studies of community-based irrigation (like most studies of common-pool resource systems generally) focused on one or two systems. We have learned a great deal from these case studies, but they do not permit generalizations about the relationship between heterogeneity and commons management. If one village with a high level of inequality also has a relatively successful management regime for its irrigation system, it is difficult to deduce from that a general relationship between heterogeneity and management. In statistical terms, the case studies do not have the degrees of freedom necessary to discern relationships among the institutions of governance, various dimensions of performance, and the structural characteristics of resource-using communities. More recently, a small number of studies have sought to complement the case study approach with information culled from relatively large numbers of resource-using systems. This chapter is unique in this volume in that it gives pride of place to this large-scale multivariate research, rather than other forms of empirical inquiry (e.g., laboratory experiments or anthropological case studies). More specifically it synthesizes lessons learned from a subset of these studies focused on irrigation systems.
The empirical context for our chapter, thus, is the poor hydraulic economy: peasant water users in conditions of low-income rural sectors. The unit of analysis is the resource-using group, of which heterogeneity is a characteristic.2
Irrigators, or users of some other common-pool resource, may be heterogeneous in economic, social, cultural, or other dimensions. There are many relevant types of economic inequality alone. Variants of economic heterogeneity include: inequality in wealth or income among the members of a resource-using group; inequalities in the sacrifices community members make in cooperating with commons-management regimes; inequalities in the benefits they derive from such regimes; and inequalities in outside earnings opportunities (“exit options”). There are other kinds of disparities that may have economic consequences, and those in turn affect cooperation. For example, locational differences, to the extent that they are not already reflected in landholding or wealth differences, might not be taken into account adequately if one considers only wealth inequality. Head-end and tail-end farmers in irrigation systems face different incentives to cooperate (Bardhan, 1984; Ostrom, 1994), as do fishers with access to more or less productive fishing spots (Berkes, 1986). For irrigators, long-run locational advantages and disadvantages will be capitalized into land values if land markets work reasonably well. Thus, the head-end/tail-end inequality is another version of wealth inequality.3 Of course, in many parts of the world, land markets notoriously do not work reasonably well. Even if head-end/tail-end differences are captured perfectly in land values, such locational differences provide strategic opportunities that are not normally available simply as a result of wealth differences. Head-end farmers, poor or not, get the water first. Similarly, differences in ability or effi-