices, i ¼ 1, 2, , m, j ¼ 1, 2, , n, and t ¼ 1, 2, , 5.

3.3.3. Distance function classification. Euclidean distance and absolute value distance methods are employed in this paper to act as a comparative analysis to DA.

Y. Kong et al. / Water Policy 16 (2014) 157–167

163

First, a 1 n Zero Matrix, C, is defined. Then the Euclidean distance from the row vectors in C to each column vector in B are calculated, and assigned as a Euclidean distance discriminant standard: vffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi uX u n Bt ¼ t (C1j b jt )2 ,

(t ¼ 1, 2, , 5; j ¼ 1, 2, , n)

(16)

j¼1

The Euclidean distance from each row vector in A to column vector in C0 (the inverse of C ) is then calculated, and is made a discriminant factor for Euclidean distance: vffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi uX u n 0 )2 , Ai ¼ t (aij C1j

(i ¼ 1, 2, m, j ¼ 1, 2, n)

(17)

j¼1

If Bt 1 , Ai Bt , sample i belongs to class t, and the larger Ai is, the worse the water quality is.

Similarly, we can use the absolute value distance to carry out water quality assessment. For the absolute value distance, we use Equation (18) to replace Equation (16) and use Equation (19) to replace Equation (17): Bt ¼

n X

C1j b jt ,

(t ¼ 1, 2, , 5; j ¼ 1, 2, , n)

(18)

(i ¼ 1, 2, m, j ¼ 1, 2, n)

(19)

j¼1

Ai ¼

n

X

0 (a

C ) ij 1j , j¼1

4. Results Water quality assessment classifies and orders the quality of water samples according to water quality standards and the values of each sample. It is an essential part of water resources management. The surface water quality standard in China (GB3838-2002) (Ministry of Environmental Protection of PRC, 2002) has been used in this paper to assess water quality in the Xinlicheng Reservoir. In this standard, there are a total of 109 standard factors, including 24 basic factors for surface water, five supplementary factors and 80 specific factors for drinking water. Based on the characteristics of surface water and water environment protection targets, the standard divides water quality into five categories: Class 1: source water and water in the national nature reserve; Class 2: the first-grade protection zone of drinking water, etc.; Class 3: the secondgrade protection zone of drinking water, etc.; Class 4: industrial water, etc.; and Class 5: agricultural water and water used for countryside and recreation areas. Here, we have chosen six factors (see Table 2) to assess the water quality of Xinlicheng Reservoir in order to compare the SF, DA and DFM methods.

4.1. SF method Using the SF method (Table 3) and based on the surface water quality standard in China, the result obtained for a water quality assessment in the Xinlicheng Reservoir is Class 4.

164

Y. Kong et al. / Water Policy 16 (2014) 157–167

Table 2. Surface water quality standard in China (GB3838–2002).

Factor

1

2

3

4

5

DO (mg/L) PI (mg/L) COD (mg/L) NHþ 4 -N (mg/L) TP (mg/L)

Saturation rate 90% (or 7.5) 2 15 0.15 0.02 0.01* 0.2

6 4 15 0.5 0.1 0.025* 0.5

5 6 20 1.0 0.2 0.05* 1.0

3 10 30 1.5 0.3 0.1* 1.5

2 15 40 2.0 0.4 0.2* 2.0

TN (mg/L)

*For lakes or reservoir.

DO: Dissolved oxygen; PI: Permanganate index; COD: Chemical oxygen demand; NHþ 4 -N: Ammonia-nitrogen; TP: Total phosphorus; TN: Total nitrogen.

Table 3. Results of the SF method for water quality assessment in the Xinlicheng Reservoir.

Site

DO

PI

COD

NHþ 4 -N

TP

TN

Classified results

SU SC SL

1 1 1

3 3 3

3 3 3

2 2 2

3 3 3

4 4 4

4 4 4

SU: Upper reaches; SC: Central reservoir; SL: Lower reaches.

DO: Dissolved oxygen; PI: Permanganate index; COD: Chemical oxygen demand; NHþ 4 -N: Ammonia-nitrogen; TP: Total phosphorus; TN: Total nitrogen.

4.2. Discriminant analysis The DA method could not be applied directly in the evaluation of water quality in the reservoir because there are no classified samples but only the classified standard. Therefore, the assumed classification should be presented before DA can be used. In this paper, we used a random data producer from MATLAB software and got 25 samples, with five samples corresponding to each individual class. A discriminant function was then established using Equations (1)–(10). The 25 assumed samples were judged based on the discrimination rules. Part of these discriminant results are presented in Table 4, which shows that the accuracy of re-classification is 100% but that the posterior probability is not very large. Using the DA method, we can then get the results of a water quality assessment in the Xinlicheng Reservoir, which shows that, for the upper reaches to the lower reaches of the reservoir, water quality belongs to Class 3 (see Table 5), which is different to the result of the SF method.

Table 4. Re-classification of assumed samples and posterior probability.

Sample order

Original class

Re-class

Posterior probability

1 2 11 12 21 22

1 1 2 2 3 3

1 1 2 2 3 3

0.5957 0.6730 0.4573 0.4638 0.4760 0.4775

Y. Kong et al. / Water Policy 16 (2014) 157–167

165

4.3. Distance function A non-dimensional data matrix was obtained using Equation (14). Then a non-dimensional classified standard matrix was obtained Using Equation (15). Using Equations (16)–(19), we can get results of a water quality assessment in the Xinlicheng reservoir, all of which belong to Class 3 (see Table 6), which is consistent with the DA results.

5. Discussion The comparative analysis shows that results using DA and DFM are consistent with each other (Tables 5 and 6) but are different to the results of SF (Table 3). Bian et al. (2011) used a Bayesian model with the average data of 2004, 2006 and 2007 and concluded the water quality in the Xinlicheng Reservoir belongs to Class 1 or 2. Although the results are not the same with DA and DFM, they show that multivariate statistical techniques classify the water quality in a better grade, as they integrate the features of all the factors.

The differences between results evaluated using the different methods should remind decision-makers to use these methods with care. A number of key points can be taken from the application of SF, DA and DFM in the Xinlicheng Reservoir, which are important for improving the methods of water quality assessment:

• First, compared with the SF method, the DA and DM methods are more useful for water quality assessment. Although the SF method is simple and feasible, it does not represent comprehensive water quality and impedes the full use of water resources. As anthropogenic pollution alters only a few evaluating factors, the SF method overestimates the importance of the altered factors. From Table 3, we can see that the high value of TN leads to the Class 4 result for the Xinlicheng Reservoir, while nitrogen is a typical contaminant from anthropogenic uses of fertilizer (Pang et al., 2013). DA and DFM give an integrated result, which can incorporate all factors and overcome the problems mentioned above.

• Second, the DA and DFM methods require appropriate factors to be selected. This is because (i) all the selected factors or evaluating indices should have an orthogonal relationship with each other; and (ii) if too few parameters are used, the results of the water quality assessment may be wrong, whilst, if there Table 5. Results of DA of water quality in the Xinlicheng Reservoir.

Sample

Classified results

Posterior probability

SU SC SL

3 3 3

0.4572 0.4488 0.4400

SU: Upper reaches; SC: Central reservoir; SL: Lower reaches.

Table 6. Results of water quality assessment using a DFM in the Xinlicheng Reservoir.

Method

SU

SC

SL

Euclidean distance Absolute value distance

3 3

3 3

3 3

SU: Upper reaches; SC: Central reservoir; SL: Lower reaches.

166

Y. Kong et al. / Water Policy 16 (2014) 157–167

are too many parameters, the calculation will become extremely complicated. Many different methods can be used to select the appropriate indices, such as the analytical hierarchy process (AHP) and PCA (Saaty, 1990; Koklu et al., 2010). PCA is one of the most powerful methods, allowing for the identification of a reduced number of latent factors with pollution sources, such as spatial and temporal sources of variation affecting quality (Kowalkowski et al. 2006; Shrestha & Kazama 2007) and exposing the important factor responsible for seasonal changes in river water quality (Ouyang et al. 2006; Koklu et al., 2010). In this case study, the selected six chemical parameters in Table 1 are common factors in water quality assessment. Of greater significance, they are dominant parameters for controlling water quality in the Xinlicheng Reservoir, whilst other parameters including oil and lead are rare (Yin et al., 2010).

• Finally, erroneous discrimination can occur whether using DA or DFM. For DA, the results are related to the number of initially assumed classified samples. However, this does not imply that the greater the number of assumed samples, the better the results. In our case, 25 initially assumed classified samples were used. But when we tried again with 50 and 250 initially assumed classified samples, the posterior probability did not always increase. The main reason is that the assumed classified samples could not give a proper coefficient of matrix discriminant function, resulting in a low posterior probability. From Tables 4 and 5, we can see that the results of the posterior probability are not ideal, although the classification results are consistent. DFM does not need initial assumed classified samples but there are still some limitations with this method. The evaluation of closeness using DFM is based on distance, which means that the water quality assessment is made by referring to a certain number or value. But sometimes certain water quality classifications cannot be determined reasonably in this way – for example, the result ‘3.2’ is closer to 3 than it is to 4 and thus, according to the rules of DFM, water quality should belong to Class 3, whilst in fact it may belong to Class 4 as the number 3.2 is larger than the number 3, which means all the values of selected indices could exceed the standard values of Class 3.

6. Conclusions A comparative study of SF, DA and DFM methods was carried out using the Xinlicheng Reservoir in Northeast China as an example. The SF method gave a result of Class 4, whilst the results of the DA and DFM methods were Class 3 for the reservoir’s water quality.

Through a comparison of the SF, DA and DFM methods, we found that: (1) the DA and DFM methods are more useful for polluted water quality assessment than the SF method; and (2) the DA and DFM methods should be used with sufficient consideration of their mathematical requirements.

Both the DA and DFM methods require the selection of appropriate factors/indices, which should be orthogonal to each other. For the DA method, it is very important to select assumed classified samples.

For the DFM method, evaluating results can show closeness between the samples and standards, but cannot always give reasonable classifying results.

Acknowledgements This study was supported by the National Natural Science Foundation of China (Grants 40672171, 41202183 and 40872162), the Innovation Program of the Chinese Academy of Sciences (Grant kzcx2-yw-127) and the China Postdoctoral Science Foundation (Grant 2013M540138).

Y. Kong et al. / Water Policy 16 (2014) 157–167

167

References Bian, J., Wang, R. & Li, X. (2011). Comprehensive assessment of Yitong River quality based on Bayesian method. Water Saving Irrigation 5, 31–33.

Dixon, W. & Chiswell, B. (1996). Review of aquatic monitoring program design. Water Research 30(9), 1935–1948.

Doria, M. F. (2010). Factors influencing public perception of drinking water quality. Water Policy 12, 1–19.

Helena, B., Pardo, R., Vega, M., Barrado, E., Fernandez, J. M. & Fernandez, L. (2000). Temporal evolution of groundwater composition in an alluvial aquifer (Pisuerga river, Spain) by principal component analysis. Water Research 34, 807–816.

Kazi, T., Arain, M., Jamali, M., Jalbani, N., Afridi, H., Sarfraz, R., Baig, J. & Shah, A. (2009). Assessment of water quality of polluted lake using multivariate statistical techniques: A case study. Ecotoxicology and Environmental Safety 72, 301–309.

Koklu, R., Sengorur, B. & Topal, B. (2010). Water quality assessment using multivariate statistical methods—a case study: Melen River system (Turkey). Water Resources Management 24, 959–978.

Kong, Y. & Wu, C. (2007). The application of distance function model used in the water quality assessment of Xinlicheng Reservoir. Water Resources and Hydropower of Northeast China 25, 61–62.

Kowalkowski, T., Zbytniewski, R., Szpejna, J. & Buszewski, B. (2006). Application of chemometrics in river water classification. Water Research 40, 744–752.

Lattin, J., Carroll, D. & Green, P. (2003). Analyzing Multivariate Data. Duxbury, New York.

Ministry of Environmental Protection of PRC (2002). Surface Water Standard in China. China Environmental Science Press, Beijing.

Otto, M. (1998). Multivariate methods. In: Analytical Chemistry. Kellner, R., Mermet, J. M., Otto, M. & Widmer, H. M. (eds).

Wiley/VCH, Weinheim, Germany, 916 pp.

Ouyang, Y., Nkedi-Kizza, P., Wu, Q., Shinde, D. & Huang, C. (2006). Assessment of seasonal variations in surface water quality. Water Research 40, 3800–3810.

Pang, Z., Yuan, L., Huang, T., Kong, Y., Liu, J. & Li, Y. (2013). Impacts of human activities on the occurrence of groundwater nitrate in an alluvial plain: a multiple isotopic tracers approach. Journal of Earth Sciences 24(1), 111–124.

Saaty, T. (1990). How to make a decision: the analytic hierarchy process. European Journal of Operational Research 48, 9–26.

Shrestha, S. & Kazama, F. (2007). Assessment of surface water quality using multivariate statistical techniques: a case study of the Fuji River Basin, Japan. Environmental Modelling & Software 22(4), 464–475.

Simeonov, V., Stratis, J. A., Samara, C., Zachariadis, G., Voutsa, D., Anthemidis, A., Sofoniou, M. & Kouimtzis, T. (2003).

Assessment of the surface water quality in northern Greece. Water Research 37, 4119–4124.

Singh, K., Malik, A. & Sinha, S. (2005). Water quality assessment and apportionment of pollution sources of Gomti river (India) using multivariate statistical techniques-a case study. Analytica Chimica Acta 538, 355–374.

William, F. (1966). An Introduction to Probability Theory and its Applications. Wiley Publishing, New York, 626 pp.

World Health Organization (2011). Guidelines for Drinking-water Quality, 4th edn. WHO, Geneva.

Xu, Z. (2005). Single factor water quality identification index for environmental quality assessment of surface water. Journal of Tongji University (Natural Science) 33, 321–325.

Xu, Z. & Lou, Y. (1990). Mathematic Geology. Peking University Press, Beijing, 323 pp.

Yin, H. & Xu, Z. (2008). Discussion on China’s single-factor water quality assessment method. Water Purification Technology 27, 1–3.

Yin, H., Chang, J., Zhang, G. & Sun, X. (2010). The reason and control measures of Xinlicheng reservoir eutrophication.

Journal of Northeast Normal University (Natural Science Edition) 42, 152–156.

Zhang, X., Xin, B., Liu, W., Guo, G., Lu, H. & Ji, Y. (2011). Comparative analysis on three evaluation methods for groundwater quality assessment. Journal of Water Resources & Water Engineering 22, 113–118.

Zhou, F., Liu, Y. & Guo, H. (2007). Application of multivariate statistical methods to water quality assessment of the watercourses in Northwestern New Territories, Hong Kong. Environmental Monitoring and Assessment 132, 1–13.

Received 7 March 2013; accepted in revised form 9 May 2013. Available online 15 October 2013

Water Policy 16 (2014) 168â&#x20AC;&#x201C;183

Energy recovery in the water industry using micro-hydropower: an opportunity to improve sustainability Aonghus McNabolaa,*, Paul Coughlanb, Lucy Corcorana, Christine Powera, A. Prysor Williamsc, Ian Harrisc, John Gallagherc and David Stylesc a

Department of Civil, Structural & Environmental Engineering, Trinity College Dublin, Ireland *Corresponding author. E-mail: amcnabol@tcd.ie b School of Business, Trinity College Dublin, Ireland c School of Environment, Natural Resources and Geography, Bangor University, Wales

Abstract The water industry as a whole consumes a considerable amount of energy in the treatment and distribution of water and wastewater. Like all sectors of society today, the industry is focusing efforts on reducing its CO2 emissions and improving the sustainability of its systems and practices. One way of achieving this is through the use of micro-hydropower (MHP) installations in water infrastructure for energy recovery purposes. This paper presents a review of energy use and CO2 emissions in the water industry as well as highlighting the opportunities and challenges for MHP energy recovery. The results indicate that significant potential exists for energy recovery in the water industry. However, many previous investigations have not considered key complexities such as variations in flows or turbine efficiency. Similarly, accurate costing and return on investment data are often absent or lacking sensitivity analysis. Further research is required to address the risks and long-term reliability of installations, alongside the development of firm policy to direct and incentivise sustainability gains in this area.

