they are important in groundwater irrigation. This refers not only to dynamic conservation of the resource, but intertemporal externalities of harvesters’ behavior, whereby my extraction this period affects your payoff next period.25
One can reasonably question whether surface-water irrigation systems are common-pool resources at all. Groundwater-based irrigation, of course, draws from a resource (an aquifer) that is subject to regeneration as well as the risk of depletion, as is the case with pastures, forests, and fish. To the extent that canal systems based on reservoirs or river diversions bear a formal resemblance to such common pools, it lies in the collectively maintained infrastructure: water source, canals, and water-control devices. Cleaning canals and repairing equipment is formally similar to replanting, recharge, and regeneration. Moreover, one person’s use of irrigation water reduces the availability for others, just as in other types of common-pool resource systems. Nevertheless, human intervention in natural systems is arguably much more invasive in irrigation than in, say, fishing or fuelwood collection. (These and other aspects of “irrigation exceptionalism” are considered in passing in Rose, this volume:Chapter 7.)
To a large extent, of course, the problems of successful commons management are not necessarily based on the characteristics of the natural resource itself—as the earlier, tragedy-of-the-commons tradition would have it—but rather the more prosaic problem of getting people to cooperate. Thus the problem is particularly closely related to those of producer and worker cooperatives. Mobilizing cooperative effort is especially problematic at the level of institutional supply, but also in the running of the institution.
Another social—rather than natural—phenomenon deserving increased attention is the effect of market failure. Market failure is said to exist when the market for a good or service fails to be efficient or, in the extreme, fails to exist at all. Such market failures in credit, insurance, and land are endemic in agrarian economies, and interact with the problem of cooperation. Optimal regulatory regimes are not difficult to describe in theory, but real-world market failures constrain the set of feasible arrangements. These constraints may be such that commons users are unable to negotiate any kind of cooperation whatsoever; or, they reach an accord that nevertheless leads to environmental degradation. Another such market failure that is frequently invoked but not quite justified is the impossibility of side payments—or equivalently, the absence of secondary markets for the common-pool resource.
Our empirical reference point limits the generalizability of our results to global-scale commons, where actors are states and international agencies rather than peasant households. Our conclusions regarding the effect of heterogeneity could be implausible in the setting of international climate-change agreements, for example; there, the presence of disproportionately powerful actors might enhance the prospects for cooperation.26
This survey illustrates the utility of the large-scale multivariate analysis of resource-using communities. Similar syntheses could be compiled for other types