Chlorine Dispensers for Safe Water – Poverty Action Lab, 2012.
Research by J-PAL affiliates Michael Kremer, Edward Miguel, and Sendhil Mullainathan, along with Clair Null, Jessica Leino, and Alix Zwane, has shown that a point-of-collection water chlorination system, in combination with encouragement from community promoters, can dramatically increase access to safe water compared to marketing bottled chlorine through retail outlets. Evidence from their studies has contributed to the scale-up of the Chlorine Dispenser System (CDS), reaching over 400,000 people in Kenya and 20,000 people in Haiti, with plans to expand the program to at least two additional countries by 2014.
From 2003 to 2010, J-PAL affiliates Michael Kremer, Edward Miguel, and Sendhil Mullainathan, along with Clair Null, Jessica Leino and Alix Zwane, conducted a suite of randomized evaluations to investigate ways to improve water quality in rural Kenya. In their first study, researchers and partner NGOs sought to find a cost-effective way to reduce diarrheal disease by improving water quality at the source. Under the “Spring Cleaning” program, community springs were encased in concrete, forcing water to flow through a pipe rather than seeping from the ground, thus preventing contamination from groundwater. This simple infrastructure investment reduced fecal contamination at the source by 66 percent. However, much of this improvement was lost in users’ home water supply due to recontamination during transport and storage. For more about this project, see the related evaluation page.
Building upon these results, the researchers conducted a second set of randomized evaluations to investigate the role of price in determining households’ use of chlorination products. In the first phase, households were provided with a seven-month supply of WaterGuard, a point-of-use chlorination product, and a storage container designed to prevent contamination. This group was subdivided into treatment arms receiving (1) coupons for a 50 percent discount on bottles of WaterGuard, or (2) additional verbal persuasion messages, or (3) no coupons or messages. The free supply increased chlorination by 52 percentage points, a large effect relative to the 2 percent starting level. However, only 10 percent of the coupons for discounted WaterGuard were redeemed, and additional persuasive messages about the product did nothing to boost demand, suggesting that both price and convenience play an important role in households’ decisions to take up water treatment products.
In the second phase, researchers compared multiple treatments designed to increase chlorine take-up, including persuasion messaging, promotion from members of the local community, and access to chlorine dispensers combined with local promotion. Paid community-based promoters who provided reminders through home visits initially increased chlorine take-up by about 30 percentage points, but this effect was not sustained after payments to promoters ended, making this an expensive long-run strategy. To combine the advantages of reduced cost, convenience, and social norms for promoting chlorine use, researchers designed and tested a point-of-collection chlorine dispenser system that provided a free supply of chlorine at local water sources. Chlorine dispensers, in combination with the paid promoters, increased take-up by 53 percentage points, and take-up was sustained 30 months into the program, even after payments to promoters had ended. For more about this project, see the related evaluation page.