Keywords: Collaboration; Energy recovery; Environmental impact; Micro-hydropower; Sustainability; Wastewater; Water supply

1. Introduction The supply of treated water in the western world is likely to be an unsustainable process in its current form due to the considerable energy consumption and CO2 emissions inherent in the various treatment and supply processes involved. With the increasing global awareness of the impacts of energy doi: 10.2166/wp.2013.164 Š IWA Publishing 2014

A. McNabola et al. / Water Policy 16 (2014) 168–183

169

consumption and CO2 emissions on climate change, humankind, finite resources and the environment as a whole, efforts to reduce such impacts are under way in all sectors of society.

The sustainability of the water supply process and its contribution to climate change are a global concern for large urban centres ( Jenerette & Larsen, 2006). Recent studies have identified key research questions in the water industry, such as: ‘How do we develop and implement low-energy water treatment processes?’ and ‘Can we optimise water supply within catchments?’ (Brown et al., 2010).

Research has also identified the need to focus innovation in wastewater as a resource for potable water, materials and energy (Kwok et al., 2010).

The record global population has resulted in an all-time high in water demand. Transporting and treating water to meet water quality standards are expensive and energy-intensive processes. The everincreasing stringency of water quality legislation across the world has added to these costs (Berndtsson & Jinno, 2008). This, coupled with the rapidly rising costs of energy, will adversely affect the extraction and conveyance of water in the future (Zilberman et al., 2008).

Many methods of improving the sustainability of water supply have been investigated. Methods aimed at reducing overall water demand and subsequently its associated energy consumption include: the reuse of grey water; water leakage reduction and pressure management schemes; rainwater harvesting schemes; water metering and hybrid water supply systems (Berndtsson & Jinno, 2008; Rygaard et al., 2011; Ramos et al., 2011). Methods to reduce the energy consumption of individual water/wastewater treatment processes have also been investigated. These include the capture of by-products such as biogas for use in combined heat and power facilities (Hernandez-Leal et al., 2010); generation of energy from wastewater treatment through anaerobic bioreactors or microbial fuel cells (Kwok et al., 2010); recovery of waste heat; recycling of dried sludge pellets in co-firing combustion systems (Park & Jang, 2010); energy-efficient desalination systems, etc.

Using micro-hydropower (MHP) technology in water pipelines and other water infrastructure has also shown potential (Gaius-obaseki, 2010). At points of high excess pressure in water supply networks, energy may be recovered using MHP technology without interfering in the water supply service (McNabola et al., 2011). However, while significant potential is reported in the literature, only limited implementation of this concept has been put in place by the water industry.

This paper presents a review of energy consumption and CO2 emissions in the water industry. In addition, an assessment of the opportunities and challenges for the recovery of energy using MHP systems is outlined. This review and assessment incorporate the technical, economic, environmental and organisational perspectives that influence the potential of this energy recovery concept.

2. Energy use and CO2 emissions in the water industry Globally, 2–3% of energy usage is reported to be associated with the production, distribution and treatment of water (Kwok et al., 2010). In the United States, it is estimated that 5% of national energy consumption is associated with water services. At city level, 30–60% of local government expenditure has been reported to be associated with water services, where the energy consumption requirement thereof is the single largest expense within budgets. Energy prices are rising and their effects on the cost of water supply have been highlighted in the literature (Zilberman et al., 2008).

The water industry is the fourth most energy-intensive industry in the United Kingdom, responsible for 5 million tonnes of CO2 emissions annually and consuming 7.9 TWh of energy in 2006–7

170

A. McNabola et al. / Water Policy 16 (2014) 168–183

(Environment Agency, 2009a, b). In Brazil, 60–80% of water industry costs are reported to be associated with the distribution of water, consuming an estimated 9.6 TWh annually at a cost of approximately US$1 billion (c.€770 million) or 14% of the annual Brazilian electricity budget (Ramos et al., 2009). In much smaller economies, such as that of Ireland, the operation of the water industry has been reported as costing over €600 million annually (Zhe et al., 2010).

In developed countries, the distribution of water typically accounts for 45% of energy use in the industry (Kwok et al., 2010). Indeed, the pumping of water in California is reported to be the largest single use of electricity in the state (Lofman et al., 2002). Water is heavy and its transport over long distances against large rises in elevation is expensive and energy-intensive. The remaining portion of energy consumption in the water industry is from wastewater management (29%) and water treatment (26%). It has been estimated that 0.8 kWh of energy is required per cubic metre of wastewater treated in Norway, twice the amount of energy required to supply the same volume of drinking water (Venkatesh & Brattebø, 2011).

The increasing political efforts to improve water quality across the globe have seen water service companies invest in high-tech, energy-intensive treatment facilities. Indeed, the ever-increasing stringency of, for example, the EU water quality directives has served to increase the energy consumption of the water industry over the past decade (Zakkour et al., 2002a). These rising monetary and energy costs in the water industry require intensified research efforts to improve the sustainability of the process overall.

As outlined earlier, the water industry is increasingly exploring the use of MHP as a means of energy recovery. The best available estimate of the hydropower potential in the UK water industry, for example, is 17 MW (Zakkour et al., 2002b), with a capacity of 9 MW installed at present (Howe, 2009). The following subsections outline reported research findings of energy recovery using MHP.

3. Energy recovery in the water industry 3.1. MHP systems MHP systems comprise a means of converting the energy of flowing fluid into mechanical and, subsequently, electrical energy on a small scale (,100–300 kW). These systems may be suitable for providing energy for a typical house or small community depending on the magnitude of the fluid resources available. As such, MHP could be considered as a form of decentralised hydroelectric energy conversion. The energy available from a particular MHP installation is a function of the fluid flow rate and available head at that particular site. It is also a function of the efficiency at which the available energy resource may be converted to electrical energy, commonly in the range of 50% to 75% depending on turbine type and flow/head conditions. The costs of MHP installation are reported to be in the region of €3,000–6,000 per kW (Gaius-obaseki, 2010).

These systems have been installed in numerous locations across the globe and have been particularly popular in developing countries; there are tens of thousands of such installations in countries like China, Nepal, Sri Lanka and other east Asian and African countries (Khennas & Barnett, 2000). In recent years, MHP installations in western countries have also become more widespread. Sites in which it is technically and economically viable to produce hydropower on a large scale have become increasingly scarce. In addition, the reduced environmental impact of MHP and the lower associated costs

A. McNabola et al. / Water Policy 16 (2014) 168–183

171

(which made it popular in developing countries) have increased its attractiveness (Da Silva et al., 2011).

Furthermore, the international focus on energy sustainability and climate change has been a driver in this activity.

3.2. Origins of MHP energy recovery in the water industry Some of the earliest published records of research in this area were carried out by Williams (1996), who identified the scope for MHP use as a form of energy recovery in water pipelines. It was identified that there are many instances in water supply networks where control valves are in place to manage downstream pressure. Here, the installation of an MHP turbine could achieve the same reduction in pressure required while simultaneously recovering some of the available energy (Williams et al., 1998). Thus, energy could be generated for use by the water service provider without reducing the level of service to consumers, thereby reducing the cost of water production and supply.

This hypothesis was tested at a control valve, acting as a pressure management control in a water supply system in the UK. The control valve was located in the vicinity of an isolated new chemical dosing plant that required 4 kW of power for operation. The available resource of 50 l/s and 36 m of head was sufficient to recover an estimated 17 kW of power. As this exceeded the local energy demand, an energy recovery MHP scheme was constructed whereby only a proportion of the flow was diverted through the turbine, sufficient to generate the 4 kW required by the chemical dosing plant. With the total investment in this energy recovery infrastructure costing €44,000 and also saving the expense of a connection to the grid (estimated at €62,000), the scheme was an obvious success (Williams et al., 1998).

Other similar installations are known to pre-date this example, with energy recovery installations in Germany (Mikus, 1984) and Scotland (Williams, 1996). In Vartry reservoir, Ireland, an MHP system was installed in 1947 to recover energy from flow between an upper storage reservoir and the treatment plant below. The MHP plant was later decommissioned and the belt-driven Pelton turbine was recently upgraded to a 90 kW plant, generating sufficient energy to operate the works and sell the excess to the grid (Figure 1). Such cases are commonplace at older water storage and treatment facilities, whereby older MHP turbine technology was sufficient to meet the power demands of treatment facilities in the earlier parts of the 20th century. However, as water treatment regulations became more stringent and hence more energy-intensive, many of these hydropower installations were no longer fit for purpose and fell into disrepair. With later advances in turbine technology and overall plant efficiencies, as well as the introduction of renewable energy Feed-in Tariff (FIT) schemes to incentivise MHP production, many schemes have since been refurbished.

Clearly, a precedent for the recovery of energy in water supply systems using MHP has existed long before international pressures on renewable energy, sustainability and climate change arose. However, in response to these pressures, renewed focus on this concept is emerging in the literature. Furthermore, of those hydropower installations in the water industry today, the vast majority could be described as the ‘low-hanging fruit’ of the total pool of potential resources available. It is likely that many suitable sites for energy recovery exist within the water infrastructure but that their potential remains untapped.

3.3. Energy recovery in water supply and wastewater infrastructure Investigations examining the recovery of energy in water supply and wastewater systems have included studies on pressure reducing valves (PRVs), control valves, break pressure tanks (BPTs),

172

A. McNabola et al. / Water Policy 16 (2014) 168â&#x20AC;&#x201C;183

Fig. 1. (a) 90 kW MHP energy recovery system (1947) Vartry Reservoir, Ireland; (b) 90 kW MHP energy recovery system upgrade, Vartry Reservoir, Ireland.

Fig. 2. Locations with energy recovery potential in the water and wastewater infrastructure.

storage/service reservoirs and wastewater treatment plants (Gaius-obaseki, 2010). These locations for energy recovery can be seen in Figure 2 and information regarding these locations are discussed in the subsequent subsections and relevant case studies summarised in Table 1.

A. McNabola et al. / Water Policy 16 (2014) 168–183

173

Table 1. Research findings from literature dealing with energy recovery in water and wastewater infrastructure.

Research studies Location

No. of sites

Water supply infrastructure Storage/service 1 reservoirs 3

Location

Potential kW energy recovery (mean, range)

Portugal

260, –

Ireland

67, 12–115

Other information 120 m3/day mean flow, €113,800 per annum (Ramos et al., 2009) Potentially generating €144,000 annually from electricity (McNabola et al., 2011)

Break pressure tanks (BPTs)

7

Ireland

12, 2–27

Economic feasibility omitted long-term uncertainties of flow and energy output (McNabola et al., 2011)

Pressure reducing valves (PRVs)

6

US

83, –

23

Brazil

10, 2.6–40

1

Italy

9.5, –

–

Canada

–

30

Ireland

8.5, 0.1–47

Produced 5–15 kW of predicted 35 kW (Rentricity, 2007) 50–110 mm pipes, efficiency of 90% overestimation, flow and pressure variations omitted (Da Silva et al, 2011) Low installation costs and no long-term sensitivity analysis (Giugni et al., 2009) Feasibility study identified uncertainties: long-term growth, diurnal and seasonal demand variations, pipe frictions and future costs. Probabilistic framework suggested (Colombo & Kleiner, 2011) Using annual average flow overestimated potential (Corcoran et al., 2012)

Switzerland

210, –

India

190, –

1

Australia

1,370, –

1

UK

180, –

1

UK

177, –

1

Ireland

133, –

Wastewater infrastructure Wastewater 1 treatment plants 1

Demonstration project using PAT from sewage outfall (Williams et al., 1998) Economically viable turbine installation at sewerage storage plant on university campus (Saket, 2008) 330 Ml/day and 60 m head to deep-sea outfall and annually generates 12 GWh of electricity and offsets 80,000 tCO2e (EcoGeneration, 2008) Two parallel Archimedes screw turbines, saving €160,000 in annual electricity costs (Engineering & Technology, 2010) 3.3 km sea outfall and 40 m head, 149–193 kW range due to turbine efficiency and tidal flows. €560,000 cost, €11,000 in annual generation, reliable system bypass necessary (Griffin, 2000) One feasible site at largest treatment works in Ireland (Power et al., 2012)

174

A. McNabola et al. / Water Policy 16 (2014) 168–183

3.3.1. Pressure reducing valves. PRVs aim to reduce the pressure of flow passing through them to a preset level. Their use in the water industry has become widespread in response to drives to reduce leakage losses in the system through pressure management. The installation of PRVs also prevents excesses in downstream hydraulic grades (Fontana et al., 2012). PRVs are a more versatile solution to pressure control than their predecessor, the BPT. Owing to their size, likely higher costs and increased risks of water contamination, the BPT has become a less popular design solution. In addition, PRVs offer the additional functionality of reducing pressure to a range of values as opposed to a single value in the case of a BPT. Replacing a PRV with an MHP turbine has been shown in the literature to be a feasible mechanism for both reducing pressure and recovering useful energy in certain circumstances. Placing an MHP turbine in parallel with a PRV and bypassing the valves allow system operators to recover energy while maintaining the integrity of the water supply system should the turbine break down (Wallace, 1996). In the USA, a commercial assessment of six PRV sites reported an estimated energy recovery potential of 500 kW. A demonstration plant was subsequently constructed at one of the sites, however, the completed MHP installation produced just 5–15 kW, lower than the initial estimation of 35 kW (Rentricity, 2007).

An investigation of the energy recovery potential of 23 PRVs in Brazil found a mean energy recovery capacity across the valves of 10 kW with a range of 2.6–40 kW (Da Silva et al., 2011). Most of these PRVs were in place on 50 mm internal diameter (ID) pipelines where the mean energy recovery capacity was typically 8 kW. One PRV in the dataset was in place on a larger 110 mm ID pipeline, which was estimated could produce over 40 kW of electricity. However, the 90% system efficiency used in this investigation could be considered an optimistic value if it is to take account of all system losses. Furthermore, this investigation and many others of this nature failed to account for the variation in flows and pressure that are likely to occur across a typical day, week, and seasonally. Thus, such estimates of energy potential do not present the full picture and may overestimate the scale of the resource.

An investigation of a PRV in a section of the water supply network in Napoli, Italy, estimated an energy recovery potential of 9.5 kW (Giugni et al., 2009). However, the estimates of cost were considerably lower than those adopted by the majority of investigations in this field. Furthermore, no consideration was given to the long-term uncertainty in any of the influencing variables of the hydro-PRV system, such as flow, pressure, energy prices, etc.

A Canadian study examined the feasibility of MHP within the water distribution network from a probabilistic perspective in order to address the issue of demand variation (Colombo & Kleiner, 2011). A number of other potential uncertainties were flagged, including long-term demand growth, diurnal and seasonal demand variations, pipe friction coefficients, and future cost fluctuations. They concluded that because demand is uncertain, a probabilistic framework should be used in calculations when deciding on the viability of a micro-turbine installation.

