Most existing economic analyses of optimal groundwater management use single-cell aquifer models, which assume that an aquifer responds uniformly and instantly to groundwater pumping. This paper demonstrates how spatially explicit aquifer response equations from the water resources engineering literature may be embedded in a general economic framework. Calibration of our theoretical model to published economic studies of specific aquifers demonstrates that, by averaging basin drawdown across the entire resource, existing studies generally understate the magnitude of the groundwater pumping externality relative to spatially explicit models. For the aquifers studied, the drawdown predicted by single-cell models may be orders of magnitude less than that predicted by a spatially explicit model, even at large distances from a pumping well. Our results suggest that single-cell models may be appropriate for analyses of the welfare effects of groundwater management policies either in small aquifers or in larger aquifers where average well spacings are tens of miles or more. However, in extensive aquifers where well spacings are on the order of a few miles or less, such as many of those of concern to groundwater managers and policy makers, use of single-cell models may result in misleading policy implications due to understatement of the magnitude and spatial nature of the groundwater externality.