Research, part of a special feature on State-Reinforced Transformative and Adaptive Governance of Sustainable Social-Ecological Systems

INTRODUCTION

The state’s role in managing the irrigation commons remains complicated, especially in sub-Saharan Africa (SSA). Since the 1980s and 1990s, decentralized irrigation schemes have increasingly received the support of governments in SSA, including legislative reforms that devolve water management authority and responsibility to water users, who are mostly farmers (Senanayake et al. 2015, Cambaza et al. 2020). Many decentralized irrigation reforms are financed by the World Bank and the International Monetary Fund (IMF). Known as Irrigation Management Transfer (IMT) or Participatory Irrigation Management (PIM), these reform initiatives aim to achieve full cost recovery, improved irrigation performance, and inclusive participation of water users (Groenfeldt and Svendsen 2000, Aarnoudse et al. 2018). Whether these decentralized reforms have lived up to expectations (and why) still presents theoretical and empirical debates in the literature.

A bit of historical context is needed before proceeding further. Some credit the 1970s as when countries began transferring or decentralizing water management functions to farmer users or cooperatives in response to failures or underperformance of large, state-led irrigation schemes (Turral 1995, Mollinga et al. 2007). These decentralization reforms ranged from fully privatizing/transferring all water management responsibilities to users (IMT) or sharing water management responsibilities between individual users (or their associations) and other public and private entities (PIM); both of these reform types are discussed extensively in the literature (Groenfeldt and Svendsen 2000, van Vuren et al. 2004, Suhardiman and Giordano 2014). Although their implementation varied by context, the primary goals of these reforms included improving irrigation infrastructure, establishing water users associations (WUAs) as essential vehicles for improved water delivery, supporting operations and maintenance, institutionalizing participatory and democratic decision making, and ensuring greater financial autonomy of WUAs by introducing irrigation service fees (Johnson 1997, Suhardiman and Giordano 2014).

In sub-Saharan Africa, these decentralized irrigation reforms occurred as part of a broader series of institutional and economic reform processes whose goals included minimizing both the responsibilities of and spending by African states in their provision of a range of public services. In the 1980s, West African countries such as Burkina Faso, Mali, and Senegal became early pioneers of WUAs and other decentralized irrigation management systems in their efforts to achieve such reductions (Wellens et al. 2013, Aarnoudse et al. 2018). Exemplifying the reform zeitgeist of the period, these decentralized irrigation systems are the product of a range of neoliberal economic and institutional reform policies, such as the structural adjustment programs encouraged (sometimes mandated) in the 1980s and 1990s by the IMF and the World Bank (Riddell 1992, Konadu-Agyemang 2000). The implementation and outcomes of these decentralized irrigation reforms leave unanswered the question of how their institutional design supports or reinforces water users’ capacities to self-govern and cooperate in managing their irrigation systems.

We address this question conceptually and empirically by examining ongoing decentralized irrigation management reforms in Ghana. These systems often exhibit closed-access common-pool resource (CPR) characteristics, with highly subtractable resource units (irrigation water) and low-excludability of resource users (e.g., farmers), often leading to excessive appropriation of the irrigation water and difficulty in getting resource users to contribute to the operation and maintenance of the irrigation system (Ostrom 1990, Ostrom et al. 1994). The institutional design of decentralized governance arrangements for these irrigation systems remains a critical factor in addressing these challenges of over-appropriation and unwillingness to contribute to the provision of irrigation systems.

In Ghana and other parts of sub-Saharan Africa, the states have played crucial roles in designing these decentralized irrigation institutions, which can be understood as “the set of working rules or ‘rules-in-use’ for supplying and using water in a particular location” (Ostrom 1992). The Government of Ghana, with the help of the World Bank, enacted new legislation in 2016 to decentralize irrigation schemes. The new Irrigation Water Users’ Association Regulations (L.I. 2230) restructures the existing large, centralized irrigation farmers’ associations within government-owned irrigation schemes into decentralized WUAs. Before 2016, a government-owned scheme might typically have a single farmers’ association serving all farmers, but the legislation reorganizes this centralized organization into a federated system with WUAs as the lowest decentralized unit; the units can organize themselves into a federation of irrigation water users’ association.

With this article we contribute to the literature by offering conceptual and methodological analyses of how and why the institutional design of these irrigation reforms matters to our diagnosis of their successes or otherwise. Specifically, it assesses the institutional design of Ghana’s current decentralized irrigation reform by testing and drawing insights from the state-reinforced self-organization (SRSG) framework (Sarker 2013, 2014, DeCaro et al. 2025) to understand how these irrigation reforms provide the needed capacities (institutional designs) for WUAs and other stakeholders to self-govern and cooperate.

We engage the literature on decentralized irrigation systems in sub-Saharan Africa and situate the findings from previous studies within discussions around the role of the state in the institutional design of decentralized governance of natural resources, focusing on the state-reinforced self-governance (SRSG) framework (Sarker 2013, 2014, DeCaro et al. 2025), an extension of the institutional analysis and development (IAD) and the social-ecological systems (SES) frameworks (Schlager and Cox 2018, McGinnis 2011, McGinnis and Ostrom 2014, Ostrom 2005, Weible 2023). In particular, the SRSG framework suggests institutional design principles (or design principles) for self-governance and cooperation (DeCaro et al. 2025). The language of design principles is borrowed from Elinor Ostrom as the defining characteristic(s) that help to account for the success of institutions “in sustaining the physical works and gaining the compliance of generations of users to the rules in use” (Ostrom 1993:1907, Cox et al. 2010). Insights from these design principles are then tested in an empirical case in Ghana for conceptual clarity and empirical insights into the capacities to self-govern and cooperate around state-sanctioned decentralized irrigation reforms.

THE ROLE OF (AFRICAN) STATES IN DECENTRALIZED IRRIGATION INSTITUTIONS: THEORETICAL AND EMPIRICAL PERSPECTIVES

The drive toward decentralized irrigation institutions in sub-Saharan Africa has been met with frameworks to critically evaluate the performance of these reforms. For instance, Aarnoudse et al. (2018) adapted the SES framework (Ostrom 2007, McGinnis and Ostrom 2014) to distill four main factors (socio-political context, water resource system, governance, and users) influencing the performance of decentralized irrigation schemes in Senegal, Ghana, Burkina Faso, Niger, Mali, Tanzania, Kenya, Ethiopia, Malawi, Zimbabwe, and South Africa. For these scholars, weak public irrigation agencies, small-scale and climate-vulnerable irrigation systems, top-down decentralized irrigation institutions, and smallholder farmers with low access to water and minimal experience with irrigation result in the poor performance of decentralized irrigation schemes routinely found in sub-Saharan Africa (Aarnoudse et al. 2018).

Similarly, Hassan et al. (2014) employed the institutional decomposition and analysis (IDA) framework (Blomquist et al. 2005, 2008, Dinar et al. 2007), which draws insights from literature on decentralized natural resource management (Ostrom 1990, 1992, 1993, Easter 2000, Agrawal 2001, Agrawal and Ostrom 2001). Here, the IDA schema was adapted and used to evaluate decentralized initiatives within water basins in South Africa, Mozambique, and Zimbabwe focusing on five parameters: contextual factor and initial conditions, characteristics of the basin’s decentralization process, characteristics of central government and basin-level relationships and capacities, internal configurations of basin-level institutional arrangements, and attributes of the basin. Their analysis revealed instances of bottom-up and top-down approaches to decentralization within the water basins and limited financial resources and autonomy for decentralized water entities, such as WUAs (Hassan et al. 2014), which are also documented in other studies (e.g., Brown 2011, Van Den Broek and Brown 2015, Aarnoudse et al. 2018). More importantly, the IDA analysis concludes that decentralized water initiatives in sub-Saharan Africa, such as those happening around irrigation schemes, represent de-concentration rather than devolution, partly because central governments have transferred neither decision-making authority and responsibility, nor financial resources to WUAs and other lower level organizations (Hassan et al. 2014).

This article contributes to the literature by offering conceptual and methodological analyses of how and why the institutional design of these irrigation reforms matters to our diagnosis of their successes or otherwise. With few exceptions (e.g., Baldwin et al. 2016, McCord et al. 2017), the cases of irrigation reforms in sub-Saharan Africa are yet to illuminate why in-depth considerations of the reforms’ institutional designs could highlight issues of local autonomy, (water) resource use efficiency, and adaptation to climate variability. For instance, by looking at the polycentric design features of Kenya’s 2002 Water Reform, some scholars concluded insightfully that the reform was deliberately designed by the national government to offer autonomy, communication, coordination, and shared decision making across multiple scales of local, regional, and national authorities (Baldwin et al. 2016). However, empirical analysis of Kenya’s polycentric reform offers some caution by noting that the reform only allowed (not encouraged) water users within 25 selected community water projects (CWPs) to experiment with rules internal to the affairs of the CWPs, and also found limited evidence of improved local conditions from such a state-led, top-down polycentric irrigation reform (McCord et al. 2017).

