How does collective action succeed in rural communities? This thesis investigates a forest planting project in the rural south of Malawi. Small scale farmers were ordered to abstain from cultivating their gardens on a mountain, as the whole mountainside land was designated to forest planting. The forest project hence turned into a conflict over land. It can be considered a collective action problem, because each individual farmer would be better off with continued cultivation on his own garden, while the community as a whole suffered from severe soil erosion due to the deforested mountain. In the long term, everybody would be better off with a forested mountain. The results of this collective action problem differed in the villages surrounding the mountain. In some villages, farmers were convinced to cooperate in the forest project, and the communities were rewarded with a valuable forest resource benefiting all inhabitants. Other villages failed, and today suffer from increasing problems of soil erosion. The main question of this thesis is why and how some villages reached the collectively successful results, while others did not. What were the differences between these villages and how may these differences have resulted in completely different outcomes in the forest project?
I analyse the decision faced by individual farmers of whether to contribute their land to the forest project or not. It is assumed that specific circumstances within the villages have affected the individual decisions and thus determined the success or failure of the forest project.
The empirical analysis is based on qualitative data collected during two months of fieldwork in the villages in 2007, with a focus on interviews and observations. The fieldwork revealed two main factors behind successful collective action; (1) a social norm supporting general co-operative behaviour, and (2) formal punishment of defectors, in a case in which the social norm was not strong enough to convince farmers of co-operation. Both regulations increase farmers’ incentives to co-operate, and thus contribute to preserve the common pool resource. A shared norm can support collective action by imposing loss of reputation for those breaking it and thus expectations about mutual co-operation. Norms develop continuously in a community and depend on legitimacy among inhabitants. Only one of the research villages had a strong norm supporting co-operational behaviour. Through the forest project and other village projects, they experienced that co-operative behaviour was beneficial, both because they became convinced that everybody else would co-operate and because the beneficial outcome of village projects was actually distributed fairly in the end. Previous successful experiences with collective action seem to have developed the social norm of co-operative behaviour in this village.
In other villages, there was little cooperation on village level and village resources were to a greater extent captured by a village elite. Villagers had less respect for their leaders and hence did not co-operate in the projects they introduced. Only one of these non-co-operative villages was able to preserve the planted forest, using strict regulation; the forest managers enforced hard punishments on non-co-operating farmers. This method had substantial drawbacks compared to the social norm method. As the punishments were not considered fair and legitimate among farmers, it actually increased the already high level of conflicts and anger towards the village leader. The strict enforcement of rules was crucial in order to preserve the forest, but it probably also gave negative external effects on social capital in the village.
There is a large literature on collective action on common pool resources. Standard economic theory proposes a result in which nobody co-operates, because everybody gains individually from defecting, given the choice of others. Hardin (1968) introduced this depressive result as the tragedy of the commons. His proposal is, however, not applicable in general, because it leans on the assumption that an individual’s choices are independent of his expectation about others’ choices (Runge, 1981). Common pool resources can be, and often are, regulated by local communities with reasonable degrees of success (Ostrom, 1990). It is these regulations which induce expectations among individuals about the behaviour of others. If individuals expect that everybody else will co-operate, they might benefit from choosing co-operation, because they know that non-co-operative behaviour may spread. Hence, in the long run, the individual might be best off choosing behaviour depending on the behaviour of others. The role of leadership becomes important as the leader will have the possibility to convince individuals to co-operate by inducing expectations about co-operation (Baland & Platteau, 1996). Leaders may also use and develop further the level of trust and trustworthiness in the community, which in itself supports co-operation (Durlauf and Fafchamps, 2004).