2. NRM, INSTITUTIONS FOR COLLECTIVE ACTION, AND DECENTRALIZATION

In the absence of a commonly agreed definition, this section begins by explaining what is generally meant by the term “natural resource management”. Following this, an analysis of the nature of natural resource goods concludes that NRM programs provide joint products wherein an activity gives rise to multiple outputs, some of which are private, purely public, and impurely public. The complex nature of natural resource goods, therefore, does not assist in determining whether the State has responsibility for their provision. However, empirical evidence from successful projects demonstrates that institutions for collective action are essential for sustainable NRM because they are able to solve the commitment and free-rider problems. These institutions can be considered to be public (or collective) entities. It is argued that the state has a responsibility for reducing environmental degradation which is a function of market failure and therefore, by default, it has a responsibility to create the enabling conditions for the supply of NRM institutions. An analysis of long enduring institutions for collective action, found that they are characterized by having considerable local authority and control over design of governance rules, decision-making and financial matters. Consequently, it is concluded that the supply of effective institutions for collective action requires state adoption of decentralized political, fiscal and institutional systems of governance.

2.1 DEFINING NATURAL RESOURCE MANAGEMENT

Natural resource management (NRM) within the Bank refers to three basic types of lending operations: (i) those designed to promote sustainable agricultural, forestry and fisheries development and/or water resource use through environmentally sound resource management techniques; (ii) those intended to conserve or protect specific ecosystems and associated biodiversity, including the establishment or consolidation of national parks, wildlife reserves, etc.; (iii) those which seek to strengthen national and/or subnational institutional capacity for improved NRM as, for example, through the use of rural land cadastres, geographic information systems, and other natural resource and environmental planning tools. A recent review of the Bank’s NRM portfolio notes that, in reality, many of the projects are “hybrids” in that they combine sustainable production, conservation and institution building objectives and components (World Bank, 1997a). Box 1 describes five types of NRM activities that fall within the first category of Bank lending operations.

The review of the Bank’s NRM portfolio (World Bank, 1997a) found that projects designed to promote sustainable agricultural production through improvements in on-farm soil and water management, such as social forestry, micro-watershed management, irrigation water management and soil conservation, had a greater likelihood of success. This is because they have considerable advantages over projects that are primarily concerned with biodiversity, forest, rangeland, or fisheries resource management and conservation. In particular, the former types of projects have two significant characteristics not always shared by the latter. These characteristics are related to the externalities affecting resource management viability:

(a) The projects are “win-win” in that they directly combine economic and environmental benefits - i.e., on-farm technical improvements in NRM enhance or extend natural soil fertility and, hence, crop productivity, with a positive potential impact on beneficiary income, while at the same time reducing erosion, sedimentation and agricultural run-off. The externalities affecting economic viability are therefore primarily local.

(b) The projects necessarily entail decentralized actions (at the municipal and micro-catchment levels) and require active beneficiary participation. Seeking commitment of resource appropriators at this micro-level through institutions for collective action is more viable than attempting to take decisions at a regional level because potential conflicts are localized. Externalities at regional levels do exist, but they do not affect the viability of local decision-making through local institutions.

In contrast, projects involving forest, rangeland, and fisheries resource management require some restriction of access to natural resources. They also involve significant trade-offs and conflicts because the externalities affecting the viability of resource use are usually not only local, but also regional, national and international. Moreover, there is insufficient understanding of the underlying ecosystem services and constraints which makes project design, implementation, and potential success, even more difficult. Box 2 provides examples of institutional channels for decision-making by level and sector.

This paper seeks to identify design characteristics of those NRM programs that lend themselves to local action by resource appropriators; the regional and national externalities therefore do not significantly impinge on the viability of local resource management interventions. Examples of these types of NRM programs are micro-watershed management, social forestry, irrigation water management, and soil conservation. These NRM projects are dependent, at least partially, upon on-farm technical improvements that result in on-farm income benefits. Obviously, there are also some regional, national and international externalities such as commodity prices, or use of common property (e.g. groundwater or common land), that do affect the viability of sustainable resource management practices. In general, however, these types of resource management activities, through micro-level planning and local action, provide greater opportunities for “win-win” outcomes as described above.

