For Clean Water, An Approach That Works
At a spring just outside Busia, Kenya—a town of 30,000 in the western part of the country, on the Ugandan border—a steady stream of women comes and goes all day. The women, most with children in tow, chat while they wait for their 20-liter yellow plastic jugs to fill, then set off toward home, jug on head.
One step in the process, so brief it almost goes without notice, distinguishes this water source from thousands of others around the country: just before filling her jug, each woman places it underneath a blue-and-white metal chlorine dispenser and, with a small twist of her wrist, releases the precise amount of chlorine needed to clean the water she’ll collect, and her family will eventually drink.
There are more than 40 such dispensers like these serving almost 10,000 people, mostly in western Kenya, in a pilot project implemented by Innovations for Poverty Action (IPA), a New Haven-based nonprofit with a presence in more than 30 countries. Michael Kremer, Gates professor of developing societies at Harvard, is helping to lead the project.
IPA aims to create and evaluate innovative—and sustainable—approaches to development. Says Karen Levy, IPA’s country director for Kenya, “One of the hallmarks of development interventions is—something works great at first, but go back a year later and no one’s using it.” IPA is seeking to change that. Once research proves a particular approach is effective, the organization disseminates information about it to policymakers, investors, donors, and development workers around the world. One previous Kremer project offers an example: in the last year, the Kenyan government has administered deworming medication to 3.6 million children at primary schools around the country, at a cost of 36 cents per child, with technical assistance from IPA and partner organization Deworm the World. Kremer's research had shown that deoworming was one of the most cost-effective ways to keep children healthy and in school.
The chlorine dispenser is another example, the fruit of a multiyear process of testing different strategies to improve child health, including different chlorine distribution schemes. Each year, nearly 2 million children die from diarrheal diseases worldwide; unsafe drinking water is a leading cause. In addition, chronic diarrhea in early childhood can contribute to malnutrition, stunting, and cognitive impairment.
IPA tried first to improve water quality by protecting springs (usually by encasing the spring in concrete so water flows out from a pipe rather than seeping from the ground, where it can be contaminted by runoff). The team found that the water coming from these protected springs was clean, but often became contaminated by the time it was consumed.
Other trials involved testing different types of containers, and distributing chlorine for people to add at home (along with advertising campaigns to encourage them to do so). In the end, placing chlorine dispensers at springs proved to be both the most cost-effective and the most effective overall approach: the dispensers cost an estimated 30 cents per person per year.
The dispensers have special metered valves that deliver just the right amount of chlorine for the size of bottle Kenyans typically use to transport water, taking the guesswork out of using the chemical. This accurate dosing reduces the likelihood that treated water will have a chemical aftertaste (eliminating yet another reason for not using the chlorine). Chlorine-treated water requires agitation and waiting time to become fully safe to drink; when the chlorine is added at the spring, purification is accomplished on the walk home. The dispenser provides a visual reminder, encouraging habit formation and the development of social norms around chlorine use. And at home, the water stays clean even if poured into a less-than-clean drinking glass or stored in an open container into which a child plunges a hand.
During unannounced home visits three to six months after dispenser installation, testing in communities with the dispensers found that between 60 and 70 percent of households had detectable levels of chlorine in their drinking water, compared to just 5 to 10 percent in communities that had to rely on store-bought chlorine for purification. The dispenser method is also less expensive: chlorine is so cheap that when it is sold or distributed in small containers, the packaging often costs more than the chemical itself. “A bottle of bleach you buy at CVS could treat water for one person for 40 years,” says Kremer.
Even as IPA continues evaluating the first 18 months of the dispensers' use, their reach is expanding: Kenya's Ministry of Education recently launched a school-based dispenser initiative that reaches more than 30,000 people, including children who drink clean water while at school, as well as their families. “If you’re looking for a public institution that goes really deep into rural areas, schools are among the furthest reaching,” says Amrita Ahuja, Ph.D. ’09, a postdoctoral fellow in the sustainability science program at the Harvard Kennedy School’s Center for International Development who is leading the effort to scale up the clean-water project.
Even while IPA conducts follow-up evaluation of some components of the dispenser program, such as pricing and the effectiveness of community promoters, the nonprofit is installing 100 more dispensers in Kenya and beginning pilots in Swaziland, Ethiopia, and Bangladesh.
By testing the approach in each new setting before scaling it up, the organization aims to aims to avoid the common development-work pitfall of throwing money at an approach before effectiveness has been proven. Says Levy: “Our agenda is about research, and about research that provides useful information. Let’s look very, very carefully at what works, and then spend money on that.”