State-civil society partnerships for improving safe water and sanitation coverage in the Northern region of Ghana: An exploratory qualitative study

Abstract State-civil society (CS) partnerships are increasingly common in the water and sanitation (W&S) sector of many low- and middle-income countries (LMICs) as a means of expanding service coverage for poor and marginalised communities. However, there has been limited research regarding the drivers, nature, successes and challenges of such partnerships. Findings from in-depth qualitative interviews of 24 participants from 16 different CS-state partnerships in the Northern region of Ghana highlight: 1) the value to both civil society organisations (CSOs) and local governments in increasing access to safe drinking W&S facilities for poor communities through pooling of resources, knowledge and expertise; 2) the distinctive role of CSOs including, reducing service inequities and facilitation of a more effective participation of poor and marginalised people in service delivery decisions; 3) three different forms of partnerships that can be harnessed for effective service delivery, comprising of project, network and strategic collaborations and 4) significant challenges, including funding, programming, legitimacy and sustainability issues that need to be overcome to enhance the success and contributions of these partnerships. The article suggests need for longitudinal and quantitative studies to assess the sustainability and impact of these partnerships on intended beneficiaries.

background spans the field of sociology, health policy and governance. His research interests include the organisation and management of public services and inter-organisational relationships involving non-governmental organisations, public sector entities and international donors. His research focuses on healthcare and other social services and makes connections between organisational and medical sociology and public management. His recent work has been published in BMC Health Services Research and BMC Public Health. E-mail: mhushie@uds. edu.gh

PUBLIC INTEREST STATEMENT
The past few years have witnessed an increased number of state-civil society partnerships in the water and sanitation sector of many developing countries as a means of organising service delivery for poor and marginalised communities. This article explores how such partnerships emerge, their nature, advantages and challenges. Findings from a qualitative study of 16 state-civil society collaborations in the Northern region of Ghana highlight how the involvement of civil society organisations contributes to increased access to services in ways that best leverage the capacities and resources of both partners. The study also identifies different models of partnerships that can be harnessed for more effective service delivery as well as challenges that must be addressed for more effective functioning of these partnerships. The findings contribute to the literature on cross-sector partnerships, by shedding more light on how such partnerships can lead to improved local service delivery.

