Econometric evaluation of the impact of agricultural conservation on nonpoint source pollution: An application to the Wabash River watershed

Agricultural production is the largest contributor of nitrogen and phosphorus pollution in lakes, rivers, and streams in the United States. The effectiveness of agricultural conservation programs that encourage farmers to adopt certain practices to reduce this water pollution, once implemented, is an open question. We develop a unique data set combining the spatial structure of the watershed river system, the timing of federal conservation contracts, water quality measurements, land use, land cover, and weather data to study the effect of conservation contracts on nitrogen and phosphorus levels in the Wabash River watershed, which drains Indiana and Illinois. We develop econometric models that generate a causal understanding of the effectiveness of these conservation contracts for reducing nitrogen and phosphorus levels in the surface water system. We find that at current treatment levels, these programs reduce surface water pollution only during relatively dry periods. The efficacy of these programs in the study area is highly sensitive to precipitation to the extent that average precipitation can eliminate the nutrient loss reduction benefits of conservation program installations at current treatment levels. Therefore, we find weak evidence to support ambient downstream water quality improvements resulting from program investment levels to date. We anticipate this work will motivate further inquiry into the manner in which these conservation programs have or have not been effective.