In Ireland, Corcoran et al. (2012) examined the potential of 30 PRVs and control valves for energy recovery purposes. The existing PRVs were found to have a mean potential for energy recovery of 8.5 kW based on average flow and head conditions, while results varied from 0.1–47 kW. Examination of the potential of the existing control valves showed higher energy potential, with a mean of 94 kW. However, it was also highlighted by the authors that the likely energy recovery potential at such water infrastructure based on yearly average flow and head data may be misleading. It was noted that flow, head and turbine efficiency vary considerably as would the subsequent power production.

3.3.2. Break pressure tanks. BPTs offer a similar functionality to that of a PRV. However, the BPT reduces pressure in a pipeline by creating a break in the system where the flow is open to the atmosphere.

A. McNabola et al. / Water Policy 16 (2014) 168–183

175

When this occurs, all the pressure that had built up in the pipeline is dispelled to the atmosphere and the continuation flow is driven by its potential energy from the break point onwards. For energy recovery purposes, an MHP turbine may be installed prior to the break point to recover energy without interfering with the level of pressure in the system downstream of the BPT. An investigation in Ireland examined the energy recovery potential of seven existing BPTs (McNabola et al., 2013). It was reported that the mean energy recovery potential was 12 kW (range 2–27 kW). Several of the BPTs examined were found to be financially viable as hydro energy recovery installations, however, again these estimates failed to address the long-term uncertainty in flow or energy-related system variables.

In addition to the earlier observations on flow variation, many investigations have also omitted the variation in turbine and system efficiency with flow rate and pressure resulting in further inaccuracies in estimations. Furthermore, studies of this nature have failed to address the long-term reliability of such energy recovery systems. MHP installation in a water supply network may be at risk of significant changes in flow and pressure conditions from the original design values. Should a new water demand arise upstream of a PRV or BPT, then flow and pressure may be significantly altered, rendering the MHP installation no longer viable.

3.3.3. Storage/service reservoirs. Storage or service reservoirs in this context comprise water storage infrastructure. Service reservoirs are typically used in a water supply system to balance the diurnal demands in a section of the distribution network while storage reservoirs are used to feed a large portion of the entire network by gravity. Service reservoirs are commonly fed by gravity but many storage reservoirs are fed via pumped mains and would be unsuitable for energy recovery in this context.

A study in Portugal of an interconnector main flowing by gravity between two reservoirs found that for a mean flow of 120 m3/day over a drop in head of 22.5 m, the annual energy production would amount to 2.28 GWh (i.e. a 260 kW plant). The value of this energy recovery was estimated at €0.05/kWh, equating to €113,800 per annum (Ramos et al., 2009). A similar investigation in Ireland, which considered a number of service reservoirs, estimated the recoverable energy at 12–115 kW depending on the particular tank in question. The reservoir with the highest energy capacity was estimated to have the potential to generate over €144,000 annually in electricity (McNabola et al., 2011).

Of the available estimates in the literature, service reservoirs have shown the highest energy recovery potential in many cases, followed by BPTs and PRVs. However, in many cases the return on investment estimation has not accounted for FIT incentivisation, available in many countries, or the savings costs of electricity to the water industry (including all taxes and charges) as opposed to the unit price.

3.3.4. Wastewater treatment plants. The flow of sewage effluent can also be directed through a penstock under pressure, through an MHP turbine to recover energy in wastewater treatment infrastructure (Gaius-obaseki, 2010). This can be carried out at treatment works outfalls or inflows, and the inlet to pumping station wet wells or in sewer mains where sufficient flow and pressure are available.

Investigators have reported on the feasibility of sewage-treatment outfalls for energy recovery using pumps as turbines (PATs). For example, a demonstration project built in 1993 in Switzerland used a PAT to produce up to 210 kW of electrical power from a sewage outfall (Williams et al., 1998). In India, a demonstration energy recovery plant has been constructed on a sewage storage tank located on a university campus. Although the plant capacity was reported as just 190 W, the project was still deemed economically viable (Saket, 2008).

176

A. McNabola et al. / Water Policy 16 (2014) 168–183

However, more notable and successful examples of energy recovery at treatment works outfalls can be seen in Sydney, Australia. Here, wastewater flow of 330 Ml/day (dry weather flow) over a 60 m drop in head at a deep-sea outfall has been used to recover energy, generating approximately 12 GWh annually. It is estimated that the plant will offset 80,000 tonnes of CO2 emissions annually as a result of this energy recovery (EcoGeneration, 2008).

In the UK, an Archimedes screw turbine has been installed to recover energy from the outfall pipe of a wastewater treatment plant. Two turbines in series are reported to produce a total of 180 kW, saving the water service provider €160,000 in annual electricity costs (Engineering & Technology, 2010).

Similarly, an energy recovery feasibility study was carried out at another wastewater treatment plant in the UK, which included a 3.3 km-long sea outfall pipe with a 40 m drop in head (Griffin, 2000).

The mean energy potential at the site was estimated to be 177 kW, within the range 149–193 kW. Energy production estimates varied with the diurnal variations in dry weather outflow from the treatment works. The effects of variations in turbine efficiency and tide levels were also included in the analysis as was the design of the plant and a suitably reliable bypass system. The investigation estimated a capital cost of €560,000 but an income from electricity generation of approximately €11,000 per annum.

In Ireland, the feasibility of MHP energy recovery at a number of wastewater treatment plant outlets was also examined. Low heads were reported for the majority of plants investigated, resulting in only those with significant daily flow demonstrating useful potential (Power et al., 2012). The largest energy recovery potential was reported as 133 kW at the country’s largest wastewater treatment facility.

No studies were found that examined the feasibility of energy recovery using MHP at the inlets to pumping station wet wells or in large sewage collector mains. Further research in this area is required to gauge the feasibility of such operations.

3.4. Variability of energy recovery potential As previously noted, the potential of energy recovery in the water infrastructure is dependent on a number of factors: in particular, the flow and pressure characteristics evident at each site (McNabola et al., 2013). As population growth has led to increased demands of the water industry, the infrastructure, that is, the locations and number of water and wastewater treatment facilities, and the continuous evolution of the distribution networks provide a significant challenge to quantify the potential energy recoverable. An example of this is the replacement of BPTs with PRVs, or BPTs becoming surplus to requirements with the optimisation of water flow and pressure characteristics in the water infrastructure.

In addition, it should be noted that the service life of the water infrastructure has also recently been highlighted as a challenge (Scholten et al., 2013), as it could affect the potential for MHP energy recovery. Water properties also vary depending on source and treatment type. Such properties (e.g. low pH, or high-pressure pipe networks (Engelhardt et al., 2000)) could affect turbines installed within some networks and hence affect energy recovery potential. However, particular issues such as ‘aggressive water’ are likely to reduce as: (i) older iron-based pipe networks are or have been replaced, which reduces corrosive particles in the networks; and (ii) most MHP sites are within the treated water distribution networks, and the latest WHO (2008) Guidelines for Drinking-water Quality report ensures that the water is of a defined quality to minimise particles.

A. McNabola et al. / Water Policy 16 (2014) 168–183

177

3.5. Hybrid systems Some of the strategies that have been investigated and proven to achieve energy savings in the water infrastructure include: water network optimisation models (Vieira & Ramos, 2009a, b), renewable energy systems used in pumped storage systems (Ramos & Ramos, 2010), and energy recovery using turbines connected to the water infrastructure (Ramos & Ramos, 2009; Carravetta et al., 2012). These strategies have also been considered as a collective or hybrid set of measures to improve the sustainability of water distribution networks (Gonçalves et al., 2011; Ramos et al., 2011), and have been investigated and implemented

by Gonçalves & Ramos (2012) and Gonçalves et al. (2011) using a neural network model. The conclusions from these studies found that a hybrid approach to achieving sustainability in the water industry has more impact than any individual energy optimisation or recovery strategy.

3.6. Economic viability The economic viability of MHP energy recovery systems in the water industry is the key question that remains to be comprehensively addressed in the literature. Numerous research investigations, as listed above, highlight the existing or theoretical energy potential of various water infrastructure sites, but fail to examine the potential variation of such average energy potential estimates (Da Silva et al., 2011). Water flow and pressure are known to vary significantly throughout a typical day, from day to day, weekday to weekend, by season and over the longer term. Water flow and pressure are also subject to significant changes due to the addition of new industries, new demands, water charging or water saving schemes, etc. Such changes in flow and pressure would have a significant influence on the efficiency of a turbine converting the excess energy to electricity.

Turbines are typically designed for a particular design flow, and variations as either increases or decreases in flow and head will reduce power production. The extent of the reduction will depend on the magnitude of the change in flow conditions and the type of turbine in question. For example, PATs have been cited in several investigations as a suitable turbine for energy recovery in water pipelines (Williams, 1996; Williams et al., 1998; Giugni, et al., 2009); but a 50% increase or decrease in the design flow for a PAT would result in a reduction in energy conversion efficiency from typically 80% to less than 30%. Changes in the average flow in a water pipeline of 50% or more are a common occurrence in water supply networks.

Aside from the aforementioned technical limitations in existing feasibility studies, sufficient scrutiny of the economic viability, in terms of revenue generation, is also lacking in many studies. Many investigations determine the annual return from an MHP installation using the unit price of electricity or using the local FIT rate of hydropower generation (Ramos et al., 2011). However, the economic viability of such installations is influenced to a very significant degree by the end use of the electricity generated.

Plants that use the electricity on site will make a saving on electricity purchases at market rates while plants that sell the electricity to the grid will do so using the local FIT rate. The savings in electricity purchases include not only the unit price but also any taxes or duties applied to the supply of electricity such as sales tax, value added tax and carbon tax. Many studies fail to account for the actual cost to the water industry of electricity savings including all taxes and charges. Furthermore, studies also fail to determine the effect on plant feasibility of future changes in energy prices. Electricity prices vary significantly across Europe, for example, from as little as €0.08/kWh in Bulgaria to as much as €0.29/kWh in Denmark.

178

A. McNabola et al. / Water Policy 16 (2014) 168–183

For plants that have no use for electricity generated on site, selling to the grid will provide a longer return on investment as FIT rates are often lower than the consumer price of electricity including all taxes and charges, but higher than the market rates. Again, FIT rates and payment models vary significantly across different countries and previous investigations have failed to examine the sensitivity of proposed energy recovery sites to the value of the FIT rate.

4. Environmental impact Primary environmental concerns in relation to the water industry are the depletion of freshwater resources and pollution arising from wastewater treatment. The availability of freshwater resources varies widely, for example, within the EU, it ranges from less than 1,000 m3 per capita in the Czech Republic and Cyprus, to over 20,000 m3 per capita in Finland and Sweden (Eurostat, 2012). It is projected that climate change will reduce the availability of freshwater in lower mid-latitude regions such as the Mediterranean and increase the frequency of severe droughts (Gössling et al., 2011). According to UNEP (2002), one-third of the world’s population currently live in countries suffering from moderate or high water stress (where water consumption is more than 10% of renewable freshwater resources), and this is projected to increase to two-thirds of the world’s population by 2030. An important environmental impact from the water industry is the consumption of energy, usually entailing the depletion of non-renewable resources (fossil fuels), and associated greenhouse gas (GHG) emissions (Rothausen & Conway, 2011).

A carbon footprint is a measure of the total amount of GHG emissions, expressed as CO2 equivalents (CO2e) according to their global-warming potential, which result from an activity or series of activities involved in the life cycle of a product or process (Shrestha et al., 2012). Most of the energy used by the water sector is in the form of electricity, with an associated carbon footprint ranging from less than 0.1 kg CO2e/kWh for nuclear and renewable generated electricity to over 0.9 kg CO2e/kWh for coalgenerated electricity. Average emission factors are 0.38 and 0.57 kg CO2e/kWh, respectively, for the EU27 and USA (DEFRA, 2011). The water industry in the United States is responsible for 5% of total US carbon emissions annually (Griffiths-Sattenspiel & Wilson, 2009). In the UK, emissions associated with water supply and treatment are estimated to average 0.34 and 0.7 kg CO2e/m3, respectively, totalling 5.01 Mt CO2e/year in 2010–11 (Water UK, 2012), equivalent to approximately 1% of UK GHG emissions. In Italy, investigations have estimated that the carbon footprint of public water supply is 0.9 kg CO2e/m3 (Botto et al., 2011). Energy consumption and GHG emissions from the water industry are related to local freshwater availability. In regions where demand exceeds availability from freshwater resources, freshwater is pumped long distances from regions of water surplus, or produced from desalination, incurring considerable energy consumption and GHG emissions. The carbon emissions associated with the water industry worldwide are likely to grow if current trends are not reversed due to: rising water demand; limited and remote locations of freshwater; more stringent and energy-intensive water treatment regulations and technology.

Investigations have highlighted the effects of numerous water management and water supply scenarios on the carbon footprint of specific systems. An investigation in the USA compared five water management scenarios to the baseline scenario and found that the carbon footprint of existing water services in Las Vegas was 0.84 million tonnes CO2e per annum. It was also found that increases in demand for water could increase this figure by over 12% by 2020, and that increasing renewable energy input

A. McNabola et al. / Water Policy 16 (2014) 168â&#x20AC;&#x201C;183

179

could reduce emissions by over 20% (Rothausen & Conway, 2011). In Florida, USA, a similar investigation examined the effect of twenty water infrastructure expansion alternatives on the carbon footprint of the service (Qi & Chang, 2012). The investigation examined the expansion of the Manatee County water supply system in the period from 2011 to 2030 using options such as exploiting further groundwater, surface water, transferring regional water and others. Transferring raw water from regional sources was estimated to result in the highest carbon footprint of 2.26 million tonnes CO2e over the assessment period. However, no investigations have been found during this review that examined the effects of widespread implementation of MHP energy recovery on carbon footprints in the water industry. Further research is therefore required to investigate the potential impact of this on the industry.

Within the UK, the water industry does contribute to national GHG emission reductions through renewable energy generation. In 2010â&#x20AC;&#x201C;11, 877 GWh were generated by the UK water industry (Water UK, 2012), equivalent to a saving of approximately 0.521 Mt CO2e. Over 90% of this energy is sourced from anaerobic digestion in wastewater treatment plants, suggesting that renewable energy generation (or, at least, energy capture) from the supply network may be underexploited. Capturing energy from the supply system could help the UK water industry meet its proposed target for 20% of energy to be sourced from renewable sources, which would exceed the advised 15% target set out by the government for 2020 (Environment Agency, 2009a, b). Investigations have also highlighted the equivalent CO2 emissions saving of a number of potential sites or demonstration projects. In Ireland, an investigation of seven BPTs and three service reservoirs estimated a potential CO2 emissions saving of 1,350 tonnes annually (McNabola et al., 2011).

5. Organisational challenges For an energy recovery project to be implemented within the water industry, it is necessary for a number of stakeholders to come together, such as local government, water utilities, electricity suppliers and regulators, turbine manufacturers, etc. Effective collaboration between this network of organisations will ensure the successful, cost- and time-effective implementation of such schemes. The development of a strong collaboration network from an early stage could also increase and encourage future collaborative projects to be implemented. In a recent paper by Gausdal & Hildrum (2012), a processbased framework for the development of inter-firm networks in the water technology industry was established. It outlined how the researchers facilitated group meetings, encouraged dialogue and promoted action from dialogue with a focus on trust building among the network.