Introduction
In the endless search for more efficient and effective methods of delivering social services, the past few decades have seen an increasing number of state-civil society (CS) partnerships in many low-and middle-income countries (LMICs) as a means of expanding safe drinking water and sanitation (W&S) coverage and improving the health of poor communities (Haque, 2004;Harvey & Reed, 2004Jiménez & Pérez-Foguet, 2010;Sansom, 2006).
The forces behind recent interest in partnership building in the W&S sector are probably numerous. However, in most sub-Saharan African countries such as Ghana, deteriorating economic conditions and monetary crises, characterised by huge public sector deficits, growing external indebtedness and balance of trade problems since the 1980s have been cited as the most significant factors driving reform of public administration systems in these countries (Bangura & Larbi, 2006;Batley & Larbi, 2004;Brinkerhoff & Brinkerhoff, 2015;World Bank, 1997). International development agencies such as the World Bank, International Monetary Fund and Asian Development Bank promoted such reforms in the form of privatisation and decentralisation policies, which entailed reallocation of resources, delegation of power and responsibility for service provision from the central government to local authorities. Increasingly, these polices became popular in many countries based on the premise that private sector participation, including CS involvement in the W&S sector would result in improved service coverage, quality and efficiency to supplement if not replace the inefficiencies of conventional public sector provision (Davis, 2005;Hall & Lobina, 2007;Herrera & Post, 2014;Prasad, 2006;Tukahirwa, Mol, & Oosterveer, 2010World Bank, 2004). To date, however, little research has focussed on the drivers, nature, barriers and facilitators of CS involvement and how their roles could be more effectively harnessed to increase access to services for poor and marginalised people.
In Ghana, W&S provisioning has traditionally been supply-driven, with a centralised para-statal institution-the Ghana Water and Sewerage Corporation (now renamed the Ghana Water Company Limited), in charge of service delivery and management of infrastructure in urban and rural areas. Following global trends, the government has since 1994 been implementing a series of public sector reforms in an attempt to become more "decentralized"-arguably with the aim of reducing the state's role and promoting private sector participation in the delivery of services (Trend Group, 2003). Particularly, the reforms have been intended to create conditions (through legal, business and regulatory interventions) to facilitate a favourable environment for increased CS participation in order to accelerate the coverage of the rural population with good drinking W&S facilities. Progressively, these reforms have culminated in the institutional re-alignment of key actors in the sector, in which an increasing number of civil society organisations (CSOs), including non-governmental organisations (NGOs), community-based organisations (CBOs) and faith-based organisations (FBOs) have become involved in service provision, collaborating with local governments to expand access to services (Chan & Effah, 2013;Fuest & Haffner, 2007;Government of Ghana [GoG], 2007;Whitfield, 2006).
Following reforms, Ghana has made relative progress, particularly towards achieving the Millennium Development Goal target of halving the proportion of population without sustainable access to safe drinking water and improving sanitation coverage. Thus, nearly 80% of the population now have access to an improved source of drinking water, while the proportion of the population using improved and unshared sanitation facilities (toilets) increased by 15% between 2006 and 2011. Despite this, however, gross disparities and inequities still exist, especially in rural areas and among the poorest groups in the three northern regions of the country (Northern, Upper East and Upper West), where access gaps in improved water sources are widening with 70% of households having no toilet facility (GSS, 2007(GSS, , 2010Osei-Assibey, 2014;United Nations Children Fund [UNICEF], 2014). Consequently, the morbidity pattern in these regions reveal a high prevalence of water-borne and poor environment-related diseases, particularly among women and children.
A major concern for W&S policymakers and implementers has, thus, focused on how to create an effective interface among key stakeholders, including, especially, CSOs with a view to integrating and harnessing their roles to improve service delivery in a context where increasing coverage of services remains a continuing challenge (GoG, 2009(GoG, , 2010Water and Sanitation Monitoring Platform [WSMP], 2009). However, there is limited understanding of how CSOs and local governments form inter-sectoral relationships, and how their interface enable or constrain the quest for more effective organisational arrangements for providing safe W&S services. Thus while, Ghana's W&S sector policies provide a relatively well-established institutional framework with clear lines of responsibility and roles for government and CSOs to engage in meaningful partnership working, not much is known about the organisational and delivery mechanisms being used and how CSOs can best work with local governments to improve services. This is in spite of the widespread belief that CSOs should assume important responsibilities in the provision and financing of such services in an era of increasing fiscal stress and inefficiency of government provision (Robinson & White, 1997). The effective involvement of CSOs in the W&S sector can contribute towards more equitable access to services for marginalised communities.

Conceptualising partnerships between the state and CSOs
Despite state-civil society partnerships' growing prominence in the W&S sector, the terms "state" "civil-society" and "partnership" are under debate. To better understand the role of these partnerships in accelerating the coverage of W&S facilities in the Northern region of Ghana, workable definitions for the terms are needed. For the purposes of this study, the following definitions are used.
The term "state" can be used to describe all public sector agencies, regulatory bodies and other entities which are established, run and financed by the Government to deliver public programmes, goods or services. This definition implies that the ownership and control of such public sector organisations lies in the hands of public authorities that do not have profit-making as primary motive as opposed to private entities. Such government organisations may exist at the national, regional and local levels (Lane, 2000). In this research, decentralised state entities, which make up the local government machinery including the metropolitan, municipal and district assembliesusually referred to as the District Assemblies (DAs) in Ghana, are conceptualised as the main collaborators of CSOs in service delivery. However, given the wider privatisation focus of reforms, it was also conceived that CSOs by design might have working relations with private sector (business) entities.
The term "civil society" refers generally to voluntary, non-profit organisations established to serve the shared interests, purposes and values of individuals, groups and communities, often seen as mediating interactions between the state, market and family. The arena of civil society encompasses a range of actors and organisational forms that vary in the degree of their formality, autonomy, power and scope of activities. These include: NGOs, CBOs, FBOs, women's organisations, professional associations, trade unions, self-help groups, social movements, business associations, networks and coalitions, lobbying and advocacy groups (Salamon, 2010).
Finally, there is no single definition of a "partnership" and the term is typically used interchangeably with collaboration, coalition and joint working. For this study, a partnership was defined as cross-sector strategic relationships that form between two or more actors who develop joint rules and structures to govern their relationships. The parties interact through formal and informal processes to achieve a shared vision and collective goals, acknowledge a high level of mutual interdependence, commit resources and mutually share in the risks and benefits of pooling resources (Cropper, 1996;Gray, 1989;Thomson, Perry, & Miller, 2007). Partnerships can afford participating organisations access to new knowledge, complementary skills, resources and technologies to provide a wider range of products/services that each organisation, working alone, could deliver easily (Powell, 1987).
This study contributes to the literature on the implementation and impact of privatisation and decentralisation reforms in LMICs that have seen increasing collaboration between CSOs and local governments to increase access to W&S services (Soublière & Cloutier, 2015) from two perspectives. First, although the salutary impact of decentralisation reforms on increasing access to services is widely acknowledged, the actual processes of CS involvement, the nature and principles of their collaboration and the role of context in influencing changes, remains poorly understood (Aslam & Yilmaz, 2011;Smoke, 2015a). This article aims to fill these gaps in the knowledge by examining case studies of state-CS collaborations in the Northern region of Ghana, where such partnerships have proliferated and, thus, offers a good setting for exploring the benefits such collaborations can bring. Finally, the article sheds light on the problems and weaknesses of the implementation of decentralisation reforms in the W&S sector in practice, with further civil society involvement (Chowns, 2015;McLoughlin, 2011;Smoke, 2015b). Apart from updating the literature on current initiatives in the sector, the Ghanaian experience can also serve as a lesson to other countries that are seeking to use partnership working as a means of providing more equitable services in deprived communities.
This study's aim was to explore the fundamental idea of why and how local governments are engaging CSOs in the W&S sector and the advantages and disadvantages they experience in their relationships. The specific research questions that were qualitatively explored are as follows: 1) what are the drivers and nature of partnerships developing between CSOs and local governments in the W&S sector? 2) what advantages accrue to CSOs and local governments working in partnership? and 3) what challenges do CSOs and local governments face in managing the sector in the Northern region of Ghana?

