Community Management of groundwater

Community Management of Groundwater
Edella Schlager

School of Public Administration and Policy

University of Arizona, USA


Abstract
Over the past several decades, farmers have increasingly invested in and utilized groundwater to irrigate crops, either as a supplement to or a substitute for canal irrigation. While groundwater aquifers and canal systems are both forms of common pool resources, they pose distinctly different governance challenges if they are to be managed sustainably. This paper draws on the emerging theory of common pool resource governance, particularly the work of Elinor Ostrom and colleagues, to examine the challenges local communities of irrigators face in managing groundwater and the conditions under which those challenges are likely to be addressed through the development of self-governing arrangements. Governance problems that groundwater users face will be contrasted with those faced by canal irrigators. In addition, the paper explores the types of productive and complementary relations between local communities and higher level governments that promote community base groundwater governance and the resolution of especially difficult groundwater problems. The challenges of governing groundwater basins and examples of self-governing arrangements are illustrated with a variety of case studies.

Paper prepared for the International Water Management Institute’s Writers Workshop for “The Agricultural Groundwater Revolution”, Anand, India, February 21-23, 2005.

Introduction
Over the past twenty years, scholars have devoted considerable attention to the ability of farmers, fishermen, pastoralists, and other types of resource users to organize, adopt, monitor and enforce institutional arrangements that govern their use of common pool resources in a sustainable manner (Ostrom, et al. 2002). During that time period, considerable progress has been made in carefully identifying and defining key theoretical concepts, developing typologies that organize diverse types of problems and institutional arrangements, identifying factors that help explain the circumstances under which resources users are likely to engage in collective action to develop governing arrangements, identifying design principles that account for durability of self-governing arrangements, and developing an impressive body of empirical work devoted to theory development and hypothesis testing. According to Stern, et al. (2002:445) the study of institutions for managing common pool resources is sufficiently developed to be recognized as a field within the social sciences. Canal irrigation systems have been a focal resource in the development of this field. Considerable attention and effort has been devoted to explaining the conditions that contribute to the emergence and persistence of farmer managed irrigation systems. In addition, comparative analyses of farmer managed and government managed systems have also been conducted.

This paper extends the work of scholars on self-governance of common pool resources to groundwater in irrigation settings. While considerable work has been conducted by such scholars on groundwater basins in the U.S, little focused attention has been paid to groundwater and irrigation. The first section of the paper covers conceptual tools and theory from the field of common pool resource governance. The tools and theoretical concepts are illustrated through case studies. In addition, empirical findings from canal irrigation systems are reviewed. The second section of the paper applies the conceptual tools and theoretical concepts to groundwater irrigation. The arguments are illustrated through several case studies. The final section explores promising types of linkages between communities of groundwater users and higher level governments. Local level governance is a key component of sustainably managing groundwater basins. How higher level governments can encourage and support local management efforts is an important topic.

A Theory of Common Pool Resources

Foundational Concepts

Common pool resources are defined as natural or human made structures characterized by costly exclusion and subtractability of units (Ostrom and Ostrom 1977). Examples include canal irrigation systems, groundwater basins, fisheries, forests, and grazing lands. Both exclusion and subtractability present challenges for governing common pool resources sustainably. Exclusion involves defining who may enter a resource and who may not, making such a determination is rarely a straightforward process. Ideally, exclusion should occur in a manner that limits access to the number of users whose use will not threaten the resource. Physical, institutional, and social issues often confound such efforts (Ostrom, Gardner and Walker 1994). The sheer size of some resources makes enforcing access limitations in any meaningful or cost effective manner virtually impossible. In other instances, national or state constitutions forbid denying citizens access to natural resources. In other settings, there may be political or economic reasons for avoiding strict access controls. For instance, a number of canal irrigation systems have been described as long and lean – the goal being to provide at least some water to as much land as is possible. Reasons for that range from equity concerns – assisting many people, to cost-benefit analysis issues – the more land included in a scheme, the better the cost-benefit ratios. In either case, too much land is included within a system with some farmers experiencing chronic water shortages.

