Previous PageTable Of ContentsNext Page

Decentralization and Devolution in Forest Management:
A Conceptual Overview

R.J. Fisher
Head of Program Development
Regional Community Forestry Training Center
Bangkok, Thailand


The terms decentralization and devolution, along with a number of associated terms (such as participation and power) are now used widely in discussions about forestry policy. The terms are, however, used to describe a wide range of different approaches and processes to forest management and are often wrongly interchanged. The first aim of this paper is to define more clearly what the various terms and concepts mean. The second aim is to look at some ways in which the concepts have been applied in practice and to identify implications from these experiences to guide future attempts to apply decentralization or devolution to forest management practices.

This paper is a critique of much of what has been described as decentralization and devolution in the past, not a critique of the ideas (or ideals) behind decentralization and devolution. It supports the call for devolution made by Dr. Banerjee (see Ajit Banerjee in this volume). The paper is not intended to be a comprehensive analysis of the different decentralization processes occurring throughout the region, but rather an effort to encourage more explicit awareness of what devolution or decentralization mean both theoretically and in practice.

Definitions and Concepts

Four key concepts will be discussed in this section: decentralization, devolution, power and participation.

The terms decentralization and devolution are often used more or less interchangeably, but it is important to differentiate between the two. Decentralization can best be defined as relocating administrative functions away from a central location. This does not necessarily involve changing the locus of decision making, or devolving power. In distinction to this, devolution can be understood as relocating power away from a central focal point.

In this context, power can be defined as the capacity to affect the outcome of decision-making processes. It is important to stress that this implies a genuine role in decision making, not just a token input in the form of "consultation". These definitions are not precise, but the distinction reflects common usage and, more importantly, it is useful for purposes of analysis.

Another important aspect of both decentralization and devolution is the direction which either one can take. In the forestry context, the terms are used to describe the relocation of administrative functions and/or power from a central location or focal point to:

Participation is another term which is used vaguely. Arnstein (1969) developed a typology of ways in which the word is used. These eight levels are progressively ranked in a ladder which groups various usages into more general categories ranging from "non-participation", "degrees of tokenism" to "degrees of citizen power" (see Figure 1). When we look at many so-called participatory projects it is clear that they fall within the lower levels of the ladder. The organizing principle of this ladder is the degree of power people have over decisions that are made. This notion of participation ranked in terms of degrees of power over decision making is a useful entry point into discussions of what is really devolved and decentralized in forest management.

Review of "Participation" in Forest Management

Several conclusions can be drawn from any modestly informed look at what passes as participatory forestry:

The last point is made clear by Banerjee (2000) who reviewed the participatory forest management experiences of a number of Asian countries where he had personal experience. He drew the following conclusions:

Discussion of people's participation in forest management often revolves around complaints that the "people do not want to participate in forest management". Sometimes the reason for this unwillingness to participate is attributed to a lack of knowledge and skills or a lack of understanding regarding the importance of forests. Sometimes it is seen as being the result of the absence of adequate "participatory methodologies" ("if we just had better tools and techniques, we could get people involved"). It is likely, however, that the more common reason is that people do not "participate" because they see no point in promoting someone else's agenda. We return, thus, to the question of power to influence management decisions.

This failure to devolve power is not just a matter of not devolving power to communities. Many attempts at administrative decentralization and devolution exhibit a common pattern in which the "periphery" is expected to implement objectives set at a central level. Very often there is a discrepancy between the responsibilities people are given and the rights and powers they have (including the power to act on their responsibilities). Devolution of responsibilities is rarely accompanied by devolution of authority. For example, forest staff are often given the responsibility to encourage local people to participate in forest development activities, for which there is a budget, but do not have the authority to make agreements about the use of that budget. Agreements made in good faith by field staff are often reversed by senior officials, with the result that confidence in field staff by local people declines and efforts to increase local support fail.

Misapplying Devolution and Decentralization

In addition to the basic problem of devolving true power responsibility to implement policies, there are three major ways in which decentralization and devolution have been misapplied:

As an example of the pressures to impose a standardized organizational model, it is useful to look at Indonesia in 1998. In the spirit of reformasi associated with the fall of the Suharto Government, there were immense pressures to make major reforms to forest policy, including strong emphasis on recognition of the needs of local people. The Minister of Foresrty at the time was a strong supporter of cooperatives and there was an assumption that local participation would flow in the form of standardized cooperative arrangements, despite the risks that cooperative arrangements would probably not be appropriate in all (or even many) situations. To work, cooperatives would need to be appropriate for a variety of situations such as cases where voluntary groups of farmers were involved in timber production or where traditional shifting cultivators were involved in regulating farming on individual plots through existing social arrangements. Cooperatives might work in some cases, but would be problematic in others.

The problem with standardized models is that they often ignore what has been learned about the social basis for effective local action. A brief history of the evolution of community forestry in Nepal illustrates some of the issues associated with applying devolution and decentralization at the community level in forestry. The Nepal study illustrates the misapplications and also shows ways they can be addressed.