Study design
This was an exploratory qualitative study, drawing on multiple interviews to provide in-depth understanding of the drivers, nature, successes and challenges of emerging CS-state linkages in the W&S sector in the Northern region of Ghana.

Study setting
Ghana is a West African country, sharing borders with la Côte d'Ivoire to the west, Burkina Faso to the north, Togo to the east and the Gulf of Guinea to the south. The Northern region of Ghana, where this study took place forms part of the Northern Savannah Zone (including Upper East and Upper West), covering an area of 70,383 km 2. It is divided into 26 DAs with a total population of 2,479,461, consisting of 17,287,49 and 750,712 rural and urban populations, respectively. Given the persistent poverty and underdevelopment trends prevailing in these three regions (World Bank, 2011aBank, , 2011b, NGOs have flourished here over the years to promote development more generally, including the provision of W&S services.

Sampling
Case selection was purposive with the intent of selecting different CSOs and their collaboration with government that might be considered "revelatory" in the sense that they represent a widespread but generally inaccessible or unexplored phenomenon in line with this study's objectives (Stake, 1995;Yin, 1994). A convenience sample based on determining whether an identified CSO was actively engaged in collaboration with any government entity to provide water or sanitation services was used as the major criterion for selecting specific partnerships and participants. The focal person of either the CSO or local government was then identified and interviewed, depending on his/her willingness to participate. Once interviewed, such a participant assisted with identifying additional partnerships and participants. In all, 21 CSOs were contacted but only 16 participated. The remaining five could not be included because the CSO representatives of the partnerships were not available at the time of the interviews.

Data collection
The data sources were comprised of in-depth interviews, organisational documents and publicly available information on the Internet, which were triangulated to enhance the reliability and external validity of the findings. In total, 24 interviews were conducted including 21 participants who were representatives or frontline staff of included CSOs and DAs and 3 donors with titles such as Program/Project Manager/Coordinator/Officer, Regional Manager, District Water and Sanitation Team (DWST) Leader and Environmental Health Officer, across case study sites. Data saturation (i.e. where no new perspectives on the subject were being elicited) was achieved after conducting 21 interviews and this was confirmed with the last three study participants.

Data analysis
Interviews were conducted in English, audio recorded and transcribed verbatim. An interpretive approach that emphasises constructing meaning within the context of individual case study partnerships was adopted. This was with the prospect that each partnership will provide highly generalisable findings regarding the provision of W&S services that is embedded in a wider process of CS-government interactions. The researcher read and re-read interview transcripts and organisational documents to gain familiarity with their content. A constant comparative analysis was done to identify common and divergent themes using codes that were developed to provide a basis for categorising and analysing data in relation to the key research questions. A sample of four transcripts (two from DAs and two from CSOs) were analysed by an independent researcher and all discrepancies in analyses between researchers were resolved.

