Research

Billions of dollars have been dedicated to mitigating non-point source pollutionfrom US agriculture. Wetlands provide an ex-post natural solution by filtering sed-iments and excess nutrients from the landscape. Quantifying the impact of wetlandeasements in the Wetland Reserve Program (WRP) and Agricultural ConservationEasement Program (ACEP) on water quality is critical for optimal fund allocationfor non-point source pollution abatement. We causally identify whether newly re-stored wetlands reduce nitrogen and phosphorus loads at the subwatershed levelin the Mississippi River Basin. Results show that wetland easements reduce am-monia and total Kjeldahl nitrogen concentrations in the local subwatershed as wellas downstream. A short-term increase in local phosphorus immediately followingrestoration likely reflects soil disturbance. While wetlands have no significant long-term impact on local phosphorus levels, downstream reductions are observed. Weestimate that this federal investment in natural infrastructure leads to lower treat-ment costs for local water systems.

 Nonpoint source pollution from agriculture is the leading cause of nutrient pollution in the US. This paper addresses whether localized, farmer-led programs can cost-effectively reduce nonpoint source pollution by increasing the adoption of agricultural conservation practices. We study this in the context of an innovative program in Wisconsin that incentivizes farmers to take collective leadership of improving water quality in their local watersheds. Using a shift-share instrumental variables design, we find that a 10 percentage point increase in farmer participation in these programs leads to a 0.03 mg/L reduction (14\%) in ambient phosphorus concentrations in local streams and rivers. We also show that farmer participation in local watershed programs causes an increase in the adoption of cover crops, conservation tillage, and more diverse crop rotations. Importantly, this localized approach achieves water quality and conservation improvements at a substantially lower cost than existing federal subsidy programs, demonstrating the potential for bottom-up approaches to address nonpoint source pollution in other contexts. 

Wetlands provide a wealth of ecosystem services, including flood mitigation, wildlife habitat, and a natural ex-post solution to surface water pollution from agricultural runoff (Taylor & Druckenmiller 2022, Karwowski & Skidmore 2026) . Wetland restoration in the United States is supported by programs including the Agricultural Conservation Easement Program. Additionally, wetlands that are part of the Waters of the United States (WOTUS) are protected under the Clean Water Act. Currently, only wetlands directly abutting another surface water feature in WOTUS are protected (Sackett v. EPA 2023). Yet, there is limited causal evidence on the ecological impact of wetlands based on their surface water connection (Cohen et al. 2016). We test the impact of restored wetlands on downstream water quality based on the distance from the wetland to the nearest protected surface water feature. We link a rich dataset on all wetlands restored under USDA easement contracts in the Mississippi River Basin, National Hydrology streams and rivers, and harmonized measures of surface water quality from nearly three decades. We use an upstream-downstream model to evaluate the changes in water quality before and after wetland restoration. We find that non-abutting wetlands within one kilometer of a surface water feature significantly reduce phosphorus and nitrogen concentrations downstream. In contrast, abutting wetlands provide only modest downstream water quality benefits, though water quality is only one of the many ecosystem services that wetlands provide. The effects are most pronounced in the Upper Mississippi region, which is a large contributor to the Gulf Hypoxic Zone (Evenson et al. 2021), and the recipient of significant federal, state, and local resources to reduce nutrient loss (Iowa DNR 2025, Berhane et al. 2020). Our results demonstrate that restored wetlands are an effective tool to improve water quality, and considering their heterogeneous benefits could be pivotal in creating optimal policy. 

As renewable energy deployment accelerates, wind turbines are increasingly installed on agricultural land, yet their effects on crop productivity remain poorly understood. This paper examines how wind turbine installations affect agricultural yields by combining parcel-level corn yield data for more than 2.6 million farmland parcels with detailed data on over 70,000 turbines and local weather conditions in the U.S. Corn Belt. Using a distance-based approach, we show that the effects of wind turbines on corn yields dissipate beyond 10 kilometers, indicating that turbine impacts are highly localized. We then exploit variation in both spatial proximity to turbines and the timing of turbine installation in a difference-in-differences framework that compares parcels located within 0–10 kilometers of a turbine to nearby parcels 10–20 kilometers away. We find that parcels near wind turbines experience an increase in corn yields of approximately 0.4–0.8 percent following installation. The effects are larger near higher-capacity turbines, in denser turbine clusters, and for parcels located downwind of installations. Mechanism analysis suggests that turbines reduce the incidence of extreme heat and precipitation events, with little effect on average weather conditions. Back-of-the-envelope calculations suggest economically meaningful gains, each megawatt-hour of wind generation (valued at $30) creates $1.46 additional agricultural revenue—approximately 4.85 percent of electricity value—highlighting an overlooked agricultural productivity spillover of wind energy infrastructure.

Agricultural runoff is the primary source of water quality impairment in the United States. The Conservation Reserve Program (CRP), which pays landowners to remove their environmentally sensitive land from production and to implement conservation covers and practices, is to a large degree aimed at reducing erosion and nutrient runoff. We study how land share in CRP impacts nitrogen, phosphorus and ammonia concentrations in the Mississippi River Basin from 2000-2018. We leverage contract-level CRP data and recently harmonized water quality measurements to evaluate CRP effectiveness at the subwatershed-level. We use an instrumental variables approach that exploits variation in the national acreage cap to mitigate the endogeneity bias. We estimate that increasing landshare in CRP significantly reduces nutrient levels. These findings contribute to a broader understanding of CRP's role in improving water quality, with implications for conservation programs and agricultural policies. By emphasizing voluntary programs like CRP, policymakers can more effectively combat agricultural non-point source water pollution, reducing its environmental and economic impacts.