To encourage collaboration between different organisations in the water and energy industries, it is necessary to first investigate and understand these organisations, including both their structure and characteristics and previous collaborative history. Over 90% of the approximately 250,000 water service systems worldwide are municipally owned water and wastewater utilities, while only 8% are privately operated and/or owned (Kwok et al., 2010). The water industry comprises asset owners, operators, engineering specialists (design and constructions), and suppliers of equipment. On the supply side, the industry includes the following technologies: water filtration membranes, UV radiation, biological water-cleansing processes, and energy-efficient recycling of sludge and industrial wastewater. On the demand side, the customer base includes waterworks, sewer plants, and construction firms. There is a significant growth potential in this industry as the global demand for clean water and the need for energy-efficient water purification increase.

180

A. McNabola et al. / Water Policy 16 (2014) 168â&#x20AC;&#x201C;183

Within the boundaries of the industry, the networks of firms may collaborate on joint research and development (R&D) projects and enhanced water-cleansing technologies. However, there may not be a history of trust to enable firms to engage in progressively more complex and risky collaboration activities. The challenges of the need for development and innovation translate into the need for collaboration among firms in the industry and, even, the establishment of new firms to exploit the new technologies.

This collaboration requires trust and can take time to emerge (Coughlan & Coghlan, 2011).

The evolution of networks of firms, with contractual bases for their relationships, brings to mind the twin concerns of competition and collaboration (Coughlan & Coghlan, 2011). Competition comes easily to firms whose focus is on the market. Further, collaboration comes (relatively) easily to firms that have an interest in a relationship. Where it becomes difficult is when the improvement imperative requires both collaboration and competition. Then, market-based relationships need to be revisited in an environment of potential reconciliation, a search for sustainability and a reduction of risk. Here, working to achieve sustainable strategic improvement and a corresponding transition from a strategic to a learning and transformational network is a problem, the resolution of which requires time and thoughtful application of resources.

Further research in this field is required to develop a model of organisational collaboration between the water and energy industries. This may then facilitate the more widespread implementation of energy recovery technology in water infrastructure.

6. Discussion and conclusions It can be seen from the available literature that energy recovery using MHP in the water industry is a growing area of research and industrial activity. Many successful demonstration projects are in existence and many promising feasibility studies have been carried out. However, to date, the examples in existence have been implemented on an ad hoc basis and little market penetration of this concept has occurred (Zakkour et al., 2002b).

From the review, it is clear that further research is required in a number of key areas. Future research should aim to address the uncertainties that exist due to the variation in water demands daily, weekly, seasonally and in the longer term. It should also address the sensitivity of projects to changes in electricity prices and/or FIT rates. Such uncertainty creates an unacceptable risk to investment in MHP infrastructure if design conditions are open to significant change during its lifetime. These uncertainties may be part of the reason why, given the vast number of suitable sites identified in the literature, only a small number of MHP installations are in existence in the industry.

In addition, more detailed information on the investment costs is required to facilitate the growth of this sector. Many studies to date have used cost estimates for MHP construction, however, the expenses associated with consulting, planning, connection to the grid, maintenance, etc. are often neglected. In essence, a more transparent and reliable model of energy recovery potential and return on investment is required for MHP energy recovery to prosper.

The environmental impact of the water industry, its carbon footprint and energy consumption have all been shown to be globally significant. Studies have set out to examine means of limiting or reducing the carbon footprint of the water industry, but none have included the option of the widespread implementation of MHP energy recovery systems. Quantification of the carbon footprint of MHP energy recovery in the water industry in comparison to other methods of energy saving or generation is an important

A. McNabola et al. / Water Policy 16 (2014) 168–183

181

missing element in the development of this concept. It has been noted that the water industry has a range of options through which it may choose to reduce its energy demand such as wind or solar power, and biogas combustion. The selection of such investment options should be informed by both the economic viability of a potential scheme and the environmental benefits of the various alternatives.

It has also been shown that the need exists for the development of collaboration models between the water and energy industries to facilitate a more widespread implementation of MHP energy recovery in water infrastructure. The governance, regulation and organisation of the water industry across jurisdictions are diverse and complex. The structure of these organisations influences the ability of the water industry to deliver on sustainable strategic improvements such as MHP energy recovery. Collaboration models shedding light on the operation of these large sets of organisations may enable the development of more effective policy to promote the implementation of MHP energy recovery in future. This, together with more dissemination of research findings to industry and policymakers, may act as a catalyst for the improvement of the sustainability of the water industry.

Acknowledgement This project was part funded by the European Regional Development Fund (ERDF) through the Ireland–Wales Programme (INTERREG 4A).

References Berndtsson, J. C. & Jinno, K. (2008). Sustainability of urban water system: examples from Fukuoka, Japan. Water Policy 10, 501–513.

Botto, S., Niccolucci, V., Rugani, B., Nicolardi, V., Bastianoni, S. & Gaggi, C. (2011). Towards lower carbon footprint patterns of consumption: the case of drinking water in Italy. Environmental Science & Policy 14, 388–395.

Brown, L. E., Mitchell, G., Holden, J., Folkard, A., Wright, N., Beharry-Borg, N., Berry, G., Brierley, B., Chapman, P., Clarke, S. J., Cotton, L., Dobson, M., Dollar, E., Fletcher, M., Foster, J., Hanlon, A., Hildon, S., Hiley, P., Hillis, P., Hoseason, J., Johnston, K., Kay, P., McDonald, A., Parrott, A., Powell, A., Slack, R. J., Sleigh, A., Spray, C., Tapley, K., Underhill, R. & Woulds, C. (2010). Priority water research questions as determined by UK practitioners and policy makers. Science of the Total Environment. 409(2), 256–266.

Carravetta, A., Del Giudice, G., Fecarotta, O. & Ramos, H. M. (2012). Energy production in water distribution networks: a PAT design strategy. Water Resource Management 26(13), 3947–3959.

Colombo, A. & Kleiner, Y. (2011). Energy recovery in water distribution systems using microturbines. Probabilistic Methodologies in Water and Wastewater Engineering. Toronto, Canada.

Corcoran, L., Coughlan, P. & McNabola, A. (2012). Energy recovery potential of the Dublin region water supply network.

Proceedings of the International Water Association World Congress on Water, Climate and Energy, Dublin, Ireland.

Coughlan, P. & Coghlan, D. (2011). Collaborative Strategic Improvement through Network Action Learning. Cheltenham, Edward Elgar, UK. Da Silva, B. L. A., Tofoli, F. L. & Scartazzini, L. S. (2011). Case study: hydroelectric generation employing distribution network in Pato Branco, Brazil. Proceedings of the IASTED International Conference on Power and Energy Systems, June 22–24, Crete, Greece.

DEFRA (2011). 2011 Guidelines to DEFRA / DECC’s GHG Conversion Factors for Company Reporting. DEFRA, UK. EcoGeneration (2008). Powering Sydney Waters carbon neutral future. News article. July–August 2008.

Engelhardt, M. O., Skipworth, P. J., Savic, D. A., Saul, A. J. & Walters, G. A. (2000). Rehabilitation strategies for water distribution networks – a literature review with a UK perspective. Urban Water 2(2), 153–170.

Engineering & Technology Magazine (2010). Power generation by Archimedes screw.

182

A. McNabola et al. / Water Policy 16 (2014) 168–183

Environment Agency (2009a). Evidence: A low carbon water industry in 2050. Resource Efficiency Programme, Evidence Directorate. Report SC070010/R3 (pp.1, 41 p.). Environment Agency, Bristol.

Environment Agency (2009b). Renewable Energy Potential for the Water Industry Evidence (pp.1, 43 p.). Environment Agency, Bristol.

Eurostat (2012). Freshwater resources in Europe. URL, accessed October 2012: http://epp.eurostat.ec.europa.eu/statistics_ explained/index.php?title=File:Freshwater_resources_per_capita_-_long-term_average_(1)_(1_000_m%C2%B3_per_ inhabitant).png&filetimestamp=20111115184708#filelinks.

Fontana, N., Giugni, M. & Portolano, D. (2012). Losses reduction and energy production in water-distribution networks.

Journal of Water Resources Planning and Management 138(3), 237–244.

Gaius-obaseki, T. (2010). Hydropower opportunities in the water industry. International Journal of Environmental Sciences 1(3), 392–402.

Gausdal, A. H. & Hildrum, J. M. (2012). Facilitating trust building in networks: a study from the water technology industry.

Systemic Practice in Action Research 25, 15–38.

Giugni, M., Fontana, N. & Portolano, D. (2009). Energy Savings policy in water distribution networks. Proceedings of the International Conference on Renewable Energies & Power Quality, Valencia, Spain.

Gonçalves, F. V., Costa, L. H. & Ramos, H. M. (2011). Best economical hybrid energy solution: model development and case study of a WDS in Portugal. Energy Policy 39(6), 3361–3369.

Gonçalves, F. V. & Ramos, H. M. (2012). Hybrid energy system evaluation in water supply systems: artificial neural network approach and methodology. Water Supply: Research and Technology – AQUA 61(2), 59–72.

Griffin, F. (2000). Feasibility of energy recovery from a wastewater treatment scheme. Journal of Power and Energy 214(1), 41–51.

Gössling, S., Peeters, P., Hall, M., Ceron, J. P., Dubois, G., Lehmann, L. V. & Scott, D. (2011). Tourism and water use: supply, demand and security. An international review. Tourism Management 33, 1–15.

Griffiths-Sattenspiel, B. & Wilson, W. (2009). The Carbon Footprint of Water. A River Network Report.

Hernandez-Leal, L., Temmink, H., Zeeman, G. & Buisman, C. J. N. (2010). Bioflocculation of grey water for improved energy recovery within decentralized sanitation concepts. Bioresource Technology 101(23), 9065–9070.

Howe, A. (2009). Renewable energy potential for the water industry. Environment Agency Report: SC070010/R5. Bristol, UK. Jenerette, G. D. & Larsen, L. (2006). A global perspective on changing sustainable urban water supplies. Global and Planetary Change 50(3–4), 202–211.

Khennas, S. & Barnett, A. (2000). Best practices for sustainable development of micro hydropower in developing countries.

Final Synthesis report. Contract R721.

Kwok, S. C., Lang, H., O’Callaghan, P. & Stiff, M. (2010). Water Technology Markets 2010 : Key Opportunities and Emerging Trends. Media Analytics Ltd, Oxford.

Lofman, D., Petersen, M. & Bower, A. (2002). Water, energy and environment nexus: The California experience. Water Resources Development 18(1), 73–85.

McNabola, A., Williams, A. P. & Coughlan, P. (2011). The technical & economic feasibility of energy recovery in water supply networks. International Conference on Renewable Energy and Power Quality, Las Palmas de Gan Canaria, 13th–15th April.

McNabola, A., Williams, A. P. & Coughlan, P. (2013). Energy recovery in water supply networks: an assessment of the potential of micro hydropower. Water and Environment Journal 27, 435–436.

Mikus, K. (1984). Erfahrungen mit Kreiselpumpenanlagen zur Energieruckgewinnung aus dem Trinkwassersystem. Das Gas und Wasserfach – Wasser/Abwasser 124(4), 159–163.

Park, S. W. & Jang, C. H. (2010). Characteristics of carbonized sludge for co-combustion in pulverized coal power plants.

Waste Management 31(3), 523–529.

Power, C., Coughlan, P. & McNabola, A. (2012). The applicability of hydropower technology in waste water treatment plants.

Proceedings of the International Water Association World Congress on Water, Climate and Energy, Dublin, Ireland.

Qi, C. & Chang, N. B. (2012). Integrated carbon footprint and cost evaluation of a drinking water infrastructure system for screening expansion alternatives. Journal of Cleaner Production 27, 51–63.

Ramos, S., Gomes, A., Cardoso, M. & Castanheira, L. (2009). Advantage of public water supply use for micro power generation. Proceedings of the International Conference on Renewable Energy and Power Quality, Valencia, Spain.

Ramos, H. M., Kenov, K. N. & Vieira, F. (2011). Environmentally friendly hybrid solutions to improve the energy and hydraulic efficiency in water supply systems. Energy for Sustainable Management 15(4), 436–442.

A. McNabola et al. / Water Policy 16 (2014) 168–183

183

Ramos, J. S. & Ramos, H. M. (2009). Sustainable application of renewable sources in water pumping systems: Optimized energy system configuration. Energy Journal 37(2), 633–643.

Ramos, J. S. & Ramos, H. M. (2010). Multi-criterion optimization of energy management in drinking systems. Water Science & Technology: Water Supply 10(2), 129–144.

Rentricity (2007). Rentricity – Aquarion Water Energy Recovery Pilot Study.

Rothausen, S. G. S. A. & Conway, D. (2011). Greenhouse-gas emissions from energy use in the water sector. Nature and Climate Change 1(4), 210–219.

Rygaard, M., Binning, P. J. & Albrechtsen, H.-J. (2011). Increasing urban water self-sufficiency: New era, new challenges.

Journal of Environmental Management 92(1), 185–194.

Saket, R. K. (2008). Design, development and reliability evaluation of micro hydro-power generation system based on municipal waste water. IEE Electrical power and Energy Conference, pp.1–8, 6–7 October, Vancouver, Canada.

Scholten, L., Scheidegger, A., Reichert, P. & Maurer, M. (2013). Combining expert knowledge and local data for improved service life modeling of water supply networks. Environmental Modelling & Software 42(0), 1–16.

Shrestha, E., Ahmad, S., Johnson, W. & Batista, J. R. (2012). The carbon footprint of water management policy options.

Energy Policy 42, 201–212.

UNEP (2002). State of the Environment and Policy Retrospective: 1972–2000. United Nations Environment Programme, Geneva.

Venkatesh, G. & Brattebø, H. (2011). Energy consumption, costs and environmental impacts for urban water cycle services: case study of Oslo (Norway). Energy 36(2), 792–800.

Vieira, F. & Ramos, H. M. (2009a). Optimization of operational planning for wind/hydro hybrid water supply systems.

Renewable Energy 34(3), 928–936.

Vieira, F. & Ramos, H. M. (2009b). Optimization of the energy management in water supply systems. Water Science & Technology: Water Supply 9(1), 59–65.

Wallace, A. R. (1996). Embedded mini-hydro generation in the water supply industry. IEE Conference Publication. 419, 168–171.

Water UK (2012). Sustainability Indicators 2010/11. Water UK, London.

WHO (2008). Guidelines for Drinking-water Quality: Incorporating 1st and 2nd Addenda, Vol.1, Recommendations (3rd edn).

World Health Organization, Geneva.

Williams, A. A. (1996). Pumps as Turbines for Low Cost Micro Hydro Power. World Renewable Energy Congress, Denver, USA. Williams, A. A., Smith, N. P. A., Bird, C. & Howard, M. (1998). Pumps as turbines and induction motors as generators for energy recovery in water supply systems. Journal of the Chartered Institute of Water and Environmental Management 12, 175–178.

Zakkour, P. D., Gaterell, M. R., Griffin, P., Gochin, R. J. & Lester, J. N. (2002a). Developing a sustainable energy strategy for a water utility. Part I: a review of the UK legislative framework. Journal of Environmental Management 66, 105–114.

Zakkour, P. D., Gaterell, M. R., Griffin, P., Gochin, R. J. & Lester, J. N. (2002b). Developing a sustainable energy strategy for a water utility. Part II: a review of potential technologies and approaches. Journal of Environmental Management 66, 115–125.

Zhe, L., Boyle, F. & Reynolds, A. (2010). Rainwater harvesting and greywater treatment systems for domestic application in Ireland. Desalination 260, 1–8.

Zilberman, D., Sproul, T., Rajagopal, D., Sexton, S. & Hellegers, P. (2008). Rising energy prices and the economics of water in agriculture. Water Policy 10, S1, 11–21.