This study builds on some of these institutional design-based analyses in two ways. It does so by testing and distilling insights from the SRSG framework (Sarker 2013, 2014) to analyze whether and how these irrigation reforms (1) provide the needed capacities for WUAs and other stakeholders to self-govern and cooperate and (2) leverage the states/central governments’ powers and resources to reinforce the capacities of WUAs to self-govern their irrigation commons. The SRSG framework leverages the analytical strengths of the Ostrom School’s IAD and SES frameworks (Ostrom 2007, McGinnis and Ostrom 2014, Schlager and Cox 2018) to carefully map how users (e.g., WUAs) negotiate institutional arrangements to self-govern the commons, especially when such users interact within coercive government authority and privatization. Conceptually, the SRSG further sketches out the Ostrom design principles by adding and explicitly accounting for the role of states as an attribute in the IAD framework and a key variable of the (irrigation) governance systems in the SES framework. Sarker’s (2014) contribution suggests that states do not always have to resort to their coercive powers in addressing the tragedy of the commons. Rather, states, “can work with natural resource users and yet furnish strategic, non-participatory, and cooperative competence to reinforce their self-governing institutional arrangements for averting the tragedy ... [this] can be termed ‘state-reinforced self-governance’ ” (Sarker 2013:727). Empirically, Japan’s irrigation water case illustrates how an explicit account of the state in the self-governance discourse sheds insights on how the state may leverage its political, technological, and financial capacities to strengthen water users’ self-governing institutional arrangements, rather than coercing users or forcefully owning the irrigation commons (Sarker 2013, 2014).

The conceptual and empirical significance of Sarker’s explicit account of the state in self-governing irrigation institutions allows for nuanced analysis of, and conversations around, the realities of decentralized WUAs in sub-Saharan Africa. As noted above, donor partners play a significant role in decentralized irrigation reforms in African countries, including steering policy reforms more toward participatory rather than centralized governance schemes. Besides providing funding, what roles and capacities do these donor entities bring to the table regarding these reforms? Similarly, because of the colonial legacies and authoritarian military regimes that have ruled some African countries, these decentralized reforms are challenged by issues such as the genuineness of central governments to truly devolve authority to lower levels (Ayee 1997, Olowu et al. 2004, Riedl 2014, Frimpong Boamah 2018), mistrust and asymmetric power and information between and among centralized and decentralized entities (Awortwi 2011, Paller 2015), and the dichotomy between statutory institutions (rules-in-form) and indigenous/customary institutions(rules-in-use) especially around irrigation water management (Nkonya 2006, Aarnoudse et al. 2018, van Koppen and Schreiner 2019).

The co-management and de/recentralization of natural resource literature also highlights some insights about the states’ role(s) in natural resource management. For instance, the government’s involvement of farmers in co-managing irrigation improvement programs improved farmers’ perceptions of reliability, flexibility, and equity of transferred water services (Salman et al. 2023), and the government’s infrastructure investments in improved water adequacy and equity of water distribution (Frey et al. 2016). Some have also highlighted the government’s role in building participatory and deliberative processes, helping (re)frame issues (e.g., defining problems, causes, and proposing solutions), and raising awareness as crucial for local collective action around water issues (Villamayor-Tomas et al. 2019). In the case of forest resources, we are reminded that state-sanctioned decentralization or recentralization efforts may generate negative impacts if they are not grounded in community-centered, transparent, accountable, and reliable procedural measures (Bullinger and Haug 2012). For some, state-sanctioned decentralization or recentralization strategies are riddled with power dynamics, including moving power from one jurisdictional level to another, often leading to regulatory vacuums, delays, and elite capture of resources (Sahide et al. 2016). In the absence of such community-centered, participatory, transparent, and accountable procedural measures, state-sanctioned de/re-centralization of natural resource management in developing countries risk falling into the conclusion that “decentralization of forest-related decision making have not yet taken root, let alone bore fruit ... central governments, regardless of official rhetoric, policy, and legislation, erect imaginative obstacles in the path of decentralized institutions and choices” (Ribot et al. 2006:1881).

For these realities concerning the complicated roles of states [and their donor counterparts], nuanced conversations are needed around self-governing institutions, such as decentralized WUAs. Along these lines, DeCaro et al. (2017) and DeCaro et al. (2025) synthesized Sarker’s conceptual lens (Sarker 2013, 2014) with Ostrom’s IAD and SES frameworks (McGinnis and Ostrom 2014, Schlager and Cox 2018, Weible 2023) into a set of design principles to critically probe self-governing institutions around the capacities granted to users of a natural resource to self-govern and cooperate. This article tests and distills insights from these synthesized design principles (DeCaro et al. 2017, 2025) in drawing relevant connections to the realities of decentralized WUAs in sub-Saharan Africa.

Capacities granted to natural resource users (e.g., WUAs) to self-govern and cooperate

DeCaro et al. (2017) posited design principles supporting capacities for self-governance and for cooperation. The basic design principles for self-governance are: authority, responsibility, balanced operational resources, and balanced mechanisms for flexibility and stability. Authority refers to whether actors (e.g., users of natural resources) are sufficiently empowered to make and implement decisions in carrying out their functions, particularly related to addressing their key challenges (social-ecological dilemmas). The authority principle signals the internal capacity of users to make and implement decisions (semi)autonomously, a foundational requirement for actors to govern themselves (Ostrom 1990, 2010, DeCaro and Stokes 2008). For instance, WUAs in South Africa are vested with powers to curb unlawful water use and decide and implement water and environmental conservation measures (Madigele 2018). However, some scholars observed that proposals for decentralized CPRs often grant no more than operational-level powers of being authorized users to the beneficiaries of these proposals while central government actors hold the significant operational-level and collective choice rights and authorities (Agrawal and Ostrom 2001). For farmers in Malawi, the formation of WUAs was perceived to have further diminished their decision-making authority, resulting in forms of farmers’ resistance (e.g., refusing to pay operation and maintenance fees) in the ongoing decentralized irrigation reforms (Veldwisch et al. 2009).

Whereas authority grants powers to decide or act, responsibility relates to the conferred or acquired duties or obligations of actors pertaining to their mission and aims/goals (ergo the common refrain that being given responsibility for outcomes but not having authority to decide and act in pursuit of outcomes is a set-up for failure). Some scholars note that the imposition of bylaws on WUAs and a reliance on state irrigation agencies to establish and supervise WUAs often provides responsibilities to WUAs, especially related to the operations and maintenance of irrigation infrastructure, but with limited/restrictive decision-making powers for these WUAs (Suhardiman and Giordano 2014, Yami 2013). The misaligned authorities and responsibilities granted to WUAs partly informed why some characterize WUAs in sub-Saharan Africa as being partial, forced, or deconcentrated (but not decentralized) irrigation institutions because governments fail to devolve decision-making authority (Aarnoudse et al. 2018), which impacts the ability of water users to evolve and adapt their mental models in response to their water situations (van Rooyen et al. 2020).

The principles of balanced operational resources and balanced mechanisms for flexibility and stability complement the earlier principles of authority and responsibility. Balanced operational resources consider whether actors (e.g., WUAs) have a sufficient mix of financial and non-financial resources from internal (e.g., user fees, fines, membership dues) and external sources (e.g., grants, loans, etc.). Being resource-sufficient is central to the self-sufficiency of actors in carrying out their core authorities and responsibilities (Agrawal and Ostrom 2001, Andersson and Ostrom 2008). In sub-Saharan Africa, heavy reliance on external resources by WUAs may compromise the authority and autonomy capacities of water users, such as in the case of Nigeria and other contexts where WUAs bylaws are designed to fit the terms of government and donor partners because meeting such requirements are often tied to WUAs receiving external funding for initial operations (Ferguson and Mulwafu 2007, Vandersypen et al. 2007, Kemerink et al. 2013, Aarnoudse et al. 2018). Conversely, heavy reliance on internal resources, especially for WUAs in low-resource contexts, may incapacitate WUAs in performing their responsibilities. For instance, most WUAs in sub-Saharan Africa do not collect enough in user fees, including documented instances of low user fees being insufficient to cover the operation and maintenance costs of managing irrigation schemes in Senegal (Le Gal et al. 2003), and farmers refusing to pay irrigation maintenance fees in Malawi (Veldwisch et al. 2009), Tanzania (Richards 2019), and Zimbabwe (Mutambara and Munodawafa 2014).

The principle of balanced mechanisms for flexibility and stability gauges the extent to which decisions or actions can be changed, altered, or updated by actors through mechanisms such as narrowly specified and rigid or broadly framed and flexible rules, procedures, and performance standards. Rigid or narrowly defined ways of framing social-ecological dilemmas, decisions, rules, procedures, solutions, and resources may constrain actors’ ability to self-govern, experiment, or adapt to their changing contexts (Cosens and Chaffin 2016, Ostrom 2009). Enforcing rigid land tenure (e.g., changing land ownership), bylaws, and irrigation scheduling, cropping calendars and types of crops, monitoring, and sanctioning are all identified as constraining factors for WUAs in sub-Saharan Africa, sometimes leading to low occupancy/participation levels in irrigation schemes, and reduced learning and experimentation among farmers (van Rooyen et al. 2020, Frimpong Boamah and Nyamekye 2021). However, having some level of stability also matters in ensuring predictability, long-term planning, commitment, and collaboration among actors. For instance, some argue that actors are more likely to invest in averting the tragedy of the commons when there is tenure security (Doss and Meinzen-Dick 2015).

