PROMOTING GROUNDWATER REFORM IN THE GUADIANA BASIN

In many parts of the world, there is a substan-tial disconnect between existing water insti-tutions and the institutions needed to ensure sustainable water supplies for the future. Im-plementing large-scale institutional change is politically challenging. In many cases, changes must emerge from political bargaining or ne-gotiation. Accordingly, there is substantial lit-erature looking at applications of negotiation theory to water policy. Carraro, Marchiori, and Sgobbi (2007) review this literature in detail. A common feature of water problems is that solutions must consider interactions between hydrology, economics, and political institu-tions. In most cases, the set of critical relation-ships is too large to be captured in a model that can be solved analytically. In these settings, numerical analysis provides a powerful tool for analyzing policy choices. Numerical mod-els can incorporate a variety of relationships and constraints that are critical to understand-ing the real problem. Moreover, numerical models do not rely on strong assumptions about interior solutions to generate results. This article illustrates how combining numerical analysis with another technique, response surface analysis, produces a tractable model of policy interactions grounded in a rich description of the real-world policy problem. Response surface analysis is frequently used in operations research; it uses a sample of full nu-merical solutions to construct an approximate mapping from policy choices to outcomes of interest. The resulting mappings or response surfaces share the advantages of analytica