Received 24 October 2012; accepted in revised form 6 August 2013. Available online 3 October 2013

Water Policy 16 (2014) 184â&#x20AC;&#x201C;196

Traditional water rights, ecology and the public trust doctrine in Hawaii C. S. Papacostas Department of Civil and Environmental Engineering, University of Hawaii at Manoa. E-mail: csp@hawaii.edu

Abstract This case study discusses the implications of imposing the doctrine of public trust to ground and surface waters within the State of Hawaii and its effects on traditional rights that had previously evolved based on common law. It traces the major events of the history of water rights and practices beginning with the system devised by the indigenous Hawaiian people prior to the adoption of the doctrine of public trust to the water resources of the State of Hawaii, applied with the most expansive interpretation of the public trust doctrine, encompassing both surface and subsurface waters and a wide assortment of protected uses and purposes. The major decisions that ensued when applying the doctrine, via legal prescriptions and administrative rules, are described. The implications of the interplay between scientific enquiry and research are presented, with legal precedent in the face of potential water shortages, competing uses, sensitivities to comprehensive resource management, considerations of ecological balance and protection of the rights of indigenous people. Many of these findings are transferable to other jurisdictions contemplating the adoption of public trust doctrine principles to their surface and ground waters.

Keywords: Ecology; Legal rights; Resource management; Water

Introduction Attributed to Roman Emperor Justinian, the public trust doctrine is predicated on the principle that certain resources can be reserved by government for the benefit of the public. When applied to water rights, this doctrine is, by its nature, antithetic to conventional rights conferred upon private parties, akin to private property rights. It imposes limitations on both riparian rights (most prevalent in the eastern states of America) that apply to lands that are contiguous to bodies of water and prior appropriation rights (commonly encountered in the western states of America) that allocate water based on the timing of use.

In the United States, the public trust doctrine had been historically applied to the protection of navigable waters until 1983 when, in the often-cited case of National Audubon Society v. Superior Court, the California Supreme Court ruled that, in connection with the Mono Lake, the state had the affirmative duty to balance conventional rights with ecological values pursuant to the public trust doctrine. doi: 10.2166/wp.2013.182 Š IWA Publishing 2014

C. S. Papacostas / Water Policy 16 (2014) 184–196

185

Opponents of this application to ecological concerns called it the wrong remedy. For example, Walston (1982) argued that although it ‘would provide a basis for protection of important environmental resources … it would jeopardize long standing water rights’. Nevertheless, the concept gradually took root and proponents have even argued in favor of broadening its scope. For instance, Bader (1992) concluded that ‘the public trust doctrine originated out of the desire to protect natural systems so as to ensure their beneficial use by society. Today we must expand the doctrine in a manner that faithfully links environmental protection and resource utilization.’ In the same vein, more recently, Brown (2006) advocated a more aggressive application by expanding ‘the waters that are subject to the public trust doctrine’ and by increasing ‘the doctrine’s reach to include additional purposes and uses within the protection of the doctrine’. At the opposite end, Proctor (2004) bemoaned her finding that ‘the surviving rights that a riparian owner can still exercise in Florida … actually only include those that do not interfere with the state’s use of sovereignty lands, the federal government’s use of the navigational servitude, or the public trust’.

In a revisionist’s view of the actual impact of the Mono Lake decision, Owen (2012) has made the case that ‘the doctrine, though important, has exerted less influence upon California water management than conventional wisdom suggests’, and that ‘achieving the greatest future impact of the public trust doctrine will require more effective integration into the larger system of administrative environmental law’ rather than ‘greater judicial involvement’.

It is relevant to point out that, as Klass & Huang (2009) succinctly advise, ‘at its core, the public trust doctrine remains a state-based doctrine, unique to each state but with lessons transferable to other states’.

To achieve this end, it is instructive to share information about the evolving experiences, both judicial and administrative, of the various states.

This paper describes the experience of the State of Hawaii, which, by many accounts, has adopted the most comprehensive application of the public trust doctrine to both surface and ground waters, and the doctrine’s effect on traditional indigenous practices prior to western contact and conventional rights that have evolved since, based on common law. The major decisions that ensued following the application of the doctrine via legal prescriptions and administrative rules, in the presence of the earlier mixture of rights, are described and their implications are discussed in view of water availability and potential water shortages, competing reasonable and beneficial uses, and sensitivity to comprehensive resource management. Many of these findings are transferable to other jurisdictions contemplating the adoption of public trust doctrine principles to their surface and ground waters.

Pre-contact water practices Prior to western contact through Captain James Cook in 1778, a feudal-like political organization had evolved with a strong religious foundation in the kapu (or taboo) system that regulated the behavior of four population segments: the Chiefs (ali`i), the priesthood (kahuna), the commoners (maka`ainana), and kauwa, an outcast class (e.g., Kamakau, 1992). Under this strictly regulated system, an indigenous population, estimated to have reached about half a million people, managed to sustain itself and thrive in the most isolated land-mass location on earth.

The kapu system was abruptly abolished in 1819, just a year before the first Protestant Calvinist missionaries, sent by the American Board of Commissioners for Foreign Missions, arrived from Boston, an event that marked the beginning of a dramatic transition from traditional to modern societal structures, including concepts of private ownership of land and water rights.

186

C. S. Papacostas / Water Policy 16 (2014) 184–196

The Ahupua`a system The western concept of privately owned land and water was unknown in old Hawaii. Under the revocable control of a hierarchy of chiefs, reaching all the way up to the King (Mo`i) who was the ultimate trustee, land divisions known as Ahupua`a were regulated by designated supervisors (konohiki), whose responsibilities included the stewardship of the natural resources and the collection of the tribute to the Chiefs from the commoners who worked the land and the sea.

Sustainable practices relating to freshwater (or wai) were of critical importance in the ability of the original inhabitants of the islands to survive and prosper. In their language, the word for ‘wealth’ was simply the reduplication waiwai, whereas the word for ‘law’ incorporated ‘water’ as a constituent of kanawai.

The native population devised an ingenious irrigation system to support the cultivation of taro within the fertile land districts. It consisted of an elaborate system of irrigation ditches (`auwai) that directed stream water to agricultural terraces (lo`i) and back to the stream for subsequent reuse downstream via levees (pani wai) and sluice gates (Dole, 1892; Nakuina, 1893; Kamakau, 1992; Lucas, 1995; Martin et al., 1996). The strict regulation of the distribution of water in terms of allotted amounts of water and irrigation time schedules was controlled by the konohiki who ‘lived in the village and was intimately familiar with its customs, resources, and current physical conditions as well as each individual’s effort and merit. When disputes arose it was the konohiki who was responsible for their resolution’ (Wilcox, 1996). These water-use practices closely resemble those associated with riparian rights to water where occupants of land at the banks of streams and rivers are afforded the reasonable use of water as long as they do not adversely affect other downstream users’ rights to the resource.

Post-contact turmoil Social and economic upheaval of the old system of land and water commenced with the arrival of Captain Cook in January 1778. Soon, the islands became a stopover and resupply station for fur traders between east and west, and this was followed by the sandalwood trade with China and the development of a major hub of the whaling industry.

The presence of traders and whalers necessitated changes in agriculture and animal husbandry, including the keeping of introduced cattle, in order to replenish the ship-holds of the westerners with supplies to their liking. It was with the direct aid of foreign weapons and strategic advice that King Kamehameha I (‘the Great’), a Chief from the island of Hawaii, finally unified the islands in 1810 under his home island’s name. The arrival of the Calvinist missionaries, beginning in 1820, added another major component to the Hawaiian tapestry.

Kamehameha’s successors adopted western ways and by 1840, under foreign influences, Kamehameha III enacted the Hawaiian Kingdom’s first constitution, completing the transition from a feudal-like to a constitutional monarchy system of governance (Kosaki, 1978; Sai, 2008) that eventually led to the private ownership of land and water rights.

Ownership of land and water Among the pressures placed on Hawaii’s government from the ever-increasing number of foreigners entering the country was their desire ‘to reside there, to engage in business … to acquire house lots and land, by

C. S. Papacostas / Water Policy 16 (2014) 184–196

187

lease or otherwise, to build houses on the lot so acquired, and to transfer their property either by sale, lease, will, or inheritance’ (Kuykendall, 1938). The upshot of such foreign advice and guidance was that the same King embarked in 1848 on a land division, known as the Mahele, between himself and 245 Chiefs, including konohiki. This was a major step that eventually ‘transformed Hawaii’s communal land tenure system to a private property system’ (Liu, 2002). A provision of the new legal system relating to water that has found its way into today’s Section 7–1 of the Hawaii Revised Statutes (HRS) states: ‘The springs of water, running water, and roads shall be free to all, on all lands granted in fee simple; provided that this shall not be applicable to wells and watercourses, which individuals have made for their own use’ (State of Hawaii, 1987).

According to an early interpretation that emerged subsequently, water rights ‘went with the land unless the owner had previously formally relinquished, transferred, divided, or neglected it for a period of over twenty years’ (Nakuina, 1893).

The primary reference to water up to this point was essentially confined to surface water generally flowing in streams. Groundwater that could be reached by the drilling of wells was, as the quote from HRS in the previous section attests, owned by the overlying property owners. In Hawaii, groundwater became a subject of contention after 1879 when the first artesian well was dug into a huge basal freshwater lens on the island of Oahu. Other categories of groundwater include dike-impounded water, perched water confined by impervious soil layers, and water flowing in underground channels. The first two categories were collectively referred to as ‘percolating’ underground water (Takemoto, 1986).

Plantation economy and water Sugar cane was cultivated by the original inhabitants of the islands, but it was in the middle of the 19th century that it became the prime component of Hawaii’s economy, as a result of the contemporaneous changes in land ownership and water rights, but also because of the appearance of strong markets in the northern US states during the American Civil War when embargos prevented the importation of sugar from southern plantations (Wilcox, 1996).

Sugar cane cultivation demands large expanses of land and great volumes of irrigation water. This combination necessitated the construction of conduits (ditches and tunnels) from surface and underground sources of water to dry lands that were often located significant distances away. The first-documented probable ‘water tunnel or aqueduct constructed in these islands for the purpose of conveying water from a stream for agricultural purposes’ (Williams, 1918) was estimated to have been completed prior to 1849. Over the next few decades, a remarkable network of irrigation ditches criss-crossed all the major islands in support of the cultivation of a very thirsty crop.

In a widely published and circulated article, O’Shaughnessy (1904), a principal hydraulic engineer in Hawaii who subsequently headed the construction of the Hetch Hetchy water system in San Francisco explained that ‘the development of irrigation projects in the Hawaiian islands has been prosecuted with the greatest vigor by private corporations owning sugar estates during the last ten years. No aid for this work has been received from either the local Territorial Government of Hawaii or the National Government at Washington.’ He noted that ‘all the streams on the island of Maui are now tapped by ditches’.

Privately owned sugar and other plantations were granted water development rights resembling the principles of prior appropriation, with provisos relating to previously established traditional and appurtenant water rights. This situation persisted through the overthrow of the Kingdom of Hawaii in 1893 by sugar growers and American businessmen with the support of US marines, the short-lived Republic of

188

C. S. Papacostas / Water Policy 16 (2014) 184–196

Hawaii that replaced the monarchy, the establishment of the Territory of Hawaii in 1898 and the attainment of statehood as the 50th state of the USA in 1959.

Period of stability For the greater part of the 20th century, Hawaiian water law was considered settled and firmly established in a unique Hawaiian way, motivating Monahan & Yamauchi (1972) to declare that ‘in general, Hawaiian surface water rights are fairly well defined and offer greater investment security’ than other options. In this environment of relative stability, the courts’ principal role was to settle disputes among private parties. It is notable that the early legal system’s view of water was based on a limited understanding of its physical and ecological attributes. For example, any water that ‘escaped’ to the ocean was often described as being allowed to ‘run to waste’ as opposed to being put to some use (Papacostas, 2001).

Common law water rights Surface water rights The evolution of the law regarding surface waters during this period has been widely documented in legal and scholarly publications. The major elements of surface water rights in Hawaii included the following: Traditional appurtenant rights. These rights, following the Mahele, were attached to the lands that enjoyed the use of water for cultivation or for domestic purposes at that time. The quantity of water so designated was fixed to the amount used at the time when a land grant was awarded by the Land Commission that was appointed in 1845 to settle issues of ownership (Clark, 1986). Through such awards, individual land parcels were converted into fee simple lands. According to Johnson (2009) ‘early cases dealing with the sugar plantations permitted transfers of appurtenant water rights but only after a showing that no harm would be done to other rights holders’.

Riparian rights. Etymologically derived from the Latin word ripa (the bank of a waterway), these rights apply to lands that are adjacent to rivers or streams. In Hawaii, limited riparian rights were fixed by common law in a 1917 Hawaii Supreme Court case (Carter v. Territory of Hawaii) where the concept was applied to surplus freshet (storm) flows only (Hutchins, 1972; Monahan & Yamauchi, 1972; Clark, 1986). Rooted in English common law, riparian rights allow an owner of land adjacent to a body of water to enjoy its use while protecting, by not diminishing, the rights of others similarly situated to do the same. In the USA, this system of water rights is prevalent in the eastern states that have adopted English common law.

Konohiki rights. In a milestone 1930 case (Territory v. Gay), the Hawaii Supreme Court decided that ‘surplus normal flow’ belonged to the owner (or konohiki) of the lands where it originated. Surplus normal flow was defined as the flow remaining after subtracting from the stream’s ‘normal flow’ the flow allotted by traditional appurtenant and prescriptive rights. This decision was affirmed by the US Ninth Circuit Court of Appeals in 1931 (Hutchins, 1972; Clark, 1986). As Monahan & Yamauchi (1972) put it, ‘surplus waters of normal flow [were] entirely owned by the land owner of the unit

C. S. Papacostas / Water Policy 16 (2014) 184–196

189

within which they [arose]’ and, consequently, ‘not being appurtenant to any part of his land’, could ‘be conveyed and used elsewhere’.

The Homestead Act provision. In what may at first glance appear to be unrelated to water rights, the US Congress adopted in 1921 the Hawaiian Homes Commission Act (HHCA), the intent of which was to rehabilitate native Hawaiians to land by providing long-term leases to them, having recognized the undesirable consequences of the Mahele in alienating the vast majority of the indigenous people from the privilege of land ownership (Liu, 2002; Sylva, 2007). The HHCA explicitly provided that ‘sufficient water shall be reserved for current and foreseeable domestic, stock water, aquaculture, and irrigation activities on lands leased to native Hawaiians’. This provision was critical ‘because the lands designated for homesteading were second-class, dry and arid lands’ (Liu, 2002).

Groundwater rights As in most other states, surface and ground water were historically defined as separate entities rather than as two components of a single hydrologic system. Furthermore, in Hawaii, groundwater rights distinguished between artesian waters and other groundwater categories.

Correlative rights. The correlative rights doctrine for groundwater was adopted in Hawaii in 1929 by the City Mill v. Honolulu Sewer and Water Commission case (Hutchins, 1972; Takemoto, 1986; MacDougal, 1988; Callies & Chipchase, 2007). According to Takemoto (1986), ‘for percolating groundwater, the courts developed three alternative doctrines: the English or absolute ownership doctrine (existing in earlier Hawaii law), the American or reasonable use doctrine, and the correlative rights doctrine’. Under the absolute doctrine, the overlying owner has unrestricted use including transporting the water elsewhere, whereas, under the reasonable ownership doctrine, ‘reasonable’ use on overlying lands is unlimited or unrestricted.