The Nepal Experience

Following the nationalization of private forests in 1957 and the collapse of the feudal system in the early 1950s, it became increasingly clear that the government could not manage forests effectively without local people being involved. It was also recognized that rural people had legitimate and urgent needs that could only be met from forests. The first real community forestry policies were initiated in 1978 with the introduction of Panchayat Forest and Panchayat Protected Forest Rules under which specified areas of forest could be handed over to local politico-administrative units called Panchayats for reforestation of degraded forests (Panchayat Forests) or protection in the case of existing forests (Panchayat Protected Forests).

This proto-community forestry was modestly successful in achieving local support, but was limited by the fact that neither category of forest allowed any significant forest use unless there was also an approved forest management plan. These were very rare. Major rethinking became evident in a watershed Workshop on Community Forestry Management in 1987 and was officially enshrined in Nepal's Master Plan for the Forestry Sector in 1988. The revised policy encouraged transfer of forests to local communities for active management and use. Emphasis in the field shifted to developing operational management plans with communities as a prerequisite to handing over forests for their use.

But major institutional barriers remained. One of these was the continuing requirement that the Panchayats should be involved, although it had already been established that smaller groups of traditional users, living close to forests and with informal locally recognized access rights, were more appropriate managers. In fact, handing forests over to the Panchayat was often seen by traditional users as giving away their forest to those who had no active interest in their management. The problem was that there was a strong official ideology that the Panchayat system was a unique and indigenous non-party solution to governance in Nepal. This made it impossible to ignore the village Panchayats in community forestry arrangements. There were also some major "institutional incompatibilities" between the forest department and local communities that limited the rate of hand over of forests to communities (Fisher 1990). These incompatibilities included:

Research into local forest management systems and lessons from field experience led to an improved understanding of these difficulties, but it was only with the collapse of the Panchayat ideology after the 1990 revolution, that community forestry was able to shift focus from local Panchayat officials towards user groups. Legislation in 1993 further entrenched the user group approach. These changes led to a rapid increase in the number of community forestry management agreements. By 1998, the government had handed over more than half a million hectares of forest to roughly 5,000 registered user groups.

The community forestry program is very popular with rural people, and demand for development of operational plans and hand over remains high. There are, however, signs that many forest officials would like to limit the program in various ways.

The main point emanating from the experience of Nepal relates to the increased success of participatory forest management which arose from a better understanding of what is involved in successful local collective action. Lessons about the factors that enable people to work together to manage forests were incorporated in policy and implementation. It was only after power was devolved to the actual users that community forestry in Nepal began to take off.

What is Known about the Social Basis for Collective Action?

There is a vast body of literature on the management of forests as common property and a great deal has been learned about the social basis for effective collective action at the local level (see Ostrom 1990; Fisher 1994). Essentially, forest management by local people requires people to agree or consent to some forms of regulated access to, and use of, forest resources. This involves cooperation, negotiation and institution building. So, what is known about the types of conditions necessary for collective action? We know that:

All of these points have great importance for how devolution or decentralization is applied at the community level. Where communities are defined in terms of the formal political and administrative structure, there are real risks that responsibility and authority will be applied to a "community" level that is inappropriate in terms of what is known about effective collective action.

The tendency to focus too much on the formal local political structures makes sense to bureaucracies, for several reasons:

But, there are major disadvantages:

Conclusions and Implications

What are the implications of these observations for the implementation of devolution and decentralization forest management policies?

The first implication (applicable whether the shifts are to local administrative units, local political structures or to "natural" user groups) is that effective decentralization or devolution requires devolved decision making and the need to support devolved responsibilities with power and authority. The second implication, applicable at the community level, is that the process needs to be informed by an understanding of the social basis of local (collective) action. A third implication (not explicitly discussed in this paper, but arising from the discussion) is that we need to think of ways to combine the concerns of local governments with those of other non-formal groups.


Arnstein, Sherry R. 1969. A Ladder of Citizen Participation. Journal of the American Institute of Planners, Vol 35: 216-224.

Banerjee, Ajit K. 2000. Foresters' Reactions to Community Forestry Paradigm's in Selected Countries. In Victor, M., Barash, A. and J. Bornemeier. (eds.), Cultivating Forests: Alternative Forest Management Practices and Techniques for Community Forestry. Proceedings of an the International Seminar held in Bangkok, Thailand from 23-25 September 1998. RECOFTC Report No. 17. Bangkok, Thailand

Fisher, R.J. 1990. Institutional Incompatibility in Community Forestry: The Case of Nepal. In Fox, J and R.J. Fisher (eds), Community Organizations and Government Bureaucracies in Social Forestry. Working Paper No 22, Environment and Policy Institute, East-West Center, Honolulu.

Fisher, R.J. 1994. Indigenous Forest Management in Nepal: Why Common Property is Not a Problem. In Michael Allen (ed), Anthropology of Nepal: People, Problems and Processes. Kathmandu: Mandala Book Point. pp 64-81.

Ostrom, Elinor. 1990. Governing the Commons: The Evolution of Institutions for Collective Action. Cambridge: Cambridge University Press.

Uphoff, Norman. 1992. Local Institutions and Participation for Sustainable Development. Gatekeeper Series. No 31. London: International Institute if Environment and Development.

1 These limitations apply even when the formal political system involves representative democracy. In representative democracy, there is often a high level of opposition to majority decisions. Where outcomes cannot be enforced, consensus or consent are necessary.

Previous PageTop Of PageNext Page