This paper examines the Olsonian thesis that group is size inversely related to successful collective action. We start with an empirical analysis based on primary data. This data gives information on a set of 21 villages in the Indian Himalayas that collectively monitor to protect and conserve community forests. This empirical analysis reveals that small and large villages fare relatively poorly, while medium size villages are much more successful, in the provision of monitoring. This finding goes against the general consensus that group size is inversely related to the likelihood of successful collective action. We identify two features of the collective good that appear critical. Both features are standard in the literature on public goods. The first feature is that the monitoring technology displays lumpiness, and must be above a certain minimum size to be worthwhile. The second feature is that the collective good is only imperfectly excludible and that this excludibility is decreasing in the size of the group. We formulate a theoretical model which incorporates these two features and develop a set of sufficient conditions on the monitoring technology under which the sustainable levels of collective good match the empirically observed patterns.