‘The correlative rights doctrine adds a further distinction between ‘shortage’ and ‘surplus’ conditions.

During surplus conditions, transporters get appropriative [i.e., first come, first served] rights in the surplus amount … During shortages, overlying users are entitled to a proportionate share; transporters get nothing.’

Ground Water Use Act of 1959. Thirty years after the establishment of correlative rights, in the face of potential overdrafts of underground water beyond ‘sustainable levels’ and the occasional detection of contaminants in basal waters, the Hawaii Legislature enacted the Ground Water Use Act of 1959 that established a permit system for all (i.e., proposed and existing) users in critical areas to be designated by the Board of Lands and Natural Resources. Designation would be based on data collection and scientific investigations of the water resource. The first such underground water management area was designated in 1979, 20 years after the law’s enactment (Takemoto, 1986).

Adoption of the public trust and water code What started as a typical water ownership dispute between two private parties in 1973, the case of McBryde v. Robinson resulted in upending the heretofore established water law in Hawaii in favor of the public trust doctrine. To quote from the case, ‘Thus by the Mahele and subsequent Land Commission Award and issuance of Royal Patent right to water was not intended to be, could not be, and was not transferred to the awardee, and the ownership of water in natural watercourses and rivers remained in

190

C. S. Papacostas / Water Policy 16 (2014) 184–196

the people of Hawaii for their common good’ (Supreme Court, 2000). This radical departure survived an appeal to the US District Court in Robinson v. Ariyoshi and was reaffirmed in Reppun v. Board of Water Supply, both in 1982 (e.g., MacDougal, 1988).

The Hawaii constitution was correspondingly amended to declare in Article XI, Section 7: ‘The State has an obligation to protect, control and regulate the use of Hawaii’s water resources for the benefit of the people.’ Moreover, ‘the legislature shall provide for a water resources agency which, as provided by law, shall set overall water conservation, quality and use policies; define beneficial and reasonable uses; protect ground and surface water resources, watersheds and natural stream environments; establish criteria for water use priorities while assuring appurtenant rights and existing correlative and riparian uses and establish procedures for regulating all uses of Hawaii’s water resources’ (State of Hawaii, 1987).

To implement the new constitutional provision following a series of contentious debates over a decade, the Hawaii Legislature enacted the state Water Code in 1987 and established the Water Commission within the Department of Land and Natural Resources (DLNR) to administer it.

Having first recognized the need to establish ‘a program of comprehensive resource planning to address the problems of supply and conservation of water’, the legislature declared that ‘the state water code shall be liberally interpreted to obtain maximum beneficial use of the waters of the State for purposes such as domestic uses, aquaculture uses, irrigation and other agricultural uses, power development, and commercial and industrial uses. However, adequate provision shall be made for the protection of traditional and customary Hawaiian rights, the protection and procreation of fish and wildlife, the maintenance of proper ecological balance and scenic beauty, and the preservation and enhancement of waters of the State for municipal uses, public recreation, public water supply, agriculture, and navigation. Such objectives are declared to be in the public interest’ (State of Hawaii, 1987).

Table 1 summarizes the principal provisions of the new Water Code.

Contested Waiahole case The new law was put to the test in a milestone case relating to the Waiahole Ditch system on the island of Oahu. According to a subsequent Hawaii Supreme Court opinion, ‘this dispute culminated in a contested case hearing of heretofore unprecedented size, duration, and complexity’ (Supreme Court, 2000; Klass & Huang, 2009). A summary of the major points of the dispute can shed light on some of the issues encountered in the process of converting to the public trust doctrine within the context of a contemporary legal and economic structure and can serve as lessons learned for other jurisdictions.

The irrigation ditch system was constructed between 1913 and 1916 by the Waiahole Water Co., Ltd. According to the project’s inspector engineer, ‘the general plan provided for collecting the water from the many streams and gulches on the windward side of Oahu by means of tunnels through the ridges or spurs, and conveying the water, after collecting, through the mountain in the main tunnel to the leeward side of the island, thence by tunnels, ditches and pipes, to the upper levels of Oahu Sugar Plantation’ (Kluegel, 1916). The Commission on Water Resource Management (1997) discovered that the system was expanded between 1925 and 1935 to capture dike-impounded high-level groundwater, and other improvements continued to be made as recently as 1992.

Interim flow standard. In April 1989, lacking the legally required solid scientific basis for establishing in-stream standards in accordance with the provisions of the Water Code, the Commission set the

C. S. Papacostas / Water Policy 16 (2014) 184–196

191

Table 1. Provisions of the 1987 Water Code.

Water Code Part

Provisions

Administrative Structure

Declares the overarching policy quoted above; designates the Commission on Water Resource Management in the Department of Land and Natural Resources, its composition, duties, procedures, and judicial review; establishes a water resource management fund for monitoring, data collection, research, reporting and dissemination of public information.

Requires all users of water anywhere within the state to declare their use and attributes, and the issuance of a certificate if the use declared is found to be reasonable and beneficial to be used in resolving existing water rights.

Mandates the development of a Hawaii Water Plan consisting of a coordinated water resource protection plan, county water use and development plans, a state water projects plan, and a water quality plan; calls for division of counties into hydrologic units along with current and future water uses, in-stream uses, and sustainable yields.

Pronounces procedures for designation, based on scientific investigations and research, of water management areas threatened by existing or proposed withdrawals and diversions of water for the purpose of administrative control; these areas to include existing areas designated via the Ground Water Use Act; establishes ground and surface water criteria for designation, modification or rescinding of designation; requires explicit permitting of all demonstrated ‘reasonable and beneficial’ water uses in designated management areas; provides for declaration of water shortages by the Commission and subsequent permit changes; and declares that appurtenant rights are preserved.

Designates the Department of Health as the subject agency in relation to water quality and the formulation of the Water Quality Plan.

Directs the Commission to establish and administer a statewide in-stream water protection program, including in-stream flow standards or interim in-stream flow standards on a stream-by-stream basis to protect adequately fishery, wildlife, recreational, aesthetic, scenic, or other beneficial in-stream uses.

Requires registration of all wells and establishes standards and permits for construction and pump installation.

Requires registration of such within and without management areas, and permits for construction or alteration.

Preserves certain provisions (such as those of the Hawaii Homes Commission Act of 1920) and the traditional and customary rights of ahupuaà tenants who are descendants of pre-western contact inhabitants; and preserves the appurtenant rights of kuleana and kalo lands.

Reports of Water Use

Hawaii Water Plan

Regulation of Water Use

Water Quality In-stream Uses of Water

Wells Stream Diversion Works Native Hawaiian Rights

interim in-stream flow standard (IIFS) for Windward Oahu at the actual flow in each stream on a specified date coinciding with water management area designations. The maintenance of minimum stream flow levels to facilitate in-stream uses, including recreation, fishing and protection of the aquatic environment, is commonly practiced in the USA and internationally ( Jowett, 1997; Utton & Utton, 1999). In most cases these requirements have been imposed by regulation, through the exercise of police power, or via acquisition/condemnation rather than by identifying in-stream uses as protected by the public trust doctrine, which is the most controversial approach. Five prior-appropriation western states accomplish these ends via direct appropriation of water, whereas Montana follows a reservation model of excluding further appropriation (Utton & Utton, 1999). In cases where the objective is the protection of the aquatic environment, several methods for determining the appropriate minimum flows in consideration of factors such as food availability, physical habitat requirements, temperature, water quality and biotic interactions have been devised. Among these methods are the historic flow method,

192

C. S. Papacostas / Water Policy 16 (2014) 184–196

hydraulic methods measuring discharge as a fraction of hydraulic geometry, and habitat methods that are ‘based on hydraulic conditions that meet specific biological requirements’ ( Jowett, 1997). In any case, the recognition that natural stream flows should be allowed to persist represents a clear departure from the earlier exploitative philosophy of considering any unused water as wasteful. In the words of the Hawaii Supreme Court, ‘We thus hold that the maintenance of waters in their natural state constitutes a distinct ‘use’ under the water resources trust. This disposes of any portrayal of retention of waters in their natural state as ‘waste’’ (Supreme Court, 2000). The major in-stream purpose that emerged in the Waiahole contested case was the protection and propagation of native species in the freshwater streams and the estuary at the stream–ocean interface.

The Windward Oahu area was so designated in July 1992. Pursuant to the provisions of the Water Code, the Waiahole Irrigation Co. applied for a combined-use permit to preserve all of its existing allocations. In June of the same year, the sugar company had announced plans to cease its operations by 1995. In response, the State Department of Agriculture petitioned to reserve the plantation’s water for future agricultural use. Petitions to reserve were also filed by six other groups, including three community groups and the Department of Hawaiian Home Lands. The three community groups filed petitions to amend the interim in-stream flow standard, and so did the state’s Office of Hawaiian Affairs (OHA) in 1995.

Following visits to investigate allegations of water waste and an agreement by 17 parties to mediate the IIFS issue, ‘the commission on January 25, 1995 ordered that a combined contested hearing be held on the [pending] petitions’ and other matters. A total of 25 parties became part of the combined hearing.

Prior to the closing arguments in September 1996, 161 written testimonies were introduced along with 567 exhibits (Commission on Water Resource Management, 1997).

Commission’s decision. Relying on 1109 enumerated findings of fact and an explanation of the legal grounds it considered, the Commission issued its 250-page decision in December 1997, which apportioned the available water among in-stream uses, various agricultural uses, and non-permitted buffer amounts of groundwater. Several requests for water-use permits were denied, among them a request for golf-course use and landscape uses related to a proposed residential housing development. Interestingly, among the conclusions was the fact that there was not sufficient scientific evidence to establish in-stream flow standards balancing the multiple competing uses, but the Commission decided to err on the side of caution by increasing the IIFS with supplemental ‘non-permitted’ flows for which no permit had been issued, and permitted, but unused, agricultural water. It considered the increased flows as the basis for long-term studies to enhance knowledge of these matters. Difficult as the decisions were, the Commission determined that the permitted uses would not effectuate a condition of shortage that would have a multiplicative effect on the balancing of reasonable and beneficial uses. Nevertheless, ‘aggrieved parties, representing both applicants for water use on the Leeward side and applicants seeking to release water into windward streams, appealed to the Hawaii Supreme Court’ (Callies & Chipchase, 2007).

Affirmation of the doctrine Having traced the modern expression of the public trust doctrine in the USA to an 1892 US Supreme Court decision relating to the protection of the navigable waters of Lake Michigan for the people of the state (in Illinois Central Railroad Co. v. Illinois), the Hawaii Supreme Court cited precedents establishing the concept to navigable waters ‘in and around the territory of the Hawaiian Government’ and later

C. S. Papacostas / Water Policy 16 (2014) 184–196

193

decisions that ‘confirmed our embrace of the public trust doctrine’ to other resources, for example, the lands under the shoreline high-water mark.

As to the application of the doctrine to freshwater resources, the court quoted an earlier decision that ‘The right to water is one of the most important usufruct of lands, and it appears clear to us … the right to water was specifically and definitely reserved for the people of Hawaii for their common good in all of the land grants.’ That decision recognized that in relation to freshwater ‘the extent of the state’s trust obligation of course would not be identical to that which applies to navigable waterways’ (Supreme Court, 2000).

The court then proceeded to outline the relationship of the public trust doctrine to the state Water Code and to enunciate the fundamental principles of the doctrine as applied to Hawaii’s water resources in terms of its scope and substance. In establishing these principles, the court asserted that ‘the public trust, by its very nature, does not remain fixed for all time, but must conform to changing needs and circumstances’.

Relationship to the Water Code Rejecting arguments by certain parties in the contested case that the passage of the Water Code and its various provisions relating to permitted uses of water extinguished the public trust doctrine, the court maintained that the doctrine was rooted in common law and was clearly and unambiguously enunciated by a valid constitutional mandate in accordance with its framers’ intent. Concluding, the court asserted that ‘the Code and its implementing agency, the Commission, do not override the public trust doctrine or render it superfluous. Even with the enactment and any future development of the Code, the doctrine continues to inform the Code’s interpretation, define its permissible ‘outer limits,’ and justify its existence. To this end, although we regard the public trust and the Code as sharing similar core principles, we hold that the Code does not supplant the protections of the public trust doctrine.’

Scope and substance In a well-reasoned conclusion, the court declared that the scope of the public trust doctrine applied to all water resources of the state ‘unlimited by any surface-ground distinction’ because ‘both categories represent no more than a single integrated source of water with each element dependent upon the other for its existence’. It thus recognized the scientific fact of the existence of an integrated water resource system rather than a dichotomy between the hydrologically connected surface and ground water. With reference to the specific Waiahole Ditch system, the connection between dike-impounded groundwater and stream flow had been established by scientific and engineering studies that had shown that water diversions of groundwater had reduced the natural flows in several streams, even though consensus regarding the exact relationship between the two is still unresolved. This relationship was poorly understood previously when common law precedent established the legal framework of regulating the two components independently of each other, a situation that remains true in the vast majority of American states (whether governed by riparian or prior-appropriation water-right regimes), despite increasing attempts to bring groundwater within the reach of the public trust doctrine (Browning, 2011). Typical objections to the expansion of the doctrine to groundwater involve arguments relating to the taking of property rights and economic dislocations of long-standing business practices and conditions.

The substance of the public trust doctrine as applied in Hawaii was defined in terms of the protected purposes or uses of the water resources on the one hand and the concomitant duties and responsibilities of the state with regard to it on the other.

194

C. S. Papacostas / Water Policy 16 (2014) 184–196

Protected trust purposes. The court enumerated various protected uses as falling under the protection of the doctrine, beginning with the original provision of the Illinois Central decision of protecting the public rights to navigation, commerce and fishing. To these were later added ‘a wide range of recreational uses, including bathing, swimming, boating, and scenic viewing’ (Supreme Court, 2000) and domestic uses, particularly for drinking, as well as resource and ecological protection (including the protection of birds and marine life), and the preservation of scenic beauty. Consistently, with state law and precedent, it affirmed appurtenant rights (as defined earlier) and ‘Native Hawaiian and customary rights’ as constituting public trust purposes. Finally, although private uses for economic development could be permitted in balancing the benefits to be derived from the use of water, the court stopped short of declaring them as protected uses under the trust doctrine, reasoning that ‘if the public trust is to retain any meaning and effect, it must recognize enduring public rights in trust resources separate from, and superior to the prevailing private interests in the resources at any given time’ (Supreme Court, 2000).

Duties and responsibilities. Interpreting Hawaii legal precedent and the state constitution identified a dual mandate to the state’s responsibility flowing from the doctrine, specifically to protect the resources for the benefit of present and future generations and to develop the state’s water resources to the maximum reasonable and beneficial use. The two objectives are by their nature inconsistent, one emphasizing the conservation of resources, the other pursuing their utilization. Refusing to establish absolute priorities among broad categories of uses (whether protected or otherwise), the court charged the Water Commission, which it characterized as the guardian of public rights under the trust, with the responsibility to ‘weigh competing public and private water uses on a case-by-case basis, according to any appropriate standards provided by law’, nevertheless reserving to itself the ultimate authority to interpret the trust’s scope and extent. As one commentator observed, ‘The most expansive judicial interpretation of the public trust doctrine has occurred in Hawaii. Hawaii applies the public trust doctrine not only to traditional public water rights, like navigation, commerce, and fishing, but also to recreational uses, such as bathing, swimming, and boating … Hawaii held that the public trust doctrine applies to all water resources within the boundaries of the state without qualification; it is not limited to navigable waters or tributaries affecting navigable waters’ (Browning, 2011).