Source: Uphoff (1992)

2.2 THE NATURE OF NATURAL RESOURCE GOODS

A theoretical analysis of the nature of natural resource goods is required in order to determine whether the responsibility for their provision is public or private. This is important because some analysts suggest that if NRM interventions are on private land and provide private benefits, then the state has no role in their provision; it should be left to the private sector. On the other hand, theory on public goods states that if the NRM goods are public, then it is unlikely that there will be a market for them. It is therefore unlikely that a private supplier will provide those goods. The responsibility for their provision is therefore public and they need to be financed out of general taxation.

Natural resource systems and whether they are public or private responsibility is particularly complex because of the multidimensional nature of the resource. In order to understand the complexity, it is useful to distinguish between resource systems and the flow of resource units produced by the system. Examples of resource systems include groundwater basins, grazing areas, irrigation canals, forests, streams, lakes, etc. (these are often referred to as common pool resources - CPRs). Resource units, in contrast, are what individuals use from resource systems. They are typified by the tons of timber harvested from a forest, the cubic meters of water withdrawn from a groundwater basin or an irrigation canal, the tons of fodder consumed from a grazing area, and so on. As long as the average rate of withdrawal of resource units does not exceed the average rate of replenishment, the resource system can be considered to be sustained over time.

Without distinguishing between the resource system and the resource units, theory on public good provision has limited potential in explaining the nature of natural resource goods. For example, the cubic meters of water withdrawn from an irrigation canal and used on a farmers land is not available for use by another individual. The resource unit is therefore not subject to joint use and can be considered to be a private good. However, the water withdrawn from an irrigation canal by one farmer will limit the amount available to another farmer. The resource system, therefore, is subject to joint use and can be considered to be a public good. Furthermore, if improvements are made to the resource system (e.g. de-silting a canal or fixing the embankments) the resultant improvements in availability of water will be simultaneously available to all users of the system. Public goods theory is based upon the notion of nonsubtractive attributes of goods; public goods being those which are nonrival and nonexcludable. Private goods, on the other hand, are fully rival and excludable.^[2] Ostrom (1990) and Ostrom and Gardner (1993) argue that it is the theory of private goods which is more applicable to the use of resource units while the use of a resource system is more related to the theory of public goods. Given current theories, it would appear that, at best, one can describe natural resource management programs as providing joint products, wherein an activity gives rise to multiple outputs, some of which are private, purely public, and impurely public. Unfortunately, this does not provide much clarity for justifying the role of the state in the provision of NRM goods.

Approaching the issue from the perspective of institutional economics offers greater clarity in defining the role of the state. Let us assume that natural resource degradation is characterized by having high social costs (i.e., private costs plus the economic costs associated with degradation of the resource base). These negative external effects of natural resource degradation imply a market failure and therefore a sub-optimal allocation of resources. From a theoretical point of view, market failure on its own is a sufficient requirement to justify state intervention. The objective of state intervention being to internalize the external costs (that is the degradation), in order to limit the production of the “public bad”. The justification for state intervention can be strengthened if one analyzes programs which have successfully fostered sustainable management of natural resource systems. In these programs, it becomes apparent that the critical factor in their success is the formation of local level institutions for collective action (this is discussed in more detail in the next section). Institutions can be viewed as a club of resource users sharing common interests; they are not “goods” but “public-like entities” that provide the means by which joint products can be provided. Since the state should promote optimal allocation of resources, it should support such institutions. As will become apparent in this paper, the state cannot supply these institutions because that would undermine the quintessential features that make them successful. Instead, the state should promote institutions through creating an enabling environment for their evolution through decentralization policies and associated legislation; directly catalyzing their establishment through program interventions; providing financial support, and so on. Thus, the “public good” provided by the state is a reduced rate of resource degradation, and the institutions are only an input in the provision of this good.

