ciency in resource extraction will affect cooperative behavior (Johnson and Libecap, 1982). These differences in many cases will be correlated closely with wealth: Fishers with more gear will have lower unit costs of harvesting. Differences in what economists call rate of time preference—essentially differences among resource users in the degree to which they consider the future in their current extraction activities (Ostrom, 1990:passim)—will lead to differential impatience among commons users in making short-run sacrifices for resource conservation.
Ethnic heterogeneity such as differences in language or caste among irrigators also will affect cooperative behavior.4 An irrigating community may be socially heterogeneous if its users come from various villages. Of course, in many cases, ethnic or social heterogeneity will be correlated with economic heterogeneity, as certain castes or ethnic groups are also more likely to be richer or poorer than other groups. Nevertheless, these noneconomic types of heterogeneity potentially have effects independent of the economic heterogeneity with which they are correlated.
Other types of inequality or heterogeneity are measured by state variables like trust or social cohesion—the absence of which Baland and Platteau (1995) called cultural heterogeneity. Generally, shared values or interpretations of social problems—cultural homogeneity—can facilitate cooperation in the use of the commons. It is conceivable that cultural homogeneity and pronounced economic heterogeneity coexist in a stable relationship. For example, highly unequal agrarian societies might sometimes exhibit widespread adherence to a hierarchical ideology that facilitates monitoring and enforcement of cooperative agreements.5
Cultural heterogeneity exists, then, when there is more than one community of interpretation or community of shared values among the members of a group. This can overlap with ethnic or social or locational heterogeneity, but need not. The experimental social-psychology literature reviewed by Kopelman et al. (this volume:Chapter 4) demonstrates the critical importance of the relative shares of “prosocial” and “proself” individuals in a group. This division could exist in the absence of other types of differentiation.6 A related type of difference arises in the game-theoretic context of Falk et al. (this volume:Chapter 5) that allows for the possibility that some players have a preference for reciprocity or equity.
The sources of heterogeneity considered in this chapter do not pretend to exhaust the possibilities for differentiation. One type of heterogeneity not considered here is different uses of the resource. In the western United States, this conflict pits residential water users against agricultural ones. In agrarian economies, an ever-present, and arguably the most important, dimension of heterogeneity is gender. For water users, Meinzen-Dick and Jackson (1996) argue that differentiation by gender and by type of use overlap substantially: In many cases, women need water for cleaning and cooking, while men need it to irrigate cash crops.7
Why is heterogeneity important? Generally speaking, if we can discern empirical regularities that link inequality to better or worse commons outcomes,