Exclusion is critical for sustainability, but also for governance. Resource users are much less likely to undertake costly and time consuming efforts to manage common pool resources if they cannot capture many of the benefits resulting from good management. Why design a water allocation scheme that conserves water if the additional water supplies may be captured and used by someone else? Why invest in groundwater recharge projects if others can pump the recharged water? Inadequate exclusion promotes free riding and free riding discourages collective action (Dietz, et al. 2002).

Even if exclusion is adequately addressed in relation to a common pool resource, sustainability is not ensured because of substractability. Subtractability means that each “unit” harvested from a CPR is not available for other users to harvest. The groundwater that a well owner pumps and uses to water his crops is not available for other well owners to pump. Since each resource user gains the value of each unit harvested but imposes some of the costs of harvesting on all resource users, resource users are likely to harvest more than is economically or ecologically desirable (Gordon 1954, Scott 1955, Dietz et al. 2002). Or, as Ostrom, Gardner, and Walker (1994:10) explain, “increased water withdrawal by one pumper reduces the water other pumpers obtain from a given level of investment in pumping inputs.” The problem of exclusion may be adequately addressed but the CPR may still be over used because of the harvesting actions of the resource users. Consequently, if common pool resources are to be governed sustainably the challenges posed by difficult and costly exclusion and substractability must together be addressed.

Considerable attention has been devoted to the problem of overuse. The earliest formal models of resource use, such as those developed for fisheries, focused on it, and most models since then have followed suit (Hardin 1968). While overuse is problematic, resource users are likely to confront a host of common pool resource dilemmas (Gardner, Ostrom, and Walker 1994). Gardner, Ostrom, and Walker (1994) define common pool resource dilemmas as suboptimal outcomes produced by the actions of resource users and the existence of feasible institutional alternatives that if adopted would lead to better outcomes (Ostrom, Gardner, Walker 1994:16).

Suboptimal outcomes are not limited to over use. Resource users may engage in a variety of actions that produce suboptimal outcomes in their use of a common pool resource. For instance, well owners may place their wells too close together interfering with one another’s pumping. Or, a farmer may install a deep tubewell near another farmer’s bore well drying up the bore well. Or, farmers may fail to maintain a tank that would otherwise serve to capture rainwater and recharge it into the underground aquifer. Ostrom, Gardner, and Walker (1994) relaxed the implicit assumptions underlying the formal models that focused on overuse to develop a typology of common pool resource dilemmas. Most models assumed a uniformly distributed resource. By relaxing that assumption and allowing resources to be patchy, so that some areas of a resource are more productive than others, assignment problems may emerge. Assignment problems involve resource users competing over productive areas and interfering with one another’s harvesting (Ostrom, Gardner, and Walker 1994:11).

Furthermore, most formal models assumed identical harvesting technologies among resource users. By relaxing that assumption and allowing diverse technology utilization, technological externalities may emerge among resource users. Technologies used by harvesters interfere with one another causing conflicts among resource users. For instance, a high capacity well may dry up a shallow tubewell (Ostrom, Gardner, and Walker 1994:12). Thus, in addition to overuse, or what Ostrom, Gardner, and Walker (1994) term appropriation externalities, resource users may experience assignment problems and technological externalities.

As Ostrom, Gardner, and Walker (1994) note, appropriation problems stemming from when, where, how, and how much to harvest are not the only problems resources users are likely to experience. Another class of dilemmas, provision problems, is also likely to emerge in many common pool resource settings from time to time. Provision problems relate to developing, maintaining, and/or enhancing the productive capacity of the common pool resource. For instance, adequately functioning canal irrigation systems require that diversion structures, headworks, canals, and outlets be regularly repaired and maintained. The productivity of a groundwater aquifer may be enhanced by capturing water during wet seasons and directing that water underground to be used during dry seasons. Provision problems are distinctly different from appropriation problems. Appropriation problems require resource users to coordinate their harvesting activities; provision problems require resource users to cooperate and contribute to the production of public goods.