STUDY CONTEXT, DATA, AND METHODS

Study context

The Kpong Irrigation Scheme (KIS) started operation in 1998, irrigating about 2125 hectares of paddy rice farms and 1140 hectares of banana farms (Ministry of Food and Agriculture 2019). This second-largest irrigation infrastructure in Ghana (following the Tono Irrigation Scheme) lies within two communities: Asutsuare in the Shai-Osudoku district of the Greater Accra Region and Akuse in the Lower Manya Krobo district of the Eastern Region. It consists of (1) irrigational canals (main canal, branch and lateral canals), (2) drainage systems (main drainage, branch and lateral drainages), and (3) access roads. There are three separate irrigation areas within the KIS: Sections A, B, and C (see Fig. 1). Sections A and B are gravity-fed irrigation servicing irrigated paddy rice farms, and Section C is serviced by a re-lift pumping station and storage reservoirs for the drip irrigation banana farms whose harvests are mostly exported to European countries.

Institutionally, the KIS comprised two main governing entities: the KIS office of the Ghana Irrigation Development Authority (GIDA), and the farmers’ cooperative group, Osudoku Agriculture Cooperative Society (OACS). The KIS Office served as the scheme management entity (SME) having authority and responsibility over the construction, operation, and maintenance of the main irrigation canal, drainage and access roads. The OACS was mainly tasked with financing loans to farmers and operating the branch and lateral canals, drainages and access roads. This institutional arrangement had numerous social-ecological dilemmas, including unreliable water supply due to blocked/destroyed sections of the main, branch, and lateral canals, delays in getting GIDA to repair and maintain main canals, OACS’s difficulty in raising revenue to maintain and restore the branch and lateral canals and drainages, poor leadership, disputes among farmers due to different cropping and water schedules and lack of communication, low crop yields, challenges in accessing ready market, and high cost of milling and storage (Ministry of Food and Agriculture 2019).

The KIS and other government-owned irrigation schemes transitioned to a new arrangement in 2016 in a stated attempt to improve water supply and management. Through the support of the World Bank under the Ghana Commercial Agricultural Project (GCAP), new legislation, L.I. 2230 (WUA Law) was enacted, mandating a new institutional arrangement for government-owned irrigation schemes. The L.I. mandates that each scheme be managed by three entities (see Fig. 1). The first is a private sector company contracted by GIDA at each irrigation scheme to serve as the Scheme Management Entity (SME) responsible for (a) operating and maintaining the main canal, drainage and access roads, and (b) supplying water to each WUA based on contractual agreements. The second is constituting WUAs, each comprising 15 or more farmers, responsible for (a) operating and maintaining the branch and lateral canals, drainages and access roads, and (b) collecting Irrigation Service Charges (ISC) and appropriate fees from its members. The third is designating GIDA’s offices at the various schemes as the Supervisory Authority responsible for (1) forming the WUAs and providing administration, accounting, and general management support for the WUAs, including assisting in drafting bylaws for the WUAs, and (2) supervising the SME and the WUAs, including conducting periodic audits of the WUAs.

The 2016 L.I. ushered in a series of institutional changes at the KIS. The OACS was dissolved in 2016 and was to be replaced by 15 WUAs through a collaboration among the Government of Ghana, the World Bank, and the Japanese International Cooperation Agency (JICA) under the Project for Enhancing Market-Based Agriculture by Smallholders and Private-Sector linkages in Kpong Irrigation Scheme (MASAPS-KIS). Of the 15 WUAs, two were formed and registered in 2018 under the MASAPS-KIS as a pilot program (after the introduction of the WUAs law), four were established in 2019, and nine were created post-2019. Close to 2800 farmers operate under this new WUA arrangement, with each WUA comprising service areas of between 59 ha to 298 ha on average, with each having between 60 to 310 farmers on average. The farm sizes are relatively small, with approximately 50% of farmers having less than one hectare of farmland, 40% with one to two hectares, and 10% with more than two hectares of farmland for crop (mostly rice) cultivation (Ministry of Food and Agriculture 2019). These WUAs are expected to manage, maintain, and operate the water at the branch and lateral levels of the canal, collect ISC and other appropriate fees, hold general assembly meetings for all WUA members, settle disputes, set-up committees (e.g., management, dispute resolution, agriculture, welfare, etc.), and coordinate with the Supervisory Authority and the SME. So far, the KIS Office of the GIDA performs both the Supervisory and SME responsibilities because a private sector company has not been contracted to play the SME role.

Data and methods

Data for the analysis included both primary and secondary data, including content analysis of key documents, stakeholder interviews, and surveys of farmers within the KIS to examine the institutional designs informing the governance of the KIS. To analyze the institutional design principles of the rules-in-form, we analyzed the 2016 L.I. 2230 and WUA bylaws for two WUAs: AK/C1 and AK/C2. The two WUAs were selected because they were established first as the pioneering WUAs (almost as pilots), with leaders well-versed in the history and dynamics of the decentralized irrigation reform and the formation of the WUAs. More importantly, they were the only ones with bylaws established and adopted when this study was being conducted. We analyzed the contents of these regulatory documents to see how well they captured the four design principles: authority, responsibility, balanced operational resources, and balanced mechanisms for flexibility and stability. We operationalized these design principles based on their definitions in the Codebook by DeCaro et al. (2025), summarized in Table 1. The statements within the L.I. 2230 and the bylaws were extracted into an Excel spreadsheet. For each statement, two coders examined (1) which of the design principles was/were captured by the statement, and (2) how explicit was/were the statements in capturing the design principles. The latter involved a three-point rating scale: explicit (3) where the design principle and the referenced actor(s) (e.g., actor[s] granted the legal capacity) are clearly/explicitly stated in the statement/rule; moderately explicit (2) where the design principle is clearly/explicitly stated in the statement/rule but the referenced actor(s) (e.g., actor[s] granted the legal capacity) is implicit or vice versa; and implicit (1) where the design principle and the referenced actor(s) (e.g., actor[s] granted the legal capacity) are implicitly captured/stated in the statement/rule.

We piloted the coding and rating on 30 institutional statements/rules in the L.I.2230 and 20 statements in the WUA bylaws to check for inter-coder and inter-rater reliability. We discussed initial disagreements after the first and second pilot coding and rating to reconcile how the design principles were being coded and rated for the L.I. and bylaws. It should be noted that two WUA bylaws, representing two of the WUAs were initially used for the piloting coding. After carefully reviewing the pilot coding, we realized that the WUA bylaws were almost identical in structure and content, with very slight differences. This was expected since the KIS Office works with each WUA to develop their bylaws to approximate the L.I. and other bylaws as much as possible. Our final analysis, therefore, focused on only the L.I 2230 and one of the WUA bylaws illustrate the capacities (for self-governance and cooperation) granted to these WUAs. We summarized and compared the design principles in the L.I. 2230 and WUA bylaws using frequency tables/charts. We also summarized the statements qualitatively in tables to compare the design principles in the L.I. 2230 and bylaws and the responses from key stakeholders and farmers/water users.

The content analysis of the regulatory documents was complemented with stakeholder interviews (n = 10) and survey of farmers (n = 39) within the KIS (see Table 2 for summary profile of farmers surveyed). The stakeholder interviews and farmer survey occurred over a period of five months, from February to June 2021. We employed purposive sampling for the stakeholder interviews, which involved asking the KIS Office to introduce us to key farmer leaders for our interviews. We interviewed one of the two staff members at the KIS Office of the GIDA (n = 1), and nine leaders of the WUAs occupying different roles within their respective WUAs: chairpersons (n = 5), vice chair (n = 1), dispute resolution committee chair (n = 1), organizer (n = 1), and secretary (n = 1). Many of these leaders (n = 5) were also part of the interim management committee formed to establish the WUAs, who provided historical and contemporary accounts of how the KIS institutional regime has evolved over time. We used a structured interview instrument, with both closed- and open-ended questions for the stakeholder interviews. In these interviews, stakeholders provided responses to questions centered around (1) the history of the previous OACS regime, including the legal capacities (design principles) granted to the OACS, farmers’ experiences and challenges under this regime, (2) their understanding and experiences under the new WUA regime, and (3) their perceptions about the legal capacities (design principles) granted to the WUAs. These perceived capacities of the WUAs were captured qualitatively and thematically analyzed to see whether stakeholders perceived these capacities as having improved or not under the WUAs.

Recruiting farmers is often difficult especially during planting seasons, so we employed snowball sampling, leveraging our connections to the stakeholders to help recruit farmers. The total farmers recruited (n = 39) were fairly evenly spread across the 15 WUAs, although the majority (76%) were members of the earlier three established WUAs (WUA C1, C2, and C3) with significantly more experience under this new WUA regime. The survey comprised socio-demographic questions, summarized in Table 2 showing a disproportionately higher percentage of male farmers (82%), mostly within the 36–55 age bracket, and the majority being Basic School Graduates (with basic reading and writing skills). The survey also used a Likert scale from 1 (strongly disagree) to 5 (strongly agree) to elicit farmers’ perceptions about the (legal) capacities (design principles) that (1) were granted to the OACS, and (2) the WUAs. The results were analyzed and summarized, comparing responses between stakeholders and farmers.

