First (Out)come, First Served? “Dispensers for Safe Water” Tests a COD Hypothesis

August 04, 2016

I’ve been working on the idea of Cash on Delivery (COD) for some years under the hypothesis that if we could define good outcome indicators, someone would step forward to buy them. Two core features of COD are that funders pay for outcomes and give recipients flexibility in how they achieve those results. Programs around the world approach this model in different ways, but the indicators they use tend to be negotiated between funders and recipients—which doesn’t really test my hypothesis. However, I recently came across a result looking for a buyer—a safe water initiative which is already set up to receive payment after achieving a desired outcome: “households utilizing safe water.” If in the next few years, a philanthropy or aid agency steps up to buy this approach, I’ll have some (anecdotal) evidence to support my hypothesis. If not, it will be time for a rethink.

Outcome indicators are a critical feature of COD. Understandably, no funder is going to sign up for a contract to deliver results if they don’t trust the measurement to reflect true progress. So an indicator has to be objectively measured and precise enough to satisfy both funders and recipients.

Another feature of COD that funders find difficult to swallow is the idea that payments are made after outcomes are achieved. Funders typically ask how the organization or country can produce results without upfront financing, or question whether their funding was necessary in the first place. They find it difficult to envision their funding as a reimbursement for actual achievements rather than an upfront investment to carry out a plan that may or may not bear fruit.

So what would happen if an organization came forward with a plan to supply a verified outcome in return for a set unit payment after delivery?

In a sense, this is what Dispensers for Safe Water is currently doing. According to Andy Narracott, a deputy director at Evidence Action, Dispensers for Safe Water has demonstrated that it can install and operate chlorine dispensers in rural African contexts. Numerous studies have found that similar interventions to provide safe water reduce the prevalence of diarrhea and other water-borne diseases. Each month, Dispensers for Safe Water verifies how many households properly chlorinate the water collected from standpipes or rivers by testing samples in randomly selected households.

Narracott told me that, initially, about 25 percent of households will chlorinate their water. The program applies a number of strategies to encourage greater use of the dispensers and finds that it can increase the share of households in a given catchment area with safe water to as high as 60 percent. As of today, the program’s 27 thousand dispensers have brought safe water to more than 4.6 million people in Uganda, Kenya, and Malawi with no associated user fees (according to their public dashboard).

And here’s the kicker: Dispensers for Safe Water’s financing model recovers about 40 percent of its costs two to three years after dispensers are installed by selling carbon credits. (Compared to boiling water with firewood, decontamination with chlorine avoids greenhouse gas emissions … a fascinating part of the story if you want further reading.) The ability to sell carbon credits suggests that the actual verification process must be pretty robust.

So here is the test: Will a funder step forward to conduct due diligence on the quality and validity of the outcome measure, establish a baseline, and agree to pay for results? Dispensers for Safe Water estimates that it could increase the number of households using clean water in other parts of rural Africa at a cost of about US $6 per household per year. The mechanism it offers has a well-defined independently verifiable indicator that measures how many households are using chlorinated water. The mechanism also lets funders pay only after the outcome is delivered. This should ring all the right bells at agencies like DFID, MCC, Sida, Norad, and the World Bank that regularly talk about paying for results, cost-effectiveness, and value for money. It should also be attractive to philanthropies that care about public health and expanding a promising initiative. And if it were structured as a matching grant for funds provided by a developing country government, it could even encourage sustainability beyond the involvement of external funders and implementers.

So the experiment has started. Will this well-measured worthy outcome find a buyer?


CGD blog posts reflect the views of the authors, drawing on prior research and experience in their areas of expertise. CGD is a nonpartisan, independent organization and does not take institutional positions.