Willingness to pay for clean water in less developed countries: systematic review of experimental evidence

Background: Diarrheal diseases kill two million children every year despite the availability of effective and inexpensive technologies to improve water quality and limit the spread of pathogens. There is a growing literature on the effectiveness of such technologies but important gaps remain in understanding the demand for these products and the adoption decision. Methods: This review expands upon and complements several existing summary articles by focusing on willingness to pay for cleaner water. Willingness to pay can be measured by price randomizations that induce people to reveal their valuation in real purchase decisions or by other methods such as contingent valuation exercises in hypothetical situations and discrete choice analysis. The review conducts a systematic search for experimental evidence on willingness to pay for cleaner water. Results: This review finds few studies that have used randomized approaches or even attempted to measure households’ willingness to pay for cleaner water, but a very clear picture emerges from the existing evidence: willingness to pay is often less than the cost of these technologies and demand is very sensitive to price. Existing evidence suggests that positive prices do not effectively target products to those who need them the most and that positive prices are a key barrier to realizing potential gains associated with water treatment. Implications: Given the evidence of low valuation for water quality, despite the impact of water-borne disease on child health, the challenge for research and policy is to identify innovative service delivery models and technological innovations that drive prices down and make public subsidies more feasible. Future willingness to pay studies should be based on real purchases and use. Experimental methods to collect estimates of willingness to pay are easily justified as promotional discounts and could be implemented via coupon programs that make it possible to assemble large datasets quickly and cheaply.