RESULTS FROM THE KIS CASE STUDY ANALYSIS

Design principles for self-governance and cooperation: authority and responsibility

The design principles of authority and responsibility are well represented in the rules-in-form (L.I. 2230 and WUA bylaws). For instance, as shown in Figure 2, around 90% or more of statements capturing the authority principles are explicit, meaning they clearly reference the principle, and the actor(s), mostly WUAs, granted such authority. Similarly, more than 80% of stakeholders and smallholder farmers (WUA members) indicated improved authority granted to the WUAs, that is, increased and improved clarity in their authority. For an in-depth analysis of the design principles, Table 3 summarizes the forms or types of authority for WUAs, comparing the L.I. 2230 and the WUA bylaws. These forms of authority are grouped into rule making, collective choice rules, general operations, and enforcement and sanctioning.

Rulemaking authority is stipulated in the L.I. 2230 and bylaws granting WUAs the authority to adopt and amend bylaws and other internal regulations, take and defend legal proceedings on behalf of members in matters of common concern, and adopt internal rules around water use and sanctions. A significant difference is that the L.I. explicitly requires the approval of GIDA/KIS Office (the supervising authority) regarding the adoption and amendment of bylaws, but such a requirement is not explicit in the bylaws. For authority around collective choice rules, the L.I. 2230 and bylaws make a difference between the procedures for elections and other general decisions and dissolution/liquidation/re-organization decisions. Elections and other general decisions are made through a simple majority vote, and decisions around dissolution/liquidation/re-organization require (1) a two-thirds majority vote, or at least 50% of members present and voting, and (2) approval of GIDA. The bylaws also allow for consensus in decision making at various committee levels within the WUA.

Both L.I. 2230 and the bylaws specify the authority of the WUAs to implement activities in line with their operations and engage in enforcement and sanctions. The L.I. has broad and generic provisions around the authority for WUAs to engage in their operational activities, such as the authority to develop a work plan, budget, and service area plan; procure and operate irrigation equipment; and take appropriate measures to combat erosion, salinity, siltation pollution, and encroachment. The bylaws take these broad and generic provisions further by specifying the WUA’s authority, such as the authority to undertake lawful activities reasonably necessary to achieve its purpose and tasks, amend its service area borders, plan the water schedule according to the cropping calendar, reduce water losses and increase efficiency; and approve lease agreements. Similarly, the L.I. 2230 has authority provisions for WUAs to enforce and sanction, including imposing sanctions on members violating bylaws or procedures, imposing fees, fines, and levies, and penalizing those who damage the association’s assets. The bylaws provide additional enforcement and sanctioning authority provisions, including removing members from the association and releasing them from their user rights over land [with approval from GIDA], imposing fines on members who fail to undertake sanitation works and interfere with water delivery, ceasing the supply of irrigation service to members in arrears for more than 21 days, and enforce measures to reduce water losses and increase efficiency as reasonably as possible.

Similarly, the design principles of responsibility are well captured in L.I. 2230 and the WUA Bylaws, with more than 90% of captured statements being explicit, that is, they clearly reference responsibilities, and the actor(s) granted such responsibilities (see Fig. 3). Around 95% or more stakeholders and farmers (in Fig. 3) indicated an improvement in responsibilities, that is, increased and improved clarity in their responsibilities. These responsibilities involve general operations of the WUAs, communication, equity, and dispute resolution (see Table 4). The L.I. 2230 stipulates some of the following responsibilities for the day-to-day general operations of the WUAs, including getting approval from GIDA to incorporate the WUAs: preparing a work plan; managing irrigation and drainage systems within its service area; maintaining, rejuvenating, and improving irrigation and drainage systems within its service area; (re)constructing works delegated to the WUAs by GIDA; and facilitating the training of its members in water-saving methods and improved farming methods and innovation. The WUA bylaws specify additional responsibilities for the WUAs, including ensuring the welfare and unity of its members, encouraging members to take full responsibility for protecting, maintaining, and promoting the association, its assets, and the entire irrigation scheme, planning the water schedule according to the cropping calendar, preparing sanitation work schedule, ensuring that member do not fix more than two pieces of four-inch diameter pipe per hectare to irrigate, and ensuring that members obey any other technical regulations

Similarly, responsibilities related to communication, equity, and dispute resolution are central to L.I. 2230 and the WUA bylaws. The L.I. specifies WUAs’ responsibilities in communicating with its members about decisions, meetings, events, and fiscal matters, but is also quite heavy on the WUAs’ communication responsibilities with GIDA—vertical communication with the supervisory authority. For instance, under the L.I., WUA is responsible for communicating with GIDA about decisions and other matters related to the irrigation and drainage systems, fiscal decisions (e.g., loans, assets, contracts, and liquidation), and getting approval for the association. The bylaws emphasize communication and information-sharing responsibilities between a WUA and its members and other associations—more of horizontal communication—including communication with other WUAs around shared/common irrigation and drainage issues (e.g., water schedule, maintenance, and repairs), and disseminating information regarding water schedule, water and land uses, and other rules to WUA members.

Both L.I. and WUA bylaws stipulate responsibilities around equity and dispute resolution in the activities of all WUA bodies/committees, including the General Assembly of each WUA (compromising members of the WUA), Management Committee, Dispute Resolution Committee, Sanitation Committee, Welfare Committee, and Agricultural Committee. For instance, the L.I. stipulates that the WUAs are responsible for ensuring fairness and equity in decisions and water allocation, avoiding discrimination in water use based on ethnicity, gender, religion, or other similar grounds, and being transparent and participatory in its decisions. The bylaws also add that the WUAs are responsible for developing a fair water distribution schedule proportionate to land sizes and adapted to the cropping calendar, supplying irrigation and drainage services fairly and timely to its members, and providing fair compensation for damages caused to the crops or land of its member resulting from the intentional or negligent act of an employee of the association or the maintenance activities undertaken by the association. For dispute resolution, the L.I. makes WUAs responsible for establishing dispute resolution measures to adjudicate disputes and appeals among its members. The WUA bylaws also specify that WUAs are responsible for adjudicating disputes through their Dispute Resolution Committees in matters related to (a) water use and distribution, (b) irrigation scheduling, (c) irrigation service charge, and (c) violations of bylaws and internal regulations. The L.I. 2230 and bylaws also designate the Supervisory Authority (GIDA) and the courts as the appellate bodies in dispute resolution, allowing for a more intricate mosaic of communication, equity, and dispute resolution channels and mechanisms for cooperation within the KIS decentralized irrigation system.

Design principles for self-governance and cooperation: balanced operational resources and balanced mechanisms for flexibility and stability

The principles of operational resources look at the internal and external resources relied on by the WUAs, making them financially self-sufficient or otherwise. From Table 5, three main internal funding sources are described in both the L.I. 2230 and WUA bylaws: (1) irrigation service charge (ISC), (2) annual membership fees and welfare contributions and fines, and (3) revenue from investments and services rendered by the association. These funding sources are backed by other fiscal mandates in the L.I. and bylaws, including mandates for WUA to determine and collect membership fees and fines; retrieve from members the cost of repairs, construction works, and desilting of drainage; and maintain a reserve fund for emergencies (in an interest-yielding bank account). It should be noted that these internal revenue sources and the fiscal mandates backing them do not necessarily translate into the on-the-ground capacities of the WUAs to raise money from these revenue sources. The L.I. and bylaws also list external resources for WUAs, including government subsidies, gifts/grants, loans, and capacity training from government and non-government sources.

Some differences exist in how some of these mandates are stipulated in the L.I. 2230 and the WUA bylaws. For instance, the L.I. gives WUAs the fiscal mandate to determine and impose an irrigation service charge (ISC) on members proportional to the volume of water used or land size. The bylaws refine this mandate further by adding that the ISC should be calculated referring to plot size in proportion to the total size of the service area as a fraction of (a) annual maintenance cost (e.g., cost of cleaning ditches), (b) cost of providing irrigation services, (c) cost and profit margin of the Irrigation Service Provider, (d) need to replenish or build the association’s Reserve Fund. Each member must pay their ISC directly into the association’s bank account and share their bank receipt with WUA’s treasurer. At least in theory, this additional provision in the bylaw provides extra conditions for the WUA to consider, potentially giving WUAs room to tailor each member’s ISC to their unique circumstances while raising extra revenue to support the Reserve Fund. We also see instances where the L.I. differs from the bylaws regarding the control given to GIDA over certain fiscal decisions of the WUAs. For instance, the L.I. requires the written approval of GIDA before WUAs can make decisions around the sale, purchase of mortgage, or pledging of any assets owned by the association, liquidation, investment, or conclusion of loans, overdrafts, or other financial liabilities, and approve contracts above certain amounts. These oversight provisions suggest a checks and balance mechanism between GIDA and the WUAs, and how they translate on the ground to ensure fiscal autonomy for the WUAs to implement their authorities and responsibilities will be revisited later in the field interviews section of this article.