The Ostrom, Gardner, and Walker (1994) typology not only provides scholars with a systematic language for identifying common pool resource dilemmas, but it also provides other advantages. First, it alerts analysts that dilemmas have distinctly different causes and dynamics. Farmers organizing to develop a groundwater recharge basin to replenish an aquifer is different from farmers organizing to develop well spacing rules. Second, because dilemmas have different causes, they often require distinct solutions. Attempting to solve appropriation externalities may not have any effect on technological externalities. Addressing technological externalities is unlikely to provide relief from provision problems. Each type of dilemma will require its own solution. Third, it is critical to carefully identify the types of dilemmas that resource users are struggling with if appropriate solutions are to be devised.

The Emergence of Self-Governing Institutional Arrangements

In attempting to resolve common-pool resource dilemmas, appropriators must work through three issues – supply, commitment, and monitoring (Ostrom 1990). Supplying institutional arrangements that may resolve common-pool resource dilemmas confronts appropriators with collective action problems – obtaining voluntary contributions of time and resources needed to identify and negotiate a set of rules acceptable to most appropriators. Even if appropriators successfully supply a set of rules, those arrangements, to be credible and effective must be followed most of the time by most appropriators. Appropriators must commit to following the rules. In many instances, however, appropriators will face temptations to disobey the rules. As Ostrom (1990:44) asks: “how does one appropriator credibly commit himself or herself to follow a rotation system when everyone knows that the temptation to break that commitment will be extremely strong in future time periods?” Monitoring to ensure that most appropriators are following the rules most of the time supports commitment to following the rules. Effective monitoring discourages rule breaking, and assures rule followers that they are not being taken advantage of by rule breakers. Yet, monitoring is itself a public good – it accrues to the benefit of all appropriators regardless of whether all appropriators contributed to monitoring. But without monitoring, commitments to following the rules are not credible, and with out credible commitments to the rules, no rules will be devised and adopted (Ostrom 1990:45). Thus, the process of devising, implementing, and sustaining institutional arrangements that resolve common-pool resource dilemmas is fraught with difficulty.

Ostrom (2001) argues that resource users are more likely to supply or invest in designing and adopting rules if they perceive that, 1) the benefits produced by the new sets of rules outweigh the costs of devising, monitoring, and enforcement, and 2) they will enjoy those benefits. Whether these two conditions hold depends on characteristics of the resource and characteristics of the resource users. For Ostrom (2001) four resource characteristics are crucial:

1. 
   Feasible improvement: Resource conditions are not at a point of deterioration such that it is useless to organize or so underutilized that little advantage results from organizing.
   

2. 
   Indicators: Reliable and valid indicators of the condition of the resource system are frequently available at a relatively low cost.
   

3. 
   Predictability: The flow of resource units is relatively predictable.
   

4. 
   Spatial extent: The resource system is sufficiently small, given the transportation and communication technology in use, that appropriators can develop accurate knowledge of external boundaries and internal microenvironments (Ostrom 2001: 40).
   

There must be a sense among resource users that governance attempts will make a difference (attribute 1). If a resource is so degraded that users believe there is little they can do to positively affect the situation, then they are unlikely to make the attempt. Conversely, appropriators may find little benefit in investing in governing arrangements if a resource is relatively abundant and of adequate quality. Whether resource users believe that feasible improvement in the productivity of the resource is possible depends on the information that they have and their ability to exercise some control over the resource. Information about a resource depends on availability of reliable and valid indicators of resource conditions, the spatial extent of the resource, and the predictability of resource units (attributes 2-4). Indicators vary from resource to resource and may be as “simple” as paying attention to wool or milk production of grazing animals or as complex as monitoring wells. The spatial extent of a resource affects both the ability of users to develop information and to assess their relative ability to capture the benefits of organization. Resource systems or subsystems that are more closely matched with the ability of resource users to monitor encourage investment in rules. Finally, predictability should be interpreted broadly to include volume and temporal and spatial patterns. Predictability not only provides resource users the opportunity to learn about the resource but to also govern harvesting activities in meaningful ways.¹

In addition to characteristics of resources, qualities of the resource users themselves affect the benefits and costs of cooperation to devise governing arrangements. Ostrom (2001) posits the following attributes of resource users:

1. 
   Salience: Appropriators are dependent on the resource system for a major portion of their livelihood or other important activity.
   