2.3 THE NECESSITY FOR COLLECTIVE ACTION TO MANAGE NATURAL RESOURCES

Policy prescriptions for preventing the erosion of the natural resource base assume that people are locked into a remorseless tragedy wherein through rational individual choices they have negative impacts on a resource. The origins of these policies can be partially attributed to the models of human behavior that were evolved in response to Garret Hardin’s seminal 1968 article in Science. This drew attention to the degradation of the environment when individuals use a scarce resource in common. His article coined the phrase “tragedy of the commons” to symbolize the problem which affects much of the world’s resources. According to Ostrom (1990), the prisoner’s dilemma game (and developments of it) have been widely used to explain the causes leading to the tragedy of the commons. The game is conceptualized as a non-cooperative game in which all players have complete information. Communication among the players is forbidden or impossible or irrelevant as long as it is not explicitly modeled as part of the game. In essence, this model postulates that individually rational strategies lead to collectively irrational outcomes thereby leading to the degradation of the commons. Contemporary to Hardin’s article was another model put forward by Olson (1965) that sought to explain the logic of collective action. This stated that the possibility of a benefit for a group is NOT a sufficient requirement to generate collective action to achieve that benefit. The pessimism of both these models is due to the fact that at the core of each of them is the free-rider problem; whenever one person cannot be excluded from the benefits that others provide, each person is motivated not to contribute to the joint effort, but to free-ride on the efforts of others. The problem is particularly severe in the case of NRM systems because of the complex nature of the goods. The result is that some people may provide while others free-ride, leading to a less than optimal level of provision of the collective benefit.

Policies seeking to address this paradox have either suggested that the State has to intervene to protect the common resources, or private rights need to be developed for some common-pool resources (CPRs). Empirical studies of the impact of these policies is leading to a growing acceptance that external authorities have limited potential to impose protection of natural systems and that private (or individual) actions only provide partial protection (Pretty, 1995). It is increasingly recognized that more coordinated responses and multiple solutions to the complexities of sustainable resource management are required.

The situation is not as gloomy as it might appear. Recent empirical evidence from analysis of long enduring institutions of collective action and also from experiences of non-governmental organizations (NGOs) which have successfully implemented community-based NRM programs is providing some scope for optimism. It is now indisputable that the promotion of sustainable natural resource systems is dependent upon involvement of stakeholders in the design, implementation, operation and maintenance of projects (Narayan, 1995; Pretty, 1995; World Bank, 1997a). Key to these community-based approaches has been the formation and strengthening of local organizational capacity (LOC) wherein resource users are organized into local institutions for collective action.^[3] Sustainable management of natural resources depends not just on the skills and knowledge of individual users, but on action taken by local groups or communities as a whole.

Current theories on human organization, unfortunately, are unable to adequately describe how a group of individuals can organize themselves voluntarily to retain the residuals of their own efforts. Most of the recent discourse on institutions for collective action has focused on aspects related to their governance and sustainability. Few authors have addressed the issue of supply of new institutions. Bates (1988) argues that, even if the benefits from collective action are symmetric and all persons are made better off from the introduction of a new institution there would still be a failure of supply. This is because the institution is a public good and rational individuals (adhering to the prisoner’s dilemma model) would seek to free-ride and secure its benefits for free. Therefore the incentive to free-ride would undermine the incentives to organize a solution to the collective dilemma. Empirically, however, it is known that some communities are able to overcome the incentive to free-ride and they do voluntarily establish new institutions. Bates’s attributes this to communities having established a sense of trust and cooperative behavior, which he suggests are the mechanisms for solving the problem of supplying new institutions. Without explicitly stating it, Bates is creating the case for social organizing and community mobilization in order to establish trust and build social capital as a means of enabling communities to overcome the incentive to secure benefits for free and to form an institution for collective action.

Having established that the critical issue is one of supplying institutions, it is necessary to ask the question: what kind of institutions are we seeking to supply? The next section looks at long enduring institutions and their similarities.

2.4 SIMILARITIES BETWEEN SUCCESSFUL AND LONG ENDURING INSTITUTIONS

This study defines effective community-based institutions as those which enable individuals to achieve productive outcomes in situations where the temptations to free-ride are ever present. These local institutions must be long enduring and continue to function after external program interventions cease. This section presents the current knowledge on the essential elements that constitute a long enduring and sustainable institution for collective action. The elements are synthesized by investigating institutions involved in the management of CPRs, farmer organizations, and institutions formed for the implementation and management of rural infrastructure.

2.4.1 Eligibility for institution membership and the resource boundary must be clearly defined

This is usually the first step in organizing for collective action. The eligibility criteria will depend upon the objectives of the institution or program which is seeking to catalyze collective action. Some programs such as the NRDP in Brazil encourage the formation of institutions that are targeted at poor households; large land-owners and rich individuals are excluded from membership. The Aga Khan Rural Support Programs (AKRSPs) in Pakistan and India, originally sought to encourage the formation of broad-based village institutions where all households would be members. However, they soon realized that women were not participating in the village institutions and consequently they adopted and strategy of fostering separate institutions for women in order to create a conducive environment in which women were able to develop their confidence levels and prioritize their own needs without external pressure from men or from cultural norms of behavior.