The design principle of balanced mechanisms for flexibility and stability examines the extent to which provisions in the L.I. and Bylaws are framed either broadly/flexible or rigid (specific/narrow requirements), permissions, or prohibitions. For instance, the WUA Bylaws have grandfathering clauses that allow for flexibility in the bylaws, including the provisions that (1) anyone with landholding right automatically qualifies to be in the association, and (2) leasing land for more than three years makes you eligible to be in the association. That said, most of the provisions in the L.I. and bylaws lean toward rigid requirements (summarized in Table 6). For instance, the bylaws state that WUAs are required not to supply water to members in arrears after 21 days. There are also moratorium clauses that limit changes or foster rigidity in the bylaws, such as WUA members cannot fix more than two pieces of four-inch diameter pipe per hectare to irrigate, members cannot use irrigation water outside the association’s water schedule, WUA-elected leaders are ineligible for re-election after two continuous terms, and WUA cannot vote if they are in arrears for more than two seasons or one year.

The rigid provisions in the L.I. differ from those in the bylaws in some ways. For instance, L.I. 2230 stipulates that the adoption and amendment of WUA bylaws require the written approval of GIDA (supervising authority). However, the bylaws do not require such written approval, except that such adoption and amendment decisions should have a two-thirds majority of votes by WUA members. The same applies to fiscal decisions where the L.I. requires all WUAs to have written approval from GIDA regarding decisions around the sale, purchase of mortgage, or pledging of any assets owned by the association, investment or conclusion of loans, overdrafts, or other financial liabilities of the association, but the bylaws do not stipulate requirements for written approval. Perhaps, the bylaws do not explicitly require GIDA’s approval because GIDA supports (and sometimes leads) the WUAs in writing, adopting, and amending their bylaws. Similarly, and as will be discussed later in the field interviews sections of this article, the WUAs and GIDA often work closely, especially on fiscal matters. The close working relationship may be an implicit avenue for the WUAs and GIDA to work toward meeting the L.I.’s explicit requirement of WUAs getting GIDA’s approval around fiscal decisions. Yet, this written approval requirement raises some questions about the extent to which the WUAs are fiscally autonomous, which we probe further in the field interviews. The clarity in the requirements also brings into question how the SRSG framework wrestles with the tension between rigidity and specificity (or clarity in command structure) in institutional design, which we revisit in the discussion section.

Grounding the rules-in-form (L.I. 2230 and WUA bylaws) in the rules-in-use (feedback from water users in the field)

The rules-in-form (the L.I. 2230 and WUA bylaws) suggest devolved authority and responsibility for WUAs. Interviews with key stakeholders and surveys of sample farmers also confirmed that there is an overall perceived improvement in (and contentment with) the authority and responsibility granted to WUAs. For instance, we coded 13 statements made by stakeholders around authority, and 85% of such statements suggest perceived improvements around authority. Examples included:

We have the authority and autonomy since GIDA doesn’t impose decisions on us anymore. We can suspend you and just write to our supervisory authority and they will agree, unlike the previous OACS model [Stakeholder 06] and

We are more independent now because we make decisions on our own and the GIDA only advises us [Stakeholder 08].

Similarly, we coded 16 statements from the stakeholder interviews around responsibility and all (100%) expressed improvements in the responsibilities granted to the WUAs:

Farmers welcomed the WUA idea because we now have the opportunity to maintain our structures and also collect the ISC [Irrigation Service Charge] [Stakeholder 11].

As WUAs, we mobilize to construct broken drains and laterals, which is our responsibility. We also put in measures to check erosion [Stakeholder 08].

Some stakeholders also highlighted how the communication responsibilities stipulated in the L.I. and WUA bylaws work on the ground:

We have WUA leaders engaging and communicating with farmers within and across associations. For instance, some WUA leaders supervise farmers in other associations to prevent encroachment ... The WUAs, I can say, is like the Parliament where we have a management committee and a general assembly. So, before you take an action, you must first consult the general assembly and ask for their approval [Stakeholder 06].

A decision taken at the management level is communicated to the farmers for them to agree or disagree [Stakeholder 05].

We meet and take decision on marketing of farm produce and decide on cost of machinery and other mechanization services [Stakeholder 11].

Two main reasons may account for this perceived improvement and contentment around the authority and responsibility granted to the farmers (and the WUAs) as part of the decentralized irrigation reforms. First, prior to the 2016 decentralized irrigation reforms, farmers mainly relied on the Ghana Irrigation Development Irrigation (GIDA) as the scheme manager to operate and maintain the irrigation scheme. For some farmers, the reliance on this state entity (GIDA) was both inefficient and ineffective, as they suffered delays in maintaining and repairing the canals and sub-canals. The post-2016 reforms, however, put the authority and responsibility of maintaining the sub-canals (and sometimes even the canal) into the hands of the farmers,

The WUA provides us the authority and responsibility to maintain the irrigation and drainage systems so we don’t have to wait for GIDA/KIS, and things work faster for us now [Stakeholder 10].

Even in the mother canal the WUA takes care of it often through communal labor. We don’t rely on them [GIDA] for maintenance [Stakeholder 03].

The second reason accounting for the perceived improvement of and contentment with the authority and responsibility granted to WUAs relates to the trans-local and cross-contexts learning between the KIS farmers and their Japanese counterparts. The Japanese government, through the Japanese International Cooperation Authority (JICA), worked with the Ghana Government and the World Bank to help establish the WUAs. This collaboration involved capacity-building training for the WUAs offered by JICA, including sending some KIS farmers (and WUA leaders) to Japan to learn about the WUAs. Recall that the conceptual lens for this article, state-reinforced self-governance, used the case of Japan’s post-war land reform between 1946 and 1950 as its empirical case (Sarker 2013, 2014). For the KIS farmers, learning about this land reform, including the reorganization of bona fide farmers into a federation of WUAs or members of designated Land Improvement Districts (LIDs) in Japan, has been instrumental in informing them about the significance and benefits of the ongoing decentralized irrigation reforms in Ghana since the L.I. 2230 was enacted in 2016. As one farmer observed,

We are in the process of forming our federation so we can replicate what the Japanese practice. Farmers here have been trained in Japan already. The Japanese model have the Land Developers Authority, so the water managers only manage water like the WUA in Ghana and their other cooperative help with loans and machinery but here the WUA deal with only water issues [Stakeholder 09].

Although farmers or water users applauded the devolved authority and responsibility legally conferred on them by the L.I. 2230 and bylaws, they also highlighted what seems missing: their limited capacity to be fiscally autonomous. This limited capacity is less about the rules-in-form: the fiscal mandates and funding sources stipulated in the L.I. 2230 and bylaws. Instead, it is more about the everyday lived experiences of WUAs not able to generate revenues from their members for reasons such as members not having the money/means to pay membership dues, welfare contributions, fines, ISC charges, sometimes because of low farm yields, increasing costs of farming, low savings, and/or low farmgate prices. Some users also refuse to pay the fees, fines, or ISC charges, constraining the capacity of WUAs to raise revenue from their members as one stakeholder noted, “We are not capable of raising enough funds to support the maintenance of the irrigation system” [Stakeholder 11].

This limited capacity to raise revenue from members informs an ongoing practice (rules-in-use) that helps WUAs to mobilize revenue in the short term but may further constrain their fiscal autonomy in the long run. Instead of determining and collecting the ISC charges and annual membership fees, welfare contributions, and fines separately, the WUAs have lumped all together into the ISC charge, around GH₵ 720 (around US$45 in July 2024 exchange rate) per hectare annually across all the WUAs. The WUAs give the total amount to the KIS (the supervisory authority and the scheme management office of GIDA). The KIS/GIDA takes 79% of each WUA’s total amount and returns 21%, which the WUAs use to run their associations and operate and maintain their lateral canals and drainage systems.

Speaking to some WUA leaders and KIS officials during our fieldwork, the following clarifications were made. First, the 79–21 split was collectively decided on by the KIS and leaders of the WUAs: KIS takes 79% of each WUA’s annual revenue generated as payment for irrigation water and to cover the cost of maintaining the main canal and drainage system. Second, the KIS uses part of the 79% to cover additional expenses or fund the deficit of WUAs based on their annual budgets, or sometimes the KIS will directly maintain the lateral canals and sub-drainage systems of WUAs who cannot cover such expenses or experience an emergency (e.g., flooding) during a year. To the KIS officials, the WUAs and farmers look like they operate on a debt snowballing or debt avalanche system:

As for these WUAs, the deficits never end ... these WUA farmers will use money from their current harvest to pay the previous ISC arrears and cost of other inputs ... we try getting them to prepay the ISC by sometimes stopping them from going to their farms or asking them to pay a percentage ... some pay, some don’t. If this happens this way, who will you release the water to? Controlling water individually, too, is a challenge for us [GIDA/KIS Office] [Stakeholder KIS].

Third, collecting 79% of each WUA’s annual revenue is insufficient for KIS to fully cover the cost of providing irrigation water and maintaining the main canals and drainage systems. To give a sense of the expenditure, a KIS official narrated this example for reference:

We spent around GH₵6000 daily cleaning the lateral canals for three WUAs (WUA/C1, WUA/C2, and WUA/C3). The GH₵6000 [$375 in July 2024 exchange rate] is how much it takes to hire the machine, labor, and buy fuel for the machine. It took us around a month or 30 days to clean. Now, multiply 6000 by 30 and see how much it costs us to clean the lateral canals for only three WUAs. So, when you do the calculations, you have no money left to do anything ... even if all the farmers in the WUAs pay their ISC of GH₵720 for the entire 3000 ha of irrigation land, how much will that be, and how much would be left if we subtract the cost of cleaning the lateral canals alone for all the WUAs?