To emphasize that the balancing of beneficial uses does not in effect eviscerate the doctrine, the Hawaii Supreme Court admonished the Commission that ‘any balancing between public and private purposes begin with a presumption in favor of public use, access, and enjoyment’ (Supreme Court, 2000). Moreover, the Commission was empowered to revisit prior allocations of water to the various uses in response to changing conditions (such as water shortages) or scientific knowledge of the system’s dynamic behavior.

Continuing debate In a seminal article that followed the passage of the Water Code, MacDougal (1988) anticipated ‘contradictions and ambiguities which, for the most part, remain to be sorted out’. As the implementation of the public trust doctrine continues to evolve via administrative rule-making and case law, the debate continues to flare up in research publications and position papers offered by the proponents of often diametrically opposed precepts. Examples of these include the argument that beneficial uses and local land-use plans should be given priority over ‘contrived preservationalism’ and that depriving owners of rights they enjoyed earlier constitutes an illegal taking that merits just compensation (Callies &

C. S. Papacostas / Water Policy 16 (2014) 184–196

195

Chipchase, 2007), and calls for the primacy of customary and traditional rights of the indigenous population (Liu, 2002; Sylva, 2007). An environmental group has even called for the outright abolition of the Water Commission, claiming that its only purpose appears to have degenerated into giving ‘[stream] diverters years, even decades of water as the [stream restoration] challenges to their use drag through the commission’s unwieldy process’ (Anonymous, 2010). Between the extremes are specific entities that wish to preserve their use of water as they have become accustomed.

Conclusion In terms of its availability and quality, water is becoming a critical resource on a global scale, partly because of population growth and partly because of increased demands to support the rising expectations of the population for ever-improving quality of life.

Through a constitutional amendment and subsequent state Supreme Court affirmation, the State of Hawaii has adopted the most comprehensive interpretation of the public trust doctrine in the United States as applied to the balancing of the dual objectives of preserving its water resources (both surface and subsurface) for the benefit of the present and future generations while maximizing the development of reasonable public and private beneficial uses.

Hawaii represents a geographically isolated region, a microcosm that can accentuate the issues associated with anticipating the possibility of water shortages and attempting to develop a working framework based on the public trust doctrine in the face of established and competing practices and uses. Moreover, decisions concerning reasonable and beneficial water uses are expected to be based on scientific investigations that address physical, social, economic and environmental considerations. This is a tall order for physical and social researchers and scientists, and for water resource engineers, particularly in their attempts to quantify the minimum flows (in the case of streams) and maximum sustainable yields (in the case of groundwater resources) that optimize the dual responsibilities of preservation and resource development.

In its basic principle, the public trust framework being attempted in Hawaii is reminiscent of the konohiki practices of the ancient Hawaiians. Just like the konohiki, land division agents appointed by the Chiefs and given strict authority to regulate the use of irrigation water, the Commission on Water Resource Management in Hawaii is charged with the responsibility of attempting to strike a balance on a grander statewide scale and across a multiplicity of social, economic and environmental considerations.

Developing this parallel further, the ancient system operated under the strict but universally accepted authority of the konohiki who were required to possess technical water management skills but also leadership skills commanding the respect of the community to maintain social harmony. Whether the permit-based bureaucratic (administrative) framework evolving in Hawaii will prove to be as practicable and long-lasting as the ancient konohiki system remains to be seen!

References Anonymous (2010). The water commission: an idea whose time has passed. Environment Hawaii 21(1), 2–3.

Bader, H. R. (1992). Antaeus and public trust doctrine: a new approach to substantive environmental protection in the common law. Boston College Envtl. Aff. L. Rev. 19(4), 749–763.

Brown, C. N. (2006). Drinking from a deep well: the public trust doctrine and the western water law. Florida State University L. Rev. 34(1), 1–39.

196

C. S. Papacostas / Water Policy 16 (2014) 184–196

Browning, J. (2011). Unearthing subterranean water rights: the environmental law foundation’s efforts to extend California’s public trust doctrine. Environ. Law Policy J. 34(2), 231–246.

Callies, D. L. & Chipchase, C. G. (2007). Water regulation, land use and the environment. U. Haw. L. Rev. 49, 49–96.

Clark, J. C. (1986). Hawaii surface water law: an analysis of Robertson v. Ariyoshi. U. Haw. L. Rev. 8, 603–618.

Commission on Water Resource Management (1997). In the Matter of Water Use Applications, Petitions for Interim Flow Standards Amendments, and Petitions for Water Reservations for the Waiahole Ditch Combined Contested Cases Hearing.

Department of Land and Natural Resources, State of Hawaii.

Dole, S. B. (1892). Evolution of Hawaiian Land Tenures. Papers of the Hawaiian Historical Society, No. 3, Honolulu, HI. Hutchins, W. A. (1972). Water Rights Laws in the Nineteen Western States. United States Department of Agriculture, Washington, DC. Johnson, J. W. (2009). United States Water Law: An Introduction. CRC Press, Coca Raton, FL. Jowett, I. G. (1997). Instream flow methods: a comparison of approaches. Regulated Rivers: Res. Manage. 13, 115–127.

Kamakau, S. M. (1992). Ruling Chiefs of Hawaii: (Revised Edition). Kamehameha Schools Press, Honolulu, HI. Klass, A. B. & Huang, L. Y. (2009). Restoring the Trust: Water Resources and the Public Trust Doctrine, A Manual for Advocates. White Paper #908, Center for Progressive Reform, pp. 1–28.

Kluegel, C. H. (1916). Engineering Features of the Water Project of the Waiahole Water Company. Hawaiian Almanac and Annual for 1917, Thos. G. Thrum Publisher, Honolulu, HI, pp. 93–107.

Kosaki, R. (1978). Constitutions and constitutional conventions of Hawaii. Hawn J. Hist. 12, 120–138.

Kuykendall, R. S. (1938). The Hawaiian Kingdom: 1778–1854. Foundation and Transformation, Vol. 1. University of Hawaii Press, Honolulu, HI. Liu, S. A. K. (2002). Native Hawaiian homestead water reservation rights: providing good living conditions for native Hawaiian homesteaders. U. Haw. L. Rev. 25, 85–130.

Lucas, P. F. N. (1995). A Dictionary of Hawaiian Legal Land-Terms. Native Hawaiian Legal Corporation, University of Hawaii Press, Honolulu, HI. MacDougal, D. W. (1988). Testing the current: the water code and the regulation of Hawaii’s water resources. U. Haw. L. Rev. 10, 205–254.

Martin, E. P., Martin, D. L., Penn, D. C. & McCarty, J. E. (1996). Cultures in conflict in Hawaii: the law and politics of native Hawaiian water rights. U. Haw. L. Rev. 18, 71–197.

Monahan, E. T. & Yamauchi, H. (1972). Hawaii’s System of Water Rights: An Economic Evaluation. Technical Report No. 57, Water Resources Research Center, Honolulu, HI. Nakuina, E. M. (1893). Ancient Hawaiian Water Rights. Hawaiian Almanac and Annual for the Year 1894. Press Publishing Co. Steam Print, Honolulu, HI, pp. 79–84.

O’Shaughnessy, M. M. (1904). Irrigation in Hawaii. Hawaiian Planters’ Monthly 23(10), 419–425.

Owen, D. (2012). The mono lake case, the public trust doctrine, and the administrative state. UC Davis L. Rev. 45, 1099–1153.

Papacostas, C. S. (2001). Water Rights in Hawaii. Wiliki o Hawaii, 36(11), Hawaii Council of Engineering Societies, Honolulu, HI. Proctor, T. B. (2004). Erosion of riparian rights along Florida’s coast. J. of Land Use 20(1), 117–158.

Sai, D. K. (2008). The American Occupation of the Hawaiian Kingdom: Beginning the Transition from Occupied to Restored State. PhD Dissertation, University of Hawaii, Honolulu, HI. State of Hawaii (1987). The Water Code. Hawaii Revised Statutes, Chapter 174C. Supreme Court (2000). Appeal From the Commission on Water Resource Management (Case No. CCH-OA95–1). State of Hawaii.

Sylva, S. (2007). Indigenizing water law in the 21st century: Na Moku Aupuni o Ko’olau Hui, a native Hawaiian case study.

Cornell J. L. Pub. Pol’y, 16, 563–580.

Takemoto, R. (1986). Groundwater issues in Hawaii. U. Haw. L. Rev. 8, 513–568.

Utton, A. E. & Utton, J. (1999). The international law of minimum stream flows. Colorado J. Int’l Law Policy 10(1), 3–37.

Walston, R. E. (1982). The public trust doctrine in the water rights context: the wrong environmental remedy. Santa Clara L. Rev. 22(1), 63–93.

Wilcox, C. (1996). Sugar Water: Hawaii’s Plantation Ditches. University of Hawaii Press, Honolulu, HI. Williams, J. N. S. (1918). A Little Known Engineering Works in Hawaii. Hawaiian Almanac and Annual for 1919.

Thos. G. Thrum Publisher, Honolulu, HI, pp. 121–126.

Received 19 November 2012; accepted in revised form 7 August 2013. Available online 24 September 2013

Water Policy 16 (2014) 197–204

Discussion The remunicipalization of Paris’s water supply service: a successful reform1 Anne Le Strat Adjointe au Maire de Paris, chargée de l’eau, de l’assainissement, et de la gestion des canaux, Présidente d’Eau de Paris

The City of Paris decided to bring to a close a 25-year period during which the public water service was delegated to the private sector, and to transfer its management to a single public operator – Eau de Paris – as of January 1, 2010. In the homeland of the major water multinationals and the delegation of public services to the private sector, this historic decision of the authorities of the capital city was seen by many as a mini revolution in the quiet world of water. It has raised considerable attention among a large number of local authorities, associations, researchers and entities involved in water, and is still hotly debated.

For instance, in his article ‘Return of drinking water supply in Paris to public control’ published in Water Policy (Barraqué, 2012) a recognized expert in the world of water, examines the mechanisms of this reform, but in terms that are misleading, to put it mildly. The arguments put forward by the opponents of public management are given remarkable prominence in his account. Such a biased account almost amounts to disinformation, and requires a response.

Without reviewing all the inaccuracies contained in this article, especially when they do not directly concern Eau de Paris – like the percentage of the French population that is served by a public water utility (which is 35%, according to Desmars (2013), and not 28%, as asserted by Barraqué) – I would like to set the record straight on the facts about the origin, the process and the consequences of the Paris water reform2.

The return to public control: beyond a political choice, a decision of good management with significant gains in terms of organization, economy and assets ‘The return of Paris water supply service under public control is clearly a political decision which requires no particular justification’: This is how Bernard Barraqué sums up the remunicipalization of 1

Right of reply by Anne Le Strat, deputy Mayor of Paris, Responsible for water, sanitation and the management of canals, Chairwoman of Eau de Paris, to the article ‘Return of drinking water supply in Paris to public control’ by Bernard Barraqué (Water Policy 14(6), 2012, 903–914).

2 Other background documents used in the preparation of this article are listed in the Appendix (available online at http://www. iwaponline.com/wp/016/085.pdf) . doi: 10.2166/wp.2013.085 © IWA Publishing 2014

198

A. Le Strat / Water Policy 16 (2014) 197–204

water in Paris. But while the reform obviously resulted from a strong political and ideological choice, it was also dictated by arguments of sound management. Water is an essential collective good that requires a sustainable and solidarity-based management under clearly accountable public control. But it was also a pragmatic decision aimed at ending the failures and limitations of the previous system. The organization of the service was split up between three operators: a semi-private company, SEM (Société d’Economie Mixte), in charge of producing and transporting bulk water to Paris; and two private operators, subsidiaries of Suez and Veolia, who were in charge of the supply and billing. The control of the water quality supplied to Parisians was the responsibility of CRECEP, a municipal centre of research and expertise in water management.

There were serious problems in auditing and evaluating the service under this system. Several reports pointed to the opacity of the two delegation contracts, which were signed in 1985 for a period of 25 years without any call for competition or consultation with the Parisians. For example, the financial reports of the distributors did not provide a full description of the financial operations performed within the framework of these contracts (Regional Audit Office, 2000). The production cost presented by the private operators in their activity reports was 25 to 30% greater than the real management cost of the service per subscriber. For instance, the production cost of the service declared by the subsidiaries of Suez and Veolia amounted to €106 and €116, respectively, and the management costs to €75 and €87 (Service Public 2000, 2002). The guarantees for renewal, a sort of insurance provision for future works on the network, were disproportionate with respect to the real cost of the works (Audit Office, 2003). As far as the margins are concerned, they were underestimated in the declarations of the private operators and could triple according to whether the added value on the supply activity was included or not in the methods of calculation (Audit Office, 2003). Although it has always been very difficult to make an exact calculation of the financial windfall that the Paris water service represented for the private operators, analyses of their accounts have shown that the profit for suppliers (after adjusting expenses and including financial incomes), amounted to at least €22 million; and that the average amount of the working capital surplus between 2000 and 2006 was €60 million per year (Du Fau de Lamothe, 2007).

Between 1985 and 2001, the subsidiary of Veolia alone was able to pay dividends of €56 million to its shareholders, 15 times the initial value of their capital. Additional obscurity was created by the companies when they sub-contracted work to other subsidiaries of the groups, allowing the groups to make profits further down the production chain which did not appear in their accounts. These examples illustrate the total absence of financial transparency of the private operators that is a consequence of the financial model they had established for their own benefit.

It was therefore quite logical, following the change of political control in 2001, that the new municipality, led by Mayor Bertrand Delanoë, rapidly decided to completely rethink the Paris Water Service.

Contrary to what Bernard Barraqué asserts, the large number of City Council debates that took place on the occasion of the annual reports on the service clearly shows that the political criticism was essentially motivated by a total absence of control over the two suppliers and their financial manipulations, and was not simply aimed at bringing down the price of water for users.

In 2003, the City of Paris took advantage of the clause permitting a renegotiation during the course of the contract to renegotiate with the two private operators and enforce a tighter control of the contracts by the public authority. The municipality calculated the financial gains made by the companies on the guarantees for renewal. In other words, the City estimated the provision for works that the private operators should have normally used for asset renewal according to the terms of their contracts and obtained a commitment from them to undertake works on the network worth up to €153 million. This included,

A. Le Strat / Water Policy 16 (2014) 197–204

199

among other things, the replacement of all lead pipes by the end of the delegation contract in 2009, and the development of a system for remote reading of meters.

We will come back to this point later, but let us consider first the issue of the lead pipes, which, in many respects, symbolizes the lack of willingness of the private operators to completely fulfill their legal and contractual obligations. Both suppliers, though they announced the eradication of lead pipes in their activity reports, stated at the end of the contracts that the number of connections not replaced was somewhat over 400. In reality, the number of lead connections still to be replaced was three times higher (nearly 1500): the most complex and costly operations had not been made by the private operators and were passed on to the new operator.

During the 2008 municipal election campaign, Bertrand Delanoë promised that he would return to a totally public management of the water service if he were re-elected. The objective was to clarify responsibilities and costs and offer a service of quality at a controlled cost and with a long-term vision for the system. After his re-election as Mayor, the remunicipalization was confirmed. The end of the concession contracts being scheduled for 2009, the whole system had to be overhauled and the transition had to be carried through in a year and a half, which is an extremely short period for such a wide-ranging reform.

The creation of the single operator, Eau de Paris, an industrial and commercial public body, was voted by the Paris Council in November 2008. The transfer of the staff and assets of the semi-private company, and of part of the research, expertise and control centre, took place on May 1, 2009. The conditions of transfer for the supply activities and the staff of the two private operators were approved by the Paris Council in November 2009 and came into force on December 31 in the same year.