Findings from the partnerships
The reported increasing involvement of CSOs in Ghana's W&S sector in response to privatisation and decentralisation reforms, has led to significant changes in the institutional arrangements for the delivery of services. The partnerships studied are presented based on the study's research questions: the drivers; nature (including the contribution of partners, management of the partnerships and their types); and their successes and challenges. Participants' anonymity is preserved by referring to them generally as "CS" or "DA" (i.e. state) participants and by their interview locations.

History and policy context of CSOs' engagement
Evidence gathered from organisational documents and interviews revealed that one of the main drivers of CSOs' engagement in the W&S sector was the passage of the national legislative instrument on decentralisation in 1994, which led to the creation and empowerment of DAs as the key institutions responsible for managing the provision of social services, including W&S, health and education. In these sectors, the trend over the years has been towards promoting a decentralised, multi-sectoral, demand-driven, private (business) and, importantly, non-profit sectororiented service delivery systems that provide reliable, affordable and sustainable services for poor people. Particularly in the W&S sector, the main antecedents for CSOs' involvement has included four principles: 1) increasing the participation of local NGOs and private (business) sector in service delivery; 2) fostering public sector facilitation of service delivery (rather than service provider); 3) promotion of demand-driven services and 4) advancing the role of poor and vulnerable people in the ownership, management and decision-making processes. Hence, as part of reforms, a National Community Water and Sanitation Programme in rural areas was launched in I994, alongside the establishment of a Community Water and Sanitation Agency (CWSA) as the implementing arm of this programme. The CWSA is mandated with the role of coordinating the work of a range of actors in the rural W&S sector, including NGOs, public sector organisations, local beneficiary communities, private sector organisations and donors (Trend Group, 2003: World Bank, 2011c. The governments' decentralisation policy, thus, provided the initial impetus and institutional framework, with authorities necessary for the interaction between local governments and CSOs, determining their new roles and responsibilities in service delivery. Subsequently, CSOs' involvement has been shaped by severe budgetary constraints that most DAs face country-wide, as decentralisation has proceeded. Although the DAs benefit from the District Assemblies Common Fund, which is a transfer grant from the central government to the DAs for development purposes, these funds tend to be inadequate and ineffectively used (GoG, 2014). Moreover, while certain sectors such as health and education operate with substantial government funding devolved from their respective line-ministries, the rural water supply and sanitation sector remains continually under-resourced. Facing such fiscal constraints, local governments have no other option but to collaborate with CSOs and development partners in order to meet the W&S needs of their constituencies. Thus, during the time of this study, several NGOsboth local and international can work in any given district collaborating with the DAs, touching on a wide range of developmental issues, including W&S, funded by a variety of multilateral and bilateral donors and international agencies (e.g. World Bank, Canadian International Development Agency, German Development Agency).
Although the reasons for the DAs' choosing to collaborate with NGOs varied across the included partnerships, this was often determined by CSOs spearheading such collaborations with external donor support; the presence of a cooperative local government; and CSOs ability to make impact in un-served or underserved communities. While no CS participant identified any limiting criteria for providing specific W&S services, most acknowledged that it was the prime responsibility of the government to meet the W&S needs of its people. However, as CSOs, they provide, support and critical services that the government for lack of resources or political will cannot or will not provide. Moreover, given the short-term and donor-funded nature of CS interventions, most CS participants viewed their collaboration with the DAs as a means of maximising use of scarce donor funds, avoiding service duplication as well as fostering local ownership of services for sustainability. Concerning this, one CS participant stated: The District Assembly is the landlord of the district; hence it is important that whatever intervention we carry out is captured by them. Secondly, we look at the sustainability of our projects because as an NGO, we have funding limitations, projects come and go for one, two or three years. . .and so definitely we are sure of withdrawing one day, then who takes over? (CS participant, Tamale).
Similarly, most DA participants acknowledged how W&S services' provision is considered the government's primary obligation and cited lack of resources as one of the main reasons for collaborating with CSOs as indicated by one participant: Our mandate is to provide social services to the people. But due to inadequate resources, we are unable to meet our targets. . .they (CSOs) come with their resources to help us meet our targets. . .if we build 2 boreholes and they build 2 or 3 more, that is a laudable idea (DA participant, Yendi).
Thus, although both CS and DA partners cited varied underlying reasons for working in collaboration, these reasons amount to the DAs' lack of legitimacy and capacity; on the one hand, and the relative power and control that CSOs have over the resources to achieve their common goals on the other.