The KIS relies on external support, including JICA, which provides heavy machinery to clear weeds in the canal and drainage system. The KIS is also exploring installing new, low-maintenance, water-efficient, and cost-saving irrigation systems at the lateral canals for the WUAs. Both WUA leaders and KIS officials noted that the 79–21 split might go a long way to cover all expenses if all WUA members had paid their ISC charges. Lastly, consolidating all fees, charges, and welfare contributions into the ISC raises concerns about how this strategy (1) becomes a short-term measure and does not address why WUA members are still not paying, and (2) makes GIDA, not the WUAs, central in fiscal decisions, which may undermine the fiscal autonomy of the WUAs in the long run.

Another challenge to fiscal autonomy was about the capacity of WUAs to secure loans. A farmer noted, “For maintenance responsibilities, I prefer WUA, but for financial autonomy, I prefer OACS because they could access loans for us” [Stakeholder 08]. Another stated, “Government should permit or license the WUAs as business ventures, so they can access loans. The WUAs are like commercial farmers” [Stakeholder 03]. Another expressed a similar concern, “We should be given the power to secure loans like the OACS did, or better still treat us as companies so we can secure loans on our own at a good rate” [Stakeholder 04]. In further probing these statements, it is important to note that the L.I. 2230 and WUA bylaws grant WUAs the fiscal mandate to secure loans. However, the limitations facing the WUAs now include (1) their membership per association is now relatively small (compared to their previous large farmer association, OACS), which limits their ability to mobilize the needed collateral for bank loans and raise the needed revenue to service loan payments, and (2) the L.I. requires written approval from GIDA before they can secure such loans. Part of the solution to the challenges of securing loans may not lie in making the rules-in-form (L.I. and bylaws) more flexible (e.g., changing the rigid requirements for GIDA’s approval or having GIDA serve as loan guarantor). For some, “farmers will abuse the rules if these rules contain flexible mechanisms” (Stakeholder 11]. For others, “rules must be as rigid as possible, and strictly adhered to without compromises” [Stakeholder 08]. Instead, the solution may still lie in addressing the WUAs’ capacity to mobilize revenue and build financial credibility (or raising the needed collateral) to secure loans when needed.

In summary, despite improved capacities/design principles in the rules-in-form, the KIS seems to face challenges like those of the previous OACS regime. For these farmers, there are (1) still challenges in getting farmers to attend meetings, (2) difficulties mobilizing resources to maintain the branch canals, drainages and access roads, (3) delays in rehabilitating and maintaining the main canal and drainage, (4) concerns about non-payment of the ISC and dues by WUA members, and (5) increased concerns about the lack of access to loans and heavy reliance on external resources, which are infrequent and less predictable. As one stakeholder remarked, the “disrespect among farmers” and “farmers not attending meetings” [Stakeholder 01] continue to threaten shared decision making and the prospect of more cooperative and federated WUAs. Another remarked, “Financial support is what we need more than any other thing because farming is business” [Stakeholder 05]. For others, “We are not capable of raising enough funds to support the maintenance of the irrigation system” [Stakeholder 11]. As some recognized the imminent realities facing the WUAs, “Unfortunately, JICA will leave us coming June [2021]” [Stakeholder 05], others made passionate appeals for external resources, “We look forward to NGOs assisting us in future” [Stakeholder 08] and “We plead with financial institutions and NGOs to help us. The government should come to our aid” [Stakeholder 10].

DISCUSSION AND CONCLUSION: SYNTHESIS AND THE STATE-FARMER RELATIONS IN SELF-GOVERNING THE KIS

The KIS case suggests an improvement in authority and responsibility principles in the institutional design of Ghana’s decentralized irrigation reforms. Legally, the L.I 2230 and WUA Bylaws granted improved authority and responsibility to the KIS WUAs, suggesting an instance of state-mandated self-governance. That is, the KIS case suggests devolved authority and responsibility to WUAs, which is unlike other instances of partial, forced, or deconcentrated decentralized irrigation institutions in sub-Saharan Africa (Yami 2013, Suhardiman and Giordano 2014, Aarnoudse et al. 2018). Similar instances of state-mandated, top-down decentralized irrigation reforms (with the support of donor partners) have been documented across Sub-Saharan Africa (Brown 2011, Van Den Broek and Brown 2015, Makaya et al. 2020). These WUAs are granted legal mandates to (1) undertake construction and reconstruction works delegated to them by the GIDA/KIS Office; (2) take measures to combat erosion, salinity, siltation pollution, and encroachment; (3) make internal rules for the use of irrigation water and collect fees from members for services provided; (4) procure, maintain, and operate irrigation equipment; and (5) facilitate training of members in irrigation techniques, irrigation farming methods, water-saving methods, and new technologies of irrigation. For many stakeholders and farmers interviewed and surveyed in the KIS case, the state’s devolution of authority and responsibility marks significant improvements in their ability to self-govern the KIS irrigation commons, which is similar to Kenya’s decentralized water reform (Baldwin et al. 2016, McCord et al. 2017), but differs from how farmers in Malawi perceived diminished decision-making authority after the formation of WUAs (see Veldwisch et al. 2009).

The KIS case also suggests that, on paper (rules-in-form), the decentralized irrigation reform improved the institutional design around operational resources (internal and external funding sources and fiscal mandates) for the WUAs to move toward fiscal autonomy and self-sufficiency. Crucially, two overlapping issues threaten this goal of fiscal autonomy. First, the WUAs remain underfunded. For most stakeholders and farmers, the limited capacity of WUAs to mobilize funding internally and from banks and reliance on external donor support (e.g., JICA and World Bank for training and other forms of support) threaten the WUAs’ ability to fully execute the fiscal mandates granted to them by the L.I. and their bylaws. As most farmers observed, the resources and capacity-building training offered by JICA and other donor partners are central to the WUAs, but there are mounting concerns about their sustenance if/when these external resources are no longer available. The challenge of achieving balanced operational resources, partly due to limited fiscal autonomy and the unpredictable inflow of external resources from national and donor partners, is a running theme across many decentralized irrigation systems in sub-Saharan Africa (Le Gal et al. 2003, Veldwisch et al. 2009, Hassan et al. 2014, Mutambara and Munodawafa 2014).

Second, the fiscal autonomy granted to the WUAs in the rules-in-form (L.I. 2230 and WUA bylaws) is impacted by two overlapping matters: WUAs’ limited capacity to raise revenue from members and their continued (over)reliance on GIDA to mobilize revenue in the short-run, and not necessarily tackle the core issues facing farmers’ fiscal well-being. The case of the KIS is similar to instances of water user fees being insufficient to cover operations and maintenance costs for irrigation schemes (Le Gal et al. 2003) and water users refusing to pay user fees (Veldwisch et al. 2009, Mutambara and Munodawafa 2014, Richards 2019). In the case of the KIS, however, WUAs have fiscal mandates in the L.I. and Bylaws, including sanctioning mechanisms to mobilize revenues from their members, such as petitioning GIDA to (re)lease the land of members in arrears to generate money to pay for unpaid fees, charges, etc. However, fully exercising these fiscal mandates also relies on WUA members having the means to pay, which is often not the case among these water users/farmers who sometimes struggle with low farm yields, increasing costs of farming, low savings, and/or low farmgate prices. Thus, the revenue generation capacities of the WUAs seem tied to their members’ fiscal well-being, which remains a core issue in the KIS.

The current solution of consolidating all fees, charges, and welfare contributions into the ISC charge and submitting everything to the KIS, which takes 79% and gives the remaining 21% to WUAs, risks centralizing too much fiscal control into GIDA in the long run, which is opposite to what the decentralized irrigation reforms were meant to achieve. Such (over)reliance of WUA on GIDA also speaks to similar instances of WUAs in sub-Saharan Africa relying on their government-sanctioned supervisory authorities for several activities, including preparing and approving WUA bylaws (see also Yami 2013, Suhardiman and Giordano 2014). The intent here is not to treat financial autonomy as the only way to secure autonomy and self-organization. The point of testing the SRSG framework in this case was to ascertain how achieving some aspects of the design principles without others may constrain actors (e.g., WUAs) in fully realizing self-governance.

Further, the KIS case offers some thoughts on procedural mechanisms in institutional designs around state-mandated decentralized irrigation reforms. Despite the challenges, farmers/water users and their associations acknowledged and appreciated the procedural mechanisms the government and its donor partners (e.g., JICA) employed to involve water users in the reform, particularly around fostering trans-local learning among water users in the KIS and Japan. Even though the reform was largely a state-sanctioned, top-down process, the state’s role in building a transparent and deliberative process among the water users and other stakeholders seemed central to what others have earlier discussed as a vital role(s) of the state in co-management and local collective action around de/recentralization efforts (Bullinger and Haug 2012, Villamayor-Tomas et al. 2019, Salman et al. 2023). The institutional designs embedded in the KIS decentralized irrigation reforms support or reinforce local collective action to the extent that water users are involved in the process and can understand and leverage them to self-govern the irrigation commons. Such procedural mechanisms are akin to the provision of recourse opportunities for water users in Kenya’s 2002 Water Act, allowing such users to hold decision makers accountable within Kenya’s polycentric water governance system (McCord et al. 2018).