Ostrom (1990), Hobley and Shah (1997) and Farrington and Thiele (1997) note that individuals who have rights to withdraw resource units from a resource system must be clearly defined, as must the boundary of the CPR. Without defining the boundaries of the CPR and closing it to “outsiders”, local users face the risk that any benefits they produce will be reaped by others who have not contributed to those efforts. If this were to occur the discount rates (see section 0, page 23) would be pushed higher and the institution would soon collapse because the incentive for collective action would become unfeasible. Membership of the institution may occasionally have to include “outsiders” or non-residents of a village if they have customary (rather than legal) rights over use of a CPR - e.g. semi-nomadic herdspeople may have customary rights over use of common and private land despite not being resident in the area.^[4] Empirically, it is known that management of a common resource by collective action is easier if the boundary of a given resource is closely related to the administrative boundary of a village or community (e.g. village boundaries should correspond closely with those of a micro-watershed).

2.4.2 Rules governing resource appropriation (use) must be locally adapted

The rules governing the use of resources have to be developed by the members in order to be appropriately adapted to the local conditions. The rules evolved by the resource users need to govern when the resource can be used; the technology that is most appropriate for managing the resource system and appropriating the resource units; and the quantity of resource units that the users are allowed to appropriate. Obviously, the rules need to reflect the specific attributes of the resource; standardized rules cannot be used. Experience demonstrates that rules designed by members of the institution are more likely to be enforceable.

Rules evolved by resource users themselves do vary according to local conditions. For example, AKRSP supported irrigation schemes in India and Pakistan had different rules for levying water fees; some charged an amount based upon the time taken to irrigate a field and others calculated the fees based upon the quantity of land to be irrigated. The rules were devised by the institution members based upon their local knowledge; the involvement of the catalytic external agency was merely to ensure that some kind of rules were in place prior to initiating construction. As will be discussed later in this paper, participatory planning methodologies are a useful means by which external agencies can intertwine local knowledge with external expertise. For example, they can advise resource users on rules which have been shown to be viable in other contexts.

2.4.3 Institutions must adhere to democratic principles

Institutions which are truly democratic are those which enable most individuals who are directly affected by the operational rules to participate in modifying those rules. Ostrom (1990) notes that institutions which adhere to this principle are better able to adapt rules to local circumstances because individuals interact with each other and the external environment. They are able to modify the operational rules over time and keep the institution relevant to the characteristics of their settings. Democratic institutions also facilitate impartial resolution of conflicts between members. The costs, however, must be kept low and the process of changing rules must be relatively simple.

As mentioned above, the rules governing resource appropriation need to be devised by members of the institution, on the other hand, there is a compulsive case for more external guidance on designing appropriate means of ensuring democratic participation. For example, in settings with resilient feudal structures, strong patron-client interdependencies and weak traditions of political freedom, it is more appropriate to have secret voting mechanisms rather than open voting arrangements. Moreover, agreements reached by so-called “consensus”, although en vogue, are fraught with danger; individuals who appear to be reaching common consent may actually be consciously deferring to the demands (or perceived wishes) of the strongest patrons or elites. However, external modes of democratization cannot be imposed on a given populace. The ideal mechanism for ensuring democratic participation needs to evolve through an iterative process whereby external knowledge is allied with traditional social and cultural practices.

2.4.4 Institutions must have authority and control over financial resources

This includes transferring responsibility and authority to local level institutions for management and allocation of financial resources, procurement of local goods and services. This relates to element No. 2 regarding rules governing appropriation of resources; if the members of the institution decide on the level of user fees they must also be responsible for collection and management of those fees. Experience has shown that governments and programs (including NGOs) are most reluctant to decentralize authority over financial matters to local institutions because they lack trust in the community’s capability to manage funds without excessive rent-seeking.

One of the indicators used by AKRSP (Pakistan) to assess the institutional maturity of village institutions was their efficiency and probity in handling of financial resources including a savings fund. Out of 120 indicators used for their study, this indicator was found to have a high correlation with institutional maturity and potential sustainability. A number of bank programs including the NRDP in Brazil and also Social Funds programs evaluated by Narayan and Ebbe (1997) have found similar results. Moreover, they have confirmed that the level of misappropriation under decentralized financial systems is no higher than under centralized control. In fact, the NRDP in Brazil found cost efficiency gains of 30% in comparison to centralized control when local procurement of goods and services was used for small infrastructure projects (World Bank, 1997b).