Contribution of the partners
Regarding partners' roles and contributions to the collaboration, both DA and CS participants cited efforts to increase access to safe W&S services, which addresses as well, cross-cutting developmental issues, such as health, education, environment, poverty and sustainable development. However, each CSO often addressed different cross-cutting issues based on their focus of activity and areas of expertise. Concerning this, one CS participant stated: For us our work is strategically tied to schools and its part of addressing our first strategic objective which is promotion and protection of child rights, using education as a tool. . .When I talk about rights it is about their ability to access basic things like food, water, shelter, sanitation. So that is why most of our work in the area of W&S is kind of channeled through schools . . .education is the bigger tool (CS Participant, Tamale).
Moreover, in recognition of the failure of most W&S projects to adequately bring about the adoption of hygiene practices or the use and sustainability of facilities, most NGOs tended to focus substantially on the "software" elements of W&S initiatives. This usually entails community engagement, social and behavioural change interventions, including hygiene/health education that focus on solving water, sanitation and waste-related hygiene problems; as well as providing training/capacity building for W&S management teams (WSMTs) for effective management of facilities. Using these software models of service delivery, NGOs add goals of providing adequate "hardware" (technical) component of services, securing funding for their delivery, building technical capacities to monitor and maintain the operation of the service and also, the capacity to raise revenues to cover, at least, the operation and maintenance costs of facilities. The hardware elements include: basic/mechanised boreholes; solar energy or electricpowered hand-dug wells; rainwater harvesting technologies; household and institutional latrines.
Apart from providing innovative services, some CSOs engage strategically in advocacy work. This often involves mobilising and empowering marginalised communities with limited access to safe W&S services to articulate their needs and exert pressure on the DAs to provide needed services. This is often achieved by facilitating the formation of Community Water Advocacy Teams (CWATs), networks and platforms that engage with local authorities to demand accountability for improved service provision. Moreover, advocacy-oriented CSOs may seek to promote dialogue among duty bearers (i.e. DAs) to replicate proven cost-effective W&S technologies such as protected hand-dug wells, low-cost rope-pump wells, sand dams, rainwater harvesting systems and ventilated improved pits.
In contrast, local governments are generally responsible for the implementation of national level W&S policies and also mostly for service provision. In particular, the DAs' contribution comprise of: 1) human resources [e.g. DWSTs, Environmental Health Officers, District Development Planners, District Chief Executives (DCEs)]; 2) partial funding and mobilising funds or labour from community members (usually 5% of facility cost); 3) facilitation of community entry and mobilisation for action which includes, forming and training of WSMTs and involving these teams throughout the project cycle to ensures service sustainability; 4) providing the land and most appropriate location for siting facilities and 5) information exchange at workshops and provision of action plans developed from community baseline surveys on service needs (e.g. hand-dug wells, boreholes; household/ community toilets; borehole, water quality testing and drilled and abandoned boreholes). Alternatively, the local communities can apply for services through the Assemblymen or Chiefs. Such information is usually fed into the DA's work plans, which are shared with NGOs or development partners for funding and co-implementation.
Moreover, private sector actors play important roles as CSO contractors and/or supply chain organisations, while education and training institutions such as Universities are also important in driving research and innovation across technical and soft system components of W&S services. At the local level, the communities are the critical stakeholders and end-users often active in the design, construction and maintenance of W&S systems. Consequently, the W&S sector in this region is characterised by multiple actors with different roles to play in providing and maintaining services.