In conclusion, testing the SRSG framework for insights into the institutional designs for decentralized irrigation reforms provokes unanswered questions/topics for future work. For instance, future work could use the results of the current work as a benchmark for a historical, pre-, and post-analysis of if/how the SRSG institutional designs (rules-in-form and rules-in-use) have evolved and the drivers of such evolution. The limitations of the SRSG principles also offer productive entry points for future studies and conversations. For instance, as highlighted in the result section, analyzing the L.I. and bylaws for the SRSG principle of rigidity also brings up instances of specificity. Some of these rigid provisions may seem to suggest instances of specificity or clarity in commands. Are clear commands in provisions necessarily rigid? And are rigid (inflexible) provisions necessarily written in clear commands? These and other tensions within the SRSG principles present productive tensions to engage, both conceptually and empirically, in future studies to further advance the SRSG framework.

Further, there is also the need to look into how the SRSG institutional design principles relate to performance outcomes around the irrigation systems (e.g., efficiency, reliability, and equity in water delivery). Such an interrogation could also look at all 15 WUAs to see how variations in performance outcomes relate to the institutional designs in the rules-in-form (L.I. 2230 and bylaws) or rules-in-use. Similarly, the extent to which historically contingent, customary property rights regimes complexly intersect with formal institutional arrangements around irrigation reforms remains important for future studies around the SRSG in sub-Saharan Africa. Finally, as African governments envision and pursue institutional reform initiatives, especially around natural resources, using the SRSG framework for a comparative historical analysis could further elucidate how such reforms can be institutionally designed and implemented for local self-governance.

LITERATURE CITED

Aarnoudse, E., A. Closas, and N. Lefore. 2018. Water user associations: a review of approaches and alternative management options for Sub-Saharan Africa. IWMI Working Paper 180. International Water Management Institute, Colombo, Sri Lanka. https://doi.org/10.5337/2018.210

Agrawal, A. 2001. Common property institutions and sustainable governance of resources. World Development 29:1649-1672. https://doi.org/10.1016/S0305-750X(01)00063-8

Agrawal, A., and E. Ostrom. 2001. Collective action, property rights, and decentralization in resource use in India and Nepal. Politics & Society 29:485-514. https://doi.org/10.1177/0032329201029004002

Andersson, K. P., and E. Ostrom. 2008. Analyzing decentralized resource regimes from a polycentric perspective. Policy Sciences 41:71-93. https://doi.org/10.1007/s11077-007-9055-6

Awortwi, N. 2011. An unbreakable path? A comparative study of decentralization and local government development trajectories in Ghana and Uganda. International Review of Administrative Sciences 77:347-377. https://doi.org/10.1177/0020852311399844

Ayee, J. R. 1997. The adjustment of central bodies to decentralization: the case of the Ghanaian bureaucracy. African Studies Review 40:37-57. https://doi.org/10.2307/525156

Baldwin, E., C. Washington‐Ottombre, J. Dell'Angelo, D. Cole, and T. Evans. 2016. Polycentric governance and irrigation reform in Kenya. Governance 29:207-225. https://doi.org/10.1111/gove.12160

Blomquist, W., A. Dinar, and K. E. Kemper. 2005. Comparison of institutional arrangements for river basin management in eight basins. Policy Research Working Paper No. 3636. World Bank, Washington, D.C., USA. https://doi.org/10.1596/1813-9450-3636

Blomquist, W., A. Dinar, and K. E. Kemper. 2008. A framework for institutional analysis of decentralization reforms in natural resource management. Society and Natural Resources 23:620-635. https://doi.org/10.1080/08941920902960396

Brown, J. 2011. Assuming too much? Participatory water resource governance in South Africa. Geographical Journal 177:171-185. https://doi.org/10.1111/j.1475-4959.2010.00378.x

Bullinger, C., and M. Haug. 2012. In and out of the forest: decentralisation and recentralisation of forest governance in East Kalimantan, Indonesia. Advances in Southeast Asian Studies 5:243-262.

Cambaza, C., J. Hoogesteger, and G. J. Veldwisch. 2020. Irrigation management transfer in sub-Saharan Africa: an analysis of policy implementation across scales. Water International 45:3-19. https://doi.org/10.1080/02508060.2019.1702310

Cosens, B., and B. C. Chaffin. 2016. Adaptive governance of water resources shared with indigenous peoples: the role of law. Water 8(3):97. https://doi.org/10.3390/w8030097

Cox, M., G. Arnold, and S. Villamayor Tomás. 2010. A review of design principles for community-based natural resource management. Ecology and Society 15(4):38. https://doi.org/10.5751/ES-03704-150438

DeCaro, D., B. C. Chaffin, E. Schlager, A. S. Garmestani, and J. B. Ruhl. 2017. Legal and institutional foundations of adaptive environmental governance. Ecology and Society 22(1):32. https://doi.org/10.5751/ES-09036-220132

DeCaro, D. A., E. C. Schlager, and E. Frimpong Boamah. 2025. The State-reinforced self-governance framework: conceptualizing and diagnosing legal and other institutional foundations of adaptive and transformative environmental governance. Ecology and Society 30(2):1. https://doi.org/10.5751/ES-15567-300201

DeCaro, D. A., E. Schlager, E. Frimpong Boamah, A. Adusei, E. Baldwin, W. Blomquist, E. Dennis, J. García, A. Garmestani, B. Hudson, A. Rissman, A. Rudolph, S. Siddiki, and M. Scown. 2020. State-reinforced self-governance project codebook.

DeCaro, D., and M. Stokes. 2008. Social‐psychological principles of community-based conservation and conservancy motivation: attaining goals within an autonomy‐supportive environment. Conservation Biology 22:1443-1451. https://doi.org/10.1111/j.1523-1739.2008.00996.x

Dinar, A., K. Kemper, W. Blomquist, and P. Kurukulasuriya. 2007. Whitewater: decentralization of river basin water resource management. Journal of Policy Modeling 29:851-867. https://doi.org/10.1016/j.jpolmod.2007.06.013

Doss, C. R., and R. Meinzen-Dick. 2015. Collective action within the household: insights from natural resource management. World Development 74:171-183. https://doi.org/10.1016/j.worlddev.2015.05.001

Easter, K. W. 2000. Asia’s irrigation management in transition: a paradigm shift faces high transaction costs. Applied Economic Perspectives and Policy 22:370-388.

Ferguson, A., and W. Mulwafu. 2007. If government failed, how are we to succeed? The importance of history and context in present-day irrigation reform in Malawi. Pages 211-227 in B. van Koppen, M. Giordano, and J. Butterworth, editors. Community-based water law and water resource management reform in developing countries. CABI, Wallingford, UK. https://doi.org/10.1079/9781845933265.0211

Frey, U. J., S. Villamayor-Tomas, and I. Theesfeld. 2016. A continuum of governance regimes: a new perspective on co-management in irrigation systems. Environmental Science & Policy 66:73-81. https://doi.org/10.1016/j.envsci.2016.08.008

Frimpong Boamah, E. 2018. Constitutional economics of Ghana’s decentralization. World Development 110:256-267. https://doi.org/10.1016/j.worlddev.2018.05.036

Frimpong Boamah, E., and A. B. Nyamekye. 2021. Caught in the rigidity trap? Limits to farmers’ adaptation to water challenges in Ghana’s smallholder irrigation schemes. Society & Natural Resources 34:866-884. https://doi.org/10.1080/08941920.2021.1897720

Groenfeldt, D., and M. Svendsen. 2000. Case studies in participatory irrigation management. World Bank, Washington, D.C., USA. https://doi.org/10.1596/0-8213-4540-0

Hassan, R., A. Mtsweni, M. Wilkinson, D. Weston, J. Mutundo, T. Magagula, P. Sithole, S. Farolfi, and A. Dinar. 2014. Water governance decentralization in Africa: a framework for reform process and performance analysis. Water Research Commission, Pretoria, South Africa.