2.4.5 There must be local monitoring of resource use and graduated sanctions

Ostrom (1990) found that in long enduring institutions the monitors who actively audit CPR rules and appropriator behaviors were accountable to the resource users or are the resource users themselves. In the AKRSPs, the village institutions are responsible for devising rules governing appropriation and also responsible for ensuring that those rules are enforced. Sanctions are employed to penalize those individuals that do not conform to the rules which are collectively agreed. Monitoring and levying sanctions therefore are carried out by the participants themselves.

The sanctions employed are generally graduated to account for the severity of the offense. When sanctions are designed by the members of the institutions, experience shows that they are often graduated because the members know the personal circumstances of the infractors and the potential harm that could be caused by excessive sanctions. Mechanisms for local monitoring include: institutions hiring local people to guard CPRs (e.g. Joint Forest Management in India); a portion of the fines are kept by the guard; and, infractors lose status and prestige.

To summarize, when institutions are designed according to the above principles, they are able to solve the commitment and free-rider problems in an interrelated manner. Resource users design their own rules which are implemented and monitored by themselves, using graduated sanctions that based on a definition of who has rights to withdraw resource units, and which reflect local conditions.

2.5 DECENTRALIZATION POLICIES AND LOCAL COLLECTIVE ACTION

Forming and strengthening local institutions is itself a strategy of decentralization to create local capacities for handling authority and responsibility (Uphoff, 1986). However, for local institutional development to proceed far and be replicated on a national scale, a degree of decentralization within the government structure itself is required.

Institutions for collective action, as already explained, are public-like entities whose benefits are limited to a specific subset of the population, members of a single community or appropriators of a specific resource system. It is generally recognized that for providing goods that benefit a subset of the population, decentralization of the public sector provides considerable advantages including pareto-efficient levels of output; typically, decentralized regimes are more responsive to citizen preferences resulting in increased allocative efficiencies (Oates, 1972; Rondinelli, 1981; Huther and Shah, 1996). This assertion is corroborated by the above analysis of long enduring institutions whose effectiveness in governing natural resource systems were greatly enhanced by having increased administrative, political and fiscal autonomy.^[5] Decentralized policies, in effect, create the conditions whereby rights over governance of natural resources are “restored” to appropriators after the state has failed in its attempt to manage them through centralized systems.

The case for increased administrative and political decentralization is strengthened by the observation that long enduring institutions are characterized by those that allow resource users to devise their own rules governing the institutions. In contrast, highly centralized regimes attempt to rely on the same operational rules for all institutions. Under centralized regimes, local resource users would need to invest a considerable amount of time and resources in seeking a change to those rules or for a special case to be made for their locality. On the other hand, under a regime that allows substantial local autonomy to engage in constitutional and collective choices, resource users may be authorized to select their own rules so long as they follow certain procedures. The procedures may vary from informal mechanisms requiring consultation to formal mechanisms necessitating signed petitions, special elections, court proceedings, and so on.

Lutz and Caldecott (1996), in looking at decentralization and biodiversity conservation, found that decentralization improves conservation and management of natural resources. Empowered local institutions that were given authority and control over revenues and expenditure decisions also had more effective control over management of natural resources. There were however some inherent dangers in decentralizing resources and authority to local institutions. For example, local elites could capture resources and use the decentralization process to further their own short term private interests. Nevertheless, decentralization, coupled with appropriate safeguards, could yield substantial benefits for sustainable conservation of biodiversity.

The advantages of decentralization policies in creating the enabling conditions for the large-scale supply of institutions for collective action are borne out by the experiences of Brazil. The 1988 Constitution set the framework for a radical decentralization of fiscal, administrative and political authority to states, municipalities and communities. The Constitution also stated that all public funds for investments in communities had to be channeled through Community Associations who had fiscal authority over the use of those funds (see section 0, page 21). This single policy decision resulted in a phenomenal growth in the number of CAs which were formed and registered. While it is certainly true that the quality and effectiveness of the CAs varies considerably, decentralization in Brazil has enhanced the ability of local appropriators to engage in effective institutional design (see Esmail and McLean, 1997 forthcoming).

To briefly sum up. This chapter has demonstrated that the state has an implicit role to play in creating the conditions for the supply of local institutions. Decentralization policies offer considerable succor in this endeavor. As this paper proceeds, the necessity for decentralizing fiscal, political and administrative authority and control to local institutions and local government will become increasingly apparent.