Managing the partnerships
Regarding how the partnerships are managed, a heterogeneity of inter-organisational relations involving CSOs, donors, local governments and other private organisations were found. Across the diversity, relationships can be usefully categorised into three broad types: a) "project"; b) "network" and c) "strategic" partnerships ( Figure 1). These arrangements are, however, not consistently applied as relationships may fit into more than one category at the same or different times and can evolve, depending on how the various actors carry out their roles, available funding and the context of a specific project. In a project partnership, an external donor or international NGO can directly contract a local NGO or CBO to implement a specific small-scale/short-term project with specific objectives to be achieved over the life of the project, working with the DAs and communities. Regarding the network partnership, a large international or local NGO can procure funding from an external donor, disburse such funds to single locally based organisations/groups or networks/coalitions of smaller local NGOs, CBOs and private actors to implement specific shortor long-term projects/programmes directly, collaborating with the DAs. The large international or local NGO funder of the network partnership does not engage in the design and direct implementation of projects. Such funders only coordinate and build the capacities of the smaller NGOs/CBOs and other organisations to design and implement projects, collaborating with the DAs and communities. Finally, the strategic partnership variant has a large international NGO obtaining funding from an external donor and acts as a sub-funder engaging in the design and direct implementation of a long-term project in collaboration with the DAs and communities with the aim of achieving sustainable impact on software and hardware elements of W&S services.
Prior to commencing any activity, CSOs would usually negotiate and sign a tripartite Memorandum of Understanding (MoU) involving the DAs and communities over specified time periods. The MoU comes with a full budget for proposed projects and specifies each party's roles, responsibilities, benefits and level of cooperation that CSOs expect from the DAs and communities during project implementation. However, on a day-to-day basis, their interactions are managed through informal consultation processes in which a CSO can contact the top hierarchy of the DAs by phone, text or WhatsApp messages or in person to discuss issues. In these relationships, being open about one's motives, lines of communication, roles and responsibilities, budget lines and intended contributions were viewed by both CS and DA participants as the bases for building trustworthy relationships needed for successful partnership working.
The process of initiating service provision is normally preceded by CSOs registering at an NGO desk established within the DAs where potential project proposals (usually a quota of W&S facilities to be provided over a specific period of time) are lodged for review, approval and presentation to donors for funding. Once funding is secured, DA and CSO partners together with other stakeholders meet with community members, to explain the processes and procedures involved in the project, including how communities can benefit from the project, as well as their  expected contribution to the investment, operation and maintenance costs. Ultimately, project implementation is untaken by CS field project coordinators and DA personnel, under the latter's supervision.

a) Project Partnerships b) Network Partnerships c) Strategic Partnerships
Established WSMTs, consisting of elected community members are responsible for ensuring that W&S systems remain functional and provide long-lasting benefit to their users. The teams determine appropriate tariff levels, considering those who can or cannot pay and need to be exempted such as the elderly, children and disabled added to determining methods of tariff collectionwhether monthly or pay-as-you-use or fetch depending on the type of facility in consultation with community members and the Chiefs. These mechanisms allow the poorest individuals and communities to have access to services without suffering financial hardship. The DAs also build the capacity of the WSMTs in areas such as book-keeping, fund-raising and management, including opening a bank account and behaviour change and communication strategies. Moreover, WSMTs are expected to invest a portion of the tariff for ongoing rehabilitation and coverage expansion. In addition, the DAs are responsible for providing post-construction support-both technical and managerial to the WSMTs.
Moreover, programme performance is assessed through interface forums and information sharing and feedback sessions that both parties organise periodically for community members, DAs and development partners. As well, CSOs usually have field workers and monitoring teams that supervise on-going projects on a regular basis, added to engaging external evaluators to appraise programme performance together with the DAs, DWSTs, EHDs and community representatives.

Successes of the partnerships
Regarding the broader successes of the partnerships, most CS participants suggested that collaborating with the DAs allows CSOs to procure credibility and legitimacy-two key qualities needed to attract resources for their work and to survive. About this, one CS participant stated: . . .for the Assembly recognizing that we have the capacity and expertise and that they need to work hand-in-hand with us has actually increased our visibility, credibility and image with donors who are always willing to fund our activities (CS participant, Tamale).
Most CS participants also emphasised the important role of the DAs in facilitating community entry, identifying the most deprived communities and supporting with requisite human resources and funding for implementing projects as significant partnering advantages. Additionally, both parties cited: 1) the joint planning and implementation of their activities; 2) information sharing, including annual reports and budgets and 3) joint periodic review and stakeholders' meetings, either organised by CSOs or DAs as contributory factors to their successes.
Overall, given that CSOs tend to be more proactive than the DAs in initiating these partnerships, most recounted several CSOs' strengths that enable successful working with the DAs. These include: 1) the commitment and honesty of CSOs' staff in investing funds for their intended purposes without misappropriating such funds; 2) adoption of proven national and international standards for delivering products and services; 3) ability to attract external donor funding; 4) decision to work through the DAs rather than independently in ways that minimises service duplication; 5) the integrated nature of their interventions and 6) established transparent and cordial relationships with the DAs.

