Files
Abstract
This study examines barriers to conservation drainage adoption despite cost-share incentives in the U.S. Midwest. Using survey data from 530 farm operators across the Corn Belt, we employ probit model, boosted regression trees, and latent class analysis to analyze factors influencing farmer non-acceptance of 50% cost-share programs for controlled drainage, saturated buffers, and wetland restoration. Results reveal substantial non-acceptance rates of 54.5-68.1% across practices. Probit models show that neighbor adoption reduces non-acceptance probability by 12.3 percentage points for controlled drainage, while higher liability-to-asset ratios consistently increase responsiveness to financial incentives across all practices. Machine learning analysis reveals that climate variables and geographic location are heavily influential in adoption decisions. Latent class analysis identifies four distinct farmer types, suggesting that uniform conservation programs are inadequate for addressing farmer heterogeneity. Understanding farmer heterogeneity and designing targeted interventions will be key to achieving widespread adoption of sustainable drainage practices.