Johnson III, S. H. 1997. Irrigation management transfer: decentralizing public irrigation in Mexico. Water International 22:159-167. https://doi.org/10.1080/02508069708686695

Kemerink, J. S., L. E. Méndez, R. Ahlers, P. Wester, and P. Van Der Zaag. 2013. The question of inclusion and representation in rural South Africa: challenging the concept of water user associations as a vehicle for transformation. Water Policy 15:243-257. https://doi.org/10.2166/wp.2012.127

Konadu-Agyemang, K. 2000. The best of times and the worst of times: structural adjustment programs and uneven development in Africa: the case of Ghana. Professional Geographer 52:469-483. https://doi.org/10.1111/0033-0124.00239

Le Gal, P.-Y., T. Rieu, and C. Fall. 2003. Water pricing and sustainability of self-governing irrigation schemes. Irrigation and Drainage Systems 17:213-238. https://doi.org/10.1023/A:1025194513819

Madigele, P. K. 2018. Efficiency of common-pool resource institutions: focusing on water users associations in South Africa. Environment, Development and Sustainability 20:825-840. https://doi.org/10.1007/s10668-017-9912-1

Makaya, E., M. Rohse, R. Day, C. Vogel, L. Mehta, L. McEwen, S. Rangecroft, and A. F. Van Loon. 2020. Water governance challenges in rural South Africa: exploring institutional coordination in drought management. Water Policy 22:519-540. https://doi.org/10.2166/wp.2020.234

McCord, P., J. Dell'Angelo, E. Baldwin, and T. Evans. 2017. Polycentric transformation in Kenyan water governance: a dynamic analysis of institutional and social‐ecological change. Policy Studies Journal 45:633-658. https://doi.org/10.1111/psj.12168

McCord, P., K. Waldman, E. Baldwin, J. Dell'Angelo, and T. Evans. 2018. Assessing multi-level drivers of adaptation to climate variability and water insecurity in smallholder irrigation systems. World Development 108:296-308. https://doi.org/10.1016/j.worlddev.2018.02.009

McGinnis, M. D. 2011. An introduction to IAD and the language of the Ostrom workshop: a simple guide to a complex framework. Policy Studies Journal 39:169-183. https://doi.org/10.1111/j.1541-0072.2010.00401.x

McGinnis, M. D., and E. Ostrom. 2014. Social-ecological system framework: initial changes and continuing challenges. Ecology and Society 19(2):30. https://doi.org/10.5751/ES-06387-190230

Ministry of Food and Agriculture. 2019. Ghana Commercial Agriculture Project (GCAP). Environmental and Social Impact Assessment (ESIA) of the Rehabilitation and Modernisation of the Kpong Irrigation Scheme (KIS). Ministry of Food and Agriculture, Accra, Ghana.

Mollinga, P. P., R. S. Meinzen‐Dick, and D. J. Merrey. 2007. Politics, plurality and problemsheds: a strategic approach for reform of agricultural water resources management. Development Policy Review 25:699-719. https://doi.org/10.1111/j.1467-7679.2007.00393.x

Mutambara, S., and A. Munodawafa. 2014. Production challenges and sustainability of smallholder irrigation schemes in Zimbabwe. Journal of Biology, Agriculture and Healthcare 4:87-96.

Nkonya, L. K. 2006. Customary laws for access to and management of drinking water in Tanzania. Law Environment and Development Journal 2:50.

Olowu, D., J. Wunsch, and J. R. A. Ayee. 2004. Local Governance in Africa: the challenges of democratic decentralization. Lynne Rienner, Boulder, Colorado, USA.

Ostrom, E. 1990. Governing the commons: the evolution of institutions for collective action. Cambridge University Press, Cambridge, UK. https://doi.org/10.1017/CBO9780511807763

Ostrom, E. 1992. Crafting institutions for self-governing irrigation systems. Institute for Contemporary Studies Press, San Francisco, California, USA.

Ostrom, E. 1993. Design principles in long‐enduring irrigation institutions. Water Resources Research 29:1907-1912. https://doi.org/10.1029/92WR02991

Ostrom, E. 2005. Understanding institutional diversity. Princeton University Press, Princeton, New Jersey, USA.

Ostrom, E. 2007. A diagnostic approach for going beyond panaceas. Proceedings of the National Academy of Sciences 104:15181-15187. https://doi.org/10.1073/pnas.0702288104

Ostrom, E. 2009. A general framework for analyzing sustainability of social-ecological systems. Science 325:419-422. https://doi.org/10.1126/science.1172133

Ostrom, E. 2010. Beyond markets and states: polycentric governance of complex economic systems. American Economic Review 100:641-672. https://doi.org/10.1257/aer.100.3.641

Ostrom, E., R. Gardner, and J. Walker. 1994. Rules, games, and common-pool resources. University of Michigan Press, Ann Arbor, Michigan, USA. https://doi.org/10.3998/mpub.9739

Paller, J. W. 2015. Informal networks and access to power to obtain housing in urban slums in Ghana. Africa Today 62:31-55. https://doi.org/10.2979/africatoday.62.1.31

Ribot, J. C., A. Agrawal, and A. M. Larson. 2006. Recentralizing while decentralizing: how national governments reappropriate forest resources. World Development 34:1864-1886. https://doi.org/10.1016/j.worlddev.2005.11.020

Richards, N. 2019. Water users associations in Tanzania: local governance for whom? Water 11:2178. https://doi.org/10.3390/w11102178

Riddell, J. B. 1992. Things fall apart again: structural adjustment programmes in sub-Saharan Africa. Journal of Modern African Studies 30:53-68. https://doi.org/10.1017/S0022278X00007722

Riedl, R. B. 2014. Authoritarian origins of democratic party systems in Africa. Cambridge University Press, Cambridge, UK. https://doi.org/10.1017/CBO9781107045484

Sahide, M. A. K., S. Supratman, A. Maryudi, Y.-S. Kim, and L. Giessen. 2016. Decentralisation policy as recentralisation strategy: forest management units and community forestry in Indonesia. International Forestry Review 18:78-95. https://doi.org/10.1505/146554816818206168

Salman, M., I. Fertő, and E. Pek. 2023. Both sides of irrigation management: an empirical analysis of smallholders’ perception of co‐management of irrigation systems. Irrigation and Drainage 72:1333-1345. https://doi.org/10.1002/ird.2894

Sarker, A. 2013. The role of state‐reinforced self‐governance in averting the tragedy of the irrigation commons in Japan. Public Administration 91:727-743. https://doi.org/10.1111/padm.12011

Sarker, A. 2014. Federated rural organization for governing the commons in Japan. Journal of Rural Studies 36:42-51. https://doi.org/10.1016/j.jrurstud.2014.06.007

Schlager, E., and M. Cox. 2018. The IAD framework and the SES framework: an introduction and assessment of the Ostrom workshop frameworks. Chapter 6 in C. M. Weible and P. A. Sabatier, editors. Theories of the policy process. Routledge, New York, New York, USA. https://doi.org/10.4324/9780429494284-7

Senanayake, N., A. Mukherji, and M. Giordano. 2015. Re-visiting what we know about irrigation management transfer: a review of the evidence. Agricultural Water Management 149:175-186. https://doi.org/10.1016/j.agwat.2014.11.004

Suhardiman, D., and M. Giordano. 2014. Is there an alternative for irrigation reform? World Development 57:91-100. https://doi.org/10.1016/j.worlddev.2013.11.016

Turral, H. 1995. Devolution of management in public irrigation systems: cost shedding, empowerment and performance. Overseas Development Institute, London, UK.

Van Den Broek, M., and J. Brown. 2015. Blueprint for breakdown? Community-based management of rural groundwater in Uganda. Geoforum 67:51-63. https://doi.org/10.1016/j.geoforum.2015.10.009

Van Koppen, B., and B. Schreiner. 2019. Viewpoint-A hybrid approach to statutory water law to support smallholder farmer-led irrigation development (FLID) in Sub-Saharan Africa. Water Alternatives 12:146-155.

Van Rooyen, A. F., M. Moyo, H. Bjornlund, T. Dube, K. Parry, and R. Stirzaker. 2020. Identifying leverage points to transition dysfunctional irrigation schemes towards complex adaptive systems. International Journal of Water Resources Development 36:S171-S198. https://doi.org/10.1080/07900627.2020.1747409

Van Vuren, G., C. Papin, and N. El Haouari. 2004. Participatory irrigation management: comparing theory with practice a case study of the Beni Amir irrigation scheme in Morocco. Séminaire sur la modernisation de l'agriculture irriguée, Rabat, Morocco.

Vandersypen, K., A. C. Keita, Y. Coulibaly, D. Raes, and J. Y. Jamin. 2007. Formal and informal decision making on water management at the village level: a case study from the Office du Niger irrigation scheme (Mali). Water Resources Research 43(6). https://doi.org/10.1029/2006WR005132

Veldwisch, G. J., A. Bolding, and P. Wester. 2009. Sand in the engine: the travails of an irrigated rice scheme in Bwanje Valley, Malawi. Journal of Development Studies 45:197-226. https://doi.org/10.1080/00220380802265587

Villamayor-Tomas, S., A. Thiel, L. Amblard, D. Zikos, and E. Blanco. 2019. Diagnosing the role of the state for local collective action: types of action situations and policy instruments. Environmental Science & Policy 97:44-57. https://doi.org/10.1016/j.envsci.2019.03.009

Weible, C. M., editor. 2023. Theories of the policy process. Taylor & Francis, New York, New York, USA. https://doi.org/10.4324/9781003308201

Wellens, J., M. Nitcheu, F. Traore, and B. Tychon. 2013. A public-private partnership experience in the management of an irrigation scheme using decision-support tools in Burkina Faso. Agricultural Water Management 116:1-11. https://doi.org/10.1016/j.agwat.2012.09.013

Yami, M. 2013. Sustaining participation in irrigation systems of Ethiopia: what have we learned about water user associations? Water Policy 15:961-984. https://doi.org/10.2166/wp.2013.031