Challenges of the partnerships
Concerning challenges, most CS participants cited lack of harmonisation in their individual work programmes, especially with the DAs often unable to keep their jointly agreed programme targets, deadlines and milestones and scheduled meetings as significant collaboration pitfalls.
In addition, most CS participants reported the DAs' reluctance to fully disclose their budget lines and the expectation that NGOs, given their numerous and varying funding sources, can and should fully fund their jointly planned projects as critical challenges. Similarly, some DA participants perceived NGOs generally as lacking transparency in their financial dealings with local government partners. This was viewed as a significant source of distrust that adversely affects the former's willingness to commit requisite funds and human resources for their work. Additionally, there were concerns that some DA staff seemed willing to undertake certain activities, such as community visits for project monitoring and evaluation, only if they received additional incentives such as per diem, fuel and transport.
Moreover, as a challenge, some DA participants tended to view CSOs' advocacy-oriented and community-empowering work as inciting community members' ill-sentiments against their jurisdiction, which often raises tension and suspicion among the parties. Such actions include public forums, which CSOs organise to track the DAs' budgets and expenditures on W&S services to promote greater transparency and accountability in service delivery.
Most CS participants also highlighted recent cut-backs in external donor funding as a major constraint. This had resulted in many NGOs, especially the local ones, winding up or shifting their strategic missions. Moreover, there has been a tendency for international NGOs to remain vibrant in terms of funding and service delivery, despite local NGO perceptions of shrinking external donor funding. This again, has tended to portray how ineffective local CSOs were across the sector, with adverse implications for their legitimacy and survival.
Additionally, some CS participants regarded the tendency for some NGOs to concentrate work in some regions, districts and communities to the disadvantage of others or duplicate their work in the same areas as added challenges. Both CS and DA participants also reported that they are hamstrung by the existence of fake CSOs whom they viewed as sources of disrepute and threat to legitimacy for credible organisations working to improve services in the sector.
Also, most CS participants reported experiencing political interference as a major barrier. This was eminent during local/national electoral campaigns, when some DAs would covertly claim sole ownership of their jointly implemented projects or request for services as means of leveraging votes from community members. Moreover, most CS participants viewed the practice in this country, where appointments to the top hierarchy of the DAs would normally be made through nominations by the governing political party rather than through local-level elections as detrimental to their efforts. As it were, such appointees tend to focus on serving the interests of their appointing authority with much less interest in addressing the development needs of their constituents. Moreover, NGOs encountered challenges with the recurrent turnover of such appointees, particularly with electoral changes in the political leadership. This often results in temporary or permanent discontinuation of programmesas vital project documents and information; established work capacities and trust relationships needed for more effective programming are lost.
Concerning the long-term sustainability of provided services, many CS and DA participants reported about how newly delivered W&S services often perform effectively for a period, and then either fall into disrepair or otherwise fail to provide continuing benefits to their users for various reasons. These include: 1) dormant WSMTs, as a result of the DAs inability to provide effective post-construction support for lack of funds and logistics; 2) late reporting of major breakdowns with W&S systems to the DAs for timely action; 3) trained mechanics abandoning their volunteering duties for paid work elsewhere; 4) local community expectation that NGO or DA partners should continue to bear post-construction repair and maintenance costs due to most WSMTs' inability to collect and save adequate funds for such purposes; 5) the tendency for some NGOs to implement programmes without consulting the DAs and 6) a new local government policy directive that favours the construction of household latrines, instead of public ones, following CLTS implementation guidelines.