2. 
   Common understanding: Appropriators have a shared image of how the resource system operates…and how their actions affect each other and the resource system.
   

3. 
   Low discount rate: Appropriators use a sufficiently low discount rate in relation to future benefits to be achieved from the resource.
   

4. 
   Trust and reciprocity: Appropriators trust one another to keep promises and relate to one another with reciprocity.
   

5. 
   Autonomy: Appropriators are able to determine access and harvesting rules without external authorities countermanding them.
   

6. 
   Prior organizational experience and local leadership: Appropriators have learned at least minimal skills of organization and leadership through participation in other local associations or studying ways that neighboring groups have organized (Ostrom 2001: 40).
   

These characteristics ease the costs of organizing, developing, and adopting a common set of rules. Attributes 1 and 3 measure how appropriators value the resource. If resource users are heavily dependent on the resource for their livelihoods and if they anticipate continued reliance on it well into the future, they are more likely to invest in new sets of rules. If appropriators share a common understanding of the resource and the effects of their actions on the resource and on each other, they are more likely to share a common understanding of the problems that they face and are more likely to agree upon a set of rules to address those problems. Trust and reciprocity and leadership provide resource users with “social capital” that they can draw upon to ease bargaining and negotiation costs. Autonomy provides appropriators with the “space” needed to engage in rule making and confidence that they will be able to capture the benefits of their institutional investments. While Ostrom (2001) separates the two sets of attributes for the sake of clarity, the attributes interact to support or discourage collective action. Resource users may have a relatively complete and accurate understanding of the resource; however, they may still be unwilling to invest in new sets of rules if the resource is of low salience to them.

These attributes are well illustrated in case studies of two farmer managed irrigation systems in North Sulawesi, Indonesia by Vermillion (2000). Both irrigation systems are small and were relatively young when Vermillion (2000) studied them, one was five years old and the other twelve years. Vermillion (2000) examined the original rules adopted and the process by which irrigators were revising the original rules as they learned more about the physical system and one another. The Mopugad system serves 28 hectares. Farmers built a concrete weir to divert water from a small river. There is no storage in the system and stream flows “can be rather erratic and unpredictable” (Vermillion 2000:59). The Werdi Agung system serves 50 hectares. Initially, farmers built a weir to divert water from a small river, however, later the Indonesian government built an irrigation system whose main canal cut through the farmers’ canal. The Werdi Agung system now receives it water from the main canal of a government built and managed system. Its water flows fluctuate less than in the Mopugad system, however, farmers still experience unpredictability and scarcity (Vermillion 2000: 61).

Farmers in both systems were part of a resettlement program of the Indonesian government. The farmers were relocated from Bali. While they are relatively homogeneous in terms of landholdings, in part because the Indonesian government allocated 2-4 hectare sized plots, the farmers are socially heterogeneous. As Vermillion (2000: 61) states: “The villages of Mopugad and Werdi Agung are not natural or indigenous groupings but are amalgams of Balinese from different parts of Bali and even include some Christians amongst the dominant Hindus”.

Aspects of the social and physical settings are conducive for collective action. While some of the farmers have access to limited, alternative sources of water, all rely on the villages’ irrigation systems to annually raise two crops of rice. The farmers have experience with local level governance through their participation in subaks in Bali. They have adopted the subak system and used proportional allocation, based on size of landholdings, as their initial water allocation rule, a rule common among Bali subaks. In addition they share common norms around sharing, reciprocity, patience, and shame (Vermillion 2000:64-65). Given the relatively modest size of both systems, farmers in each have a basic understanding of the physical system and the types and quality of landholdings. The farmers, however, are still learning about the physical system as they develop additional rice paddy, and as they come to a better understanding of the soil qualities over the course of several growing seasons.