Many people thought that the commitment of the Mayor of Paris was a mere electoral promise. Until the end, the private companies thought that the Paris Municipality would never implement the reform.

When they realized this was not the case, they tried to make the best of it and asserted that the decision was only guided by political considerations (AquaFed, 2010), meaning campaign rhetoric that had nothing to do with the water service and the way in which it had been managed by the private companies.

The fact that Bernard Barraqué highlights this argument and considers it as the main lesson from his analysis on the Paris water reform is astonishing.

It is however clear that the transfer to a public operator has allowed the elimination of overlapping functions and activities that were frequent in the past, and to optimize the functioning of the service.

The synergies from pooling the production, supply and billing functions has enabled an increase in efficiency. It is now possible to monitor every drop of water from the catchment area right up to the tap. It is worth remembering that the creation of a single operator was also advocated by the private sector – but not in the form of a public structure.

The economic gains generated by the Paris reform range between €35 and 40 million per year, including the reintegration of the profits made by the private operators, lower operation costs and lower taxes. These gains are entirely reinvested into the water service. The operational savings include those made on the contracts for works on the distribution network. The private operators, during the whole course of their contracts, used the services of subcontractors that were part of their group for such works as maintenance and repair of pipes, renewal and renovation of water pipes, replacement of lead pipes and installation of automated meter reading systems. All these works were sub-contracted by mutual agreement, without competition on the technical options or the prices, to subsidiaries of these large groups. The cost of these contracts was then integrated into the expenses recorded by the service providers as part of the water service operation costs, which minimized the apparent margin of the private operators. In reality, these works generated a significant additional margin that indirectly went to the group of which the service provider was a subsidiary. Such works are now subject to public procurement procedures, and significant reductions in their costs have

200

A. Le Strat / Water Policy 16 (2014) 197–204

already been observed: 25 to 30% savings on unitary prices compared with those charged by the private operators, which is an annual gain of between €3 and 5 million.

In short, the economic gains did not result, contrary to what Bernard Barraqué says, from savings on the dividends paid to shareholders of SEM, because this never happened.

The delegation contracts provided a windfall for Suez and Veolia, since profits were made all along the chain of value: through the subcontracting of works, the quantification of costs of capital maintenance, the payment of wages to staff who were not assigned full time to the service in Paris, and the delay in payments for the water purchased from the semi-private company in charge of production. All these margins were consolidated and used to inflate the dividends of the shareholders of the two groups.

Concerning the maintenance of the network, the efficiency of which was so highly praised by some people, the private operators limited their action to the replacement of joints and pipes attachments. The technical teams of Eau de Paris considered that this approach, which suited a private operator during the course of the contract, was not sustainable over the long term. This is why the municipal service worked out a long-term assets management plan with an investment budget for the supply network that was higher than during the last years of the delegation.

In addition, the network performance considerably improved after the 2003 negotiations, as it was made even more interesting financially for the private operators to tackle leaks. The network yield rose by 10% between 2003 and 2007. The companies reduced their efforts from 2008 when they found out their contracts would not be renewed. As a result, fewer leaks were repaired and the network efficiency started to decline.

The way the network yield was calculated also accounted for its increase after 2003. The suppliers used the volumes of water billed annually and not the volumes of water actually consumed to obtain this figure.

Moreover private operators would bill volumes consumed a given year the year after. This allowed them to control to a certain extent the yield they wanted to publish. Some years they would even lower its value on purpose for financial reasons. The 2003 riders to their contract stipulated that they had to compensate the SEM if the network performance improved, as bulk water sales would then drop. This financial compensation was proportional to the increase in the network efficiency.

As far as the meters are concerned, they were effectively fitted out with remote metering systems by the private operators. But the Paris authority has been obliged to buy them all back at an excessive price – €17.6 million – but the equipment had been defined as ‘compensatable returnable assets’, which means that the municipality was entitled to recover the equipment at any time provided it compensated the private operators for the value that was not yet amortized at the date of the transfer. Moreover, the equipment installed by the private operators proved to be neither compatible with each other, nor usable without resorting to the computer system and staff of Suez and Veolia. In short, Eau de Paris has developed its own information and operation system, operational since 2011, that covers the management of customers, billing, the management of the meters, remote metering and consumption data, the online agency, the management of technical interventions, the management of operations on the network, and reporting. The municipal service now has a system at its disposal that corresponds to its needs and over which it has a full control. This has allowed it to significantly modernize the functioning of the water service.

A three-fold control of the service: political, municipal and by citizens Bernard Barraqué underlines that monitoring water services is always difficult, irrespective of the management method. This was particularly true in Paris during the period of delegation to the private

A. Le Strat / Water Policy 16 (2014) 197–204

201

sector. The lack of any control and power of the local authority over its service was well known. The municipality had entrusted SEM to monitor the private suppliers who were themselves administrators of SEM! There was an obvious conflict of interest. Several reports had pointed to this conflict (Regional Audit Office 2000; City of Paris General Inspection, 2003): hence the decision taken by the Paris Council in December 2003 to transfer the monitoring of the private water suppliers to the City administration.

This was followed in March 2007 by a vote by the Council of Paris – not by the Mayor, Bertrand Delanoë – that approved the withdrawal of the private operators from the capital of SEM and their replacement by the Caisse des Dépôts et Consignations.

Like many other water authorities and semi-public companies – particularly SEM from 2008 to 2010 – Eau de Paris is chaired by the deputy Mayor in charge of water. The deputy Mayor in charge of finance is also responsible for monitoring Eau de Paris, against the objectives set by the municipality, as well as the results achieved.

Bernard Barraqué’s criticism concerning an excessive political control is difficult to understand and he seems to repeat the arguments put forward by the private groups.

Then again, owing to his expertise as a member of the Paris Water Observatory, he could have mentioned the monitoring system that was implemented with the return to municipal management. This extra-municipal committee, made up of a wide range of players from various backgrounds – social housing agencies, associations of tenants and users, environmental NGOs, trade unions, neighborhood councillors, researchers, etc. – is entirely focused on water and sanitation. It is chaired by a non-City of Paris person. It allows every citizen to be informed about all the important resolutions that concern the water service, and express an opinion prior to their study by the Council of Paris.

This monitoring by citizens supplements the monitoring by the municipality, which is based on a performance agreement signed with Eau de Paris for a period of five years – a true innovation in the French water sector. This document specifies the powers of the local authority and sets objectives to be reached by the public utility. It includes a large number of technical, financial, social, and assets indicators, and is a genuine tool of control at the service of the Paris administration. In fact, this agreement, which is regularly assessed by the city’s technical and financial services, is submitted every year to the Paris Council and the Paris Water Observatory.

The control over the public utility is also ensured through its board, which includes civil society representatives among its members. Two delegates from associations and a member of the Paris Water Observatory sit on the board with voting rights. There are also elected representatives of Paris, who are appointed by proportional representation of the political groups at the Paris Council; representatives of the staff; and other qualified people. Far from being symbolic, the presence of the civil society at the board has significantly changed the proceedings of board meetings. It also compels the public utility to give a clearer account of its management and to be more transparent.

Drop in the price of water for all users At the time of the reform, the first commitment made by the Mayor was to stabilize the price of water in current euros during his term in office, that is to say until 2015, in spite of a continuous drop in operating revenues because of lower water consumption by Parisians, due to a lower consumption of domestic appliances (dish-washers and washing machines) and the shift of industrial activities from Paris to the suburbs.

Furthermore, Parisians have become more vigilant with their water usages. It has been possible to go beyond this objective and reduce the price of water for all users by 8% as of July 1, 2011, thanks to the

202

A. Le Strat / Water Policy 16 (2014) 197–204

economic gains from the return of the service to municipal management. Thus, Parisians will have benefitted from a €76 million saving during the whole course of the performance agreement.

These decisions have put an end to the soaring of prices during the previous 25 years, with a 174% increase in the price of drinking water between 1985 and 2009. In his article, Bernard Barraqué stresses that the share going to the private operators increased slightly less than the share going to SEM. This assertion is incorrect because he fails to mention that the largest part of the investments made on the water supply system – production, transportation and network until 2003 – was borne by the semi-public company. The need to modernize equipment and the evolution of drinking water standards led SEM to make massive investments throughout the duration of the concession. In 1980, the water service was in a risky situation since two drinking water production plants in the Paris region, at Ivry-sur-Seine and Joinville-le-Pont, were showing signs of obsolescence. SEM renovated these plants – from 1988 to 1993 for Ivry-sur-Seine and from 1993 to 1998 for Joinville-le-Pont. It also modernized the Orly plant in 1992. During the same period, it renovated the control centre and started renovation works on the aqueducts after working out a monitoring system for the pipes.

Three pumping stations for underground water were also modernized. Then, in 2002, it started the construction of four underground water treatment plants for a total amount of €150 million. In Paris, it has improved the safety of the water supply by creating two inter-reservoir connections. During these 20 years of activity, SEM has also undertaken the automation of installations which ended the system based on three eight-hour shifts.

The price increase applied by the private operators was directly computed according to a particularly rewarding price indexation formula, which provided for compensation in case of lower revenues, that is to say in case of lower bulk water purchases due, in particular, to the reduction of the leakage rate in the network.

Nearly 80% of the fixed expenses were borne by SEM alone, while the most profitable part of the service was held by the private operators.

All users are at the core of the new Paris water service Restoring the place of users at the core of the service has been one of the pillars of the reform. How can Bernard Barraqué ignore it? Beyond the opening of the board of Eau de Paris to the civil society and the involvement of the Paris Water Observatory in the service operation, the return to municipal control of the management of customers has allowed Eau de Paris to re-establish links with the users.

The internalization of this relationship with users and subscribers was implemented after a unanimous vote of the board of Eau de Paris – including the municipal opposition. This new service, which has proved to be less costly than when it was entrusted to private operators, has been rewarded with the award for ‘Best Customer Service of the Year’ in 2013 and 2014. The new services it offers – better monitoring of consumption, more transparent information on the functioning of the service, more attention to consumers’ expectations – have re-established direct contact with users.

The public service now guarantees access to water for everybody Making the right to water effective in any circumstances is another strong commitment made at the time of the reform of the water public service in Paris that Bernard Barraqué totally fails to mention.

As soon as the return to a city operated-system was effective, the municipality implemented, with the support of Eau de Paris, a whole range of social measures targeted to the most underprivileged. This started with a water solidarity allowance to help households encountering difficulties in paying their

A. Le Strat / Water Policy 16 (2014) 197–204

203

water bills. This was the first initiative of its kind in France. For households who have unpaid bills, a support fund is available to which Eau de Paris contributes up to €500,000, which represents a contribution almost three times higher than that of the private operators.

Eau de Paris has gone even further by developing partnerships with Paris social housing agencies to install water-saving kits in order to bring down water and energy bills for the most underprivileged.

Homeless people are not forgotten in this package of measures: the installation of water fountains in the city that provide permanent and free access to water; the distribution of jerry cans, tumblers, and flasks to the social organizations who are in contact with these populations; the water supply maintained in squats; etc, all these actions contribute to the implementation of the right to water for all, for which Paris is considered as one of the leading French local authorities (Smets, 2011).

The remunicipalization of water as an example The remunicipalization of water in Paris is obviously a crack in the commercial façade of French water multinationals, even if Bernard Barraqué does not recognize it by claiming that ‘This change, however, was not as dramatic as presented in the media.’ This was explicitly recognized by the representatives of Suez and Veolia: the loss of the Paris market, of course, had a detrimental effect on them from a financial point of view but its impact was even more negative on their image. In the entire world, these companies always presented the management of water in Paris as a model.

Has the Paris experience rocked the French water sector to its core? Can it bring other French local authorities in its wake? France is historically the heartland of the delegation of water services to the private sector, which has permitted groups such as Suez and Veolia to gain the stature they now have. The supporters of public management have so far been a minority but the fact that Paris returned to public management has inevitably added force to their arguments. Many delegation contracts are due for renewal and renegotiation in the coming years. More and more local authorities are thinking about returning to municipal management because this system seems appropriate to all the water services, even in complex situations where large investments are needed. The cities of Rouen and Brest have already returned to a city operated system, others, such as Bordeaux, Rennes and Nice, have announced it.

Of course, a number of local authorities have not yet made the decision, but they have used the threat of a remunicipalization to renegotiate existing contracts and, in particular, to obtain lower prices and a better control by the organizing authority. Why then does Bernard Barraqué not recognize the impact of the Paris reform about which nobody disagrees, not even its opponents?

According to Bernard Barraqué, the remunicipalization only ‘opened a Pandora’s box of water production and reorganisation at the regional level’. For an equivalent quality of water and service, the price of water in Paris remains much lower than that charged by other services in the Paris urban area.

The wish of other local authorities to get their water from Eau de Paris facilities is therefore perfectly legitimate. For instance, the Communauté d’Agglomération des Lacs de l’Essonne (CALE), to take only this example cited by Bernard Barraqué, signed a cooperation agreement with the Paris public utility in June 2012. According to this partnership, Eau de Paris will provide water to the distribution service recently created by the Communauté d’Agglomération at a price of €0.42/m3, taxes not included.

This kind of partnership allows greater efficiency in the production of drinking water in a context where consumption continues to diminish, without impacting the safe water supply in the capital. It also enables the pooling of production facilities, which is consistent with the wish of the City of

204

A. Le Strat / Water Policy 16 (2014) 197–204

Paris to preserve water resources over the long term. The closure of the Ivry-sur-Seine water treatment plant in 2010 reflects this concern to save water, and the continuous drop in water consumption – minus 28% in 20 years – and has absolutely not been decided after a conflict with the local authorities of the sector, as implied by Bernard Barraqué!

Conclusion Today, the water market is not as stable as it used to be. The economic parameters have changed, the democratic demands are stronger and strategies for the sector are evolving. The case of Paris typifies these changes in many respects.

Through its decision to remunicipalize its water supply service and to create a single public operator, the City of Paris has succeeded in offering Parisians a service close to its users, transparent and taking fully into account sustainability issues (environmental, social and assets management).

References AquaFed, International Federation of Private Water Operators (2010). The Water Operator’s World AquaFed Summer Newsletter.

Audit Office (2003). La gestion des services publics d’eau et d’assainissement – Cour des comptes, décembre 2003.

Barraqué, B. (2012). Return of drinking water supply in Paris to public control. Water Policy 14(6), 903–914.

City of Paris General Inspection (2003). Le contrôle par la Ville de Paris de sa filière eau (production – distribution)- Inspection Générale de la Ville de Paris, June 2003.

Desmars, M. (2013). Modes de gestion des services d’eau et d’assainissement en France – Paris. Paper presented at the France Eau Publique conference, Paris, 22 March 2013.

Du Fau de Lamothe, P. (2007). Observations sur les résultats des distributeurs de l’eau potable et non potable et leur contrôle par la ville.

Regional Audit Office (2000). Lettre d’observations définitives du 07/09/2000 sur la gestion de la production et de la distribution de l’eau potable et non potable à Paris. Chambre Régionale des Comptes d’Ile-de-France.

Service Public 2000 (2002). Gestion du service commercial des eaux de Paris.

Smets, H. (2011). La mise en œuvre du droit à l’eau: les solutions à Paris. Johanet, Paris.

Discussion Water Policy welcomes short formal and informed responses to published articles. When submitted, responses will be reviewed by the Editorial Board and shared with original authors and published according to recommendations of editors. The Editors feel that responses can enrich understanding of important and timely water policy topics (Editor in Chief).