Discussion
This article has provided an overview of on-going state-CS partnerships for improving W&S coverage in the Northern region of Ghana, highlighting the influence of context in determining how these partnerships emerge, their nature, advantages and challenges. It started from the belief, based on the theoretical and empirical arguments in the public management and governance literature in favour of decentralisation reforms in LMICs that CS involvement in the W&S can improve access to services at the local community level. Emerging themes highlight that the willingness of local governments and CSOs to work with external donor support deploying decentralised cooperative mechanisms and policies results in more equitable, effective, efficient and sustainable access to services for marginalised communities. As an invaluable model for organising W&S systems in rural areas, state-CS partnerships allow the two parties to pool resources and combine the managerial and technical skills of both to expand service coverage than working independently. Moreover, this study has identified three distinctive forms that emerging state-civil society partnerships in the W&S can take, consisting of project, network and strategic partnerships.
It has also highlighted the distinctive role that CSOs are playing in extending W&S services to poor communities and by extension improving the health of local citizens. These roles can be broadly grouped into seven categories: 1) facilitation of a more effective participation of poor and marginalised people throughout the project cycle, i.e. needs assessment, planning, implementation, monitoring and evaluation of W&S management; 2) promoting the political participation of local people in decision-making processes through their advocacy work; 3) community education, awareness-raising, sanitation and hygiene promotion; marketing including stimulating demand, supply chain strengthening and implementing behaviour change programmes; 4) research and innovation including demonstration and piloting of innovative best practices/know-how and locally adapted approaches and technologies among local governments that contributes to greater ownership, accountability and sustainability; 5) fostering collaboration and promoting networking between different sector actors including donors, local governments, local CBOs, private organisations, education and research organisations for improved service provision; 6) facilitating the translation and implementation of national W&S policies at the local level and 7) capacity building for local governments, CBO networks, WSMTs and end users (households and communities) that together, contribute to greater sustainability of implemented initiatives.
Additionally, this study has identified challenges associated with these partnerships, including: 1) lack of synchronisation in their individual work programmes; 2) inability of local government authorities to keep their jointly agreed programme timelines; 3) distrust, suspicion and lack of transparency about each other's financial contributions; 4) local political interference; 5) high attrition rate of local government political leadership; 6) reduced external donor funding; 6) presence of quack CSOs and 7) programme identification, design, implementation and sustainability concerns that affect the effective functioning of these partnerships.

Limitations
This study suffers limitations. First, the data and interpretation of this study are heavily based on the experiences of state, CS and donor participants, because these actors have key roles in influencing access to services for marginalised communities. Including data from other relevant stakeholders such as private sector organisations, education and research institutions often engaged by CSOs in service delivery might have led to other explanations and findings. Moreover, despite efforts made, this study was unable to provide any quantitative data on the number of W&S infrastructure supported by NGO and DA partners, as there were no effective local or regional W&S monitoring, reporting systems or database in place. This made data obtained from participating organisations inconsistent and unreliable to report on. Finally, given the exploratory nature of this study, it is difficult to assess whether or not CS-state partnerships are sustainable modes of governance in the W&S sector over time. Evidently, further research within a longitudinal and quantitative framework is needed to study the impact of these partnerships on intended beneficiaries.

Conclusions
With the emergence and flourishing of various forms of CS involvement in the W&S sector in the Northern region of Ghana, the state's traditional structure for W&S governance has changed considerably. This article has provided evidence of the contribution of these new partnerships to increased access to funding and investment, and especially of improved service provision for disadvantaged communities. This way, the original goals of the Ghanaian government to engage CSOs in W&S provisioning as part of decentralisation reforms are being realised. Moreover, this study has identified three different partnership forms in use: project, network and strategic models; which can serve as a valuable resource for international donors and NGOs seeking to expand W&S services to poor communities around the globe.
Additionally, this article has unveiled several obstacles that limit the ability of CS and DA partners to maximise their resources and will to improve the W&S needs of underserved populations, including funding, programming, legitimacy and sustainability concerns. These challenges may, however, be viewed as organisational and context-specific that can be addressed through new polices and mechanisms to more profitably engage CSOs in service provision. Concerning, funding in particular, most CS participants were of the view that resourcing the sector need not be dependent on international sources, as is the current situation and, therefore, advocate for increased government funding to make these partnerships effective models for local service delivery. Others also suggested the need for the government to rid the NGO sector of quacks, added to empowering credible CSOs to serve as the DAs' watchdogs, monitoring and tracking their activities for improved resource allocation and service delivery. Finally, although the involvement of CSOs in the W&S sector allow for improved service provision, it is crucial to accelerate the establishment of systematic and comprehensive local government coordination, monitoring, reporting and data collection systems that efficiently integrate the work of CSOs and various actors, as the current ad hoc, fragmented and diverse system endangers effective programming. Of necessity, both CS and DA participants attested to how the absence of such monitoring and reporting systems constituted significant barriers to reducing service overlaps and duplication in the sector. Consequently, improving such systems will enable the DAs to evaluate the impact of CSOs' programmes and target available resources to achieve maximum results.