We use the Becker-DeGroot-Marschak (BDM) mechanism to estimate willingness to pay (WTP) for and heterogeneous impacts of clean water technology through a field experiment in Ghana. Although WTP is low relative to cost, demand is inelastic at low prices. Short-run treatment effects are positive throughout the WTP distribution. After 1 year, use and benefits are both increasing in WTP, with negative effects on low-WTP households. Combining estimated treatment effects with households’ WTP implies valuations of health benefits much smaller than typically used by policy makers. We explore differences between BDM and take-it-or-leave-it valuations and make recommendations for implementing BDM in the field.