Debates on institutions for collective action in general

What is the effect of group size on the level of cooperation?



The original insights by Olson (1965) and Hardin (1982) were presented in a categorical way: ‘large groups will fail; small groups may succeed’ (Hardin 1982, 38). According to them, since the increase in the number of users inevitably led to decreasing individual gains relative to the cost of providing a good, free-riding would be much more pervasive in larger groups and decreasing individual contributions, the ultimate fate. This original aim of universality has been, however, severely contested by a diversity of scholars working in the social sciences. As a result, the negative relation between size and collective action seems now much weaker than some decades ago.


Chamberlin (1974) was one of the first to point at the flaws in Olson’s argument. If it is assumed that the collective good is private, that is, if the gains it provides the users are divisible and excludable, Olson is right. However, in the case of pure public goods, an increase in the number of users does not lead to lower individual gains. The incentives to provide the good remain strong and, thus, a positive relation between the size of the group and the levels of collective action emerges. Oliver and Marwell (1988) refined Chamberlin’s analysis. According to them, the negative relation posited by Olson may hold even in the case of public goods but only when the cost of providing the collective good increases as a result of a higher number or users (scale diseconomy or zero jointness of supply). When costs are independent from the number of users (pure jointness of supply) or even when they increase less than proportionately (partial jointness), the probability of someone providing the collective good as well as the total amount of contributions increases as well.


Recently, Esteban and Ray (2001) and Ostrom and Poteete (2004) have provided us with new insights which keep calling into question Olson´s findings. Esteban and Ray (2001) have demonstrated that, even when the gains of the good are excludable (private good), increasing marginal costs of individual efforts (whether time or money) make possible a positive relation between group size and levels of collective action. Ostrom and Poteete (2004), for their part, highlight the significance of the context as well as the role of the institutions in tempering the harmful effects of a large size (or a heterogeneous composition).



> Back to overview of debates