Vermillion (2000) was struck by the differences between the formal rules adopted by the subaks and working rules that the farmers appeared to be developing. As Vermillion (2000:61-62) explains:

Preliminary interviews with farmer leaders indicated that service area proportionality was indeed the basic rule for water allocation among farmers’ fields. But the first several inspections of both systems revealed something else going on in practice…There was an apparent rampant disregard for the traditional tektek, or temuku – a wooden log cut out so as to divide water proportionately.
Through further investigation Vermillion (2000) discovered that farmers frequently, but temporarily, took additional water. Neighboring farmers affected by such actions would be informed, if they were present in their fields. It was up to the farmers affected to return the diversions to the standard setting because farmers should regularly check their fields and their water diversions. Accepted reasons for taking additional water centered on physical circumstances. For instance, a farmer’s fields may consist of particularly sandy soils, or a farmer may be constructing new rice terraces (Vermillion 2000:66). The farmers used the newness of the irrigation systems to justify experimenting with variations of the proportional allocation rule. They were still learning about variation in the physical system and one another’s farming practices (Vermillion 2000:65). Vermillion (2000:77) notes that the adjustments in water allocation have tended to support more equitable allocations of water. Fields at the end of canals tend to benefit. He anticipates that over time the working rules for water allocation that farmers are developing will eventually be formally adopted by the subaks (Vermillion 2000:67).

The Persistence of Self-Governing Institutional Arrangements

Ostrom (2000) proposes that the factors that support institutional change are different from the factors, or principles, that account for the persistence of self-governance and the institutional arrangements adopted by irrigators. Whether the subaks and the sets of rules that the farmers of Mopugad and Werdi Agung, are experimenting with and adopting will persist depends on a number of design principles. Ostrom (1990:90) was not willing to propose the principles as necessary conditions for long term success, however, the principles do support the continued success of institutional arrangements for sustaining common-pool resources and for gaining the compliance of generations of appropriators. The design principles are (Ostrom 1990:90):

1. 
   Individuals or households who have rights to withdraw resource units from the CPR must be clearly defined, as must the boundaries of the CPR itself.
   

2. 
   Appropriation rules restricting time, place, technology, and/or quantity of resource units are related to local conditions and to provision rules requiring labor, material, and/or money.
   

3. 
   Most individuals affected by the operational rules can participate in modifying the operational rules.
   

4. 
   Monitors, who actively audit CPR conditions and appropriator behavior, are accountable to the appropriators or are the appropriators.
   

5. 
   Appropriators who violate operational rules are likely to be assessed graduated sanctions (depending on the seriousness and context of the offense) by other appropriators, by officials accountable to these appropriators, or by both.
   

6. 
   Appropriators and their officials have rapid access to low-cost local arenas to resolve conflicts among appropriators or between appropriators and officials.
   

7. 
   The rights of appropriators to devise their own institutions are not challenged by external governmental authorities.
   

8. 
   Appropriation, provision, monitoring, enforcement, conflict resolution, and governance activities are organized in multiple layers of nested enterprises (Ostrom 1990:90).
   

Design principle one, exclusion, is critical if appropriators are to commit to following a set of institutional arrangements over time and investing in modifying them as circumstances warrant. Appropriators must be assured that they will capture the benefits of their actions. Exclusion, while critical, is insufficient, however, to ensure long-term commitment to the rules. The rules themselves must make sense, they must be crafted to the exigencies of the situation, and as the situation changes, appropriators must have the ability to modify the rules. Accountable monitors and graduated sanctioning maintain appropriators’ commitment to institutional arrangements. In many instances, the rules support monitoring by appropriators while they are using the common-pool resource. Monitoring is important to assure resource users that most of them are following the rules most of the time. Graduated sanctioning works to bring rule violators back in line with the rules. Only if an appropriator persists in violating rules do sanctions become severe and are social ties broken. Finally, conflict resolution mechanisms and at least a minimal recognition of the right to organize prevent these institutional arrangements from unraveling due to internal strife or interference from external governmental authorities.