Changing the way we change the world

Predictors of non-functionality for community-managed handpumps: a simple summary of Foster’s analysis

By Susan Davis, Executive Director

This is a summary of a useful study by Tim Foster.  It was actually titled “Predictors of Sustainability for Community-Managed Handpumps in Sub-Saharan Africa: Evidence from Liberia, Sierra Leone, and Uganda,” but as usual I find the reasons for failure useful to share.

Failed water point near Debre Zeit, Ethiopia. We were told the borehole collapsed (credit Susan Davis, July 2012)

Failed water point near Debre Zeit, Ethiopia. We were told the borehole collapsed (credit Susan Davis, July 2012)

The study was published in 2013, but I’ve found that some people who are interested in this kind of rigorous analysis have not heard of it.  So I’ve summarized the findings below, in plain English. If you prefer the statistical language, please see the original paper.

Where did the data come from?

  • Comprehensive water supply inventories were created by the governments of Sierra Leone, Liberia, and Uganda, with the support of development partners.
  • All three inventories were nation-wide and have technical, institutional, financial, and geographical information associated with rural water supply systems.
  • Liberian data were collected in 2011 for 10,001 improved water points, 9,388 of which were equipped with handpumps.
  • In Sierra Leone, data on 28,845 water supply systems were collected in 2012. Of those points, 12,003 were handpumps.
  • Ugandan data were collected in 2009-2010. Within the data set are 42,151 water points fitted with handpumps.
  • Foster analyzed data for handpumps that (a) had a water committee established and (b) had been installed less than nine years prior to the time of data collection. The latter was to avoid looking at handpumps installed during the civil wars in Sierra Leone and Liberia.

How was functionality defined?

  • Handpumps considered “nonfunctional” included those broken down due to technical problems relating to the handpump and other reasons such as changes in water level. Handpumps described as partially working, or in working order but not in use, were considered to be functional.

What did the analysis show?

  • In Liberia, 18.2% of the subsample of water points were non-functional.
  • In Sierra Leone, 17.9% of the subsample of water points were non-functional.
  • In Uganda, 19.1% of the subsample of water points were non-functional.
  • In all three countries, non-functionality rates rose with the age of the water point.
  • Across all countries, rates of breakdown were considerably higher where water users were not paying fees.
  • In Uganda, handpumps with regular servicing were more likely to be functional.
  • Handpumps in Sierra Leone were more likely to be working the closer to the community that spare parts were stored or sold, and where a community had access to a trained handpump mechanic.
  • In all three countries, failure rates were greater the further that water points were located from a district or county capital.

What wasn’t surprising?

The following factors match what has been commonly found in previous research on water point functionality:

  • Collection of user fees
  • Women in key water committee positions (Uganda)
  • The closeness of the community to spare parts (Sierra Leone)
  • Positive user perception of water quality (Liberia & Sierra Leone)

What was surprising?

The following factors (and a few others) were not found to be associated with handpump functionality.

  • Training of water committees (Uganda). This may indicate that training of water committees is poorly designed and delivered, or that even if done well it
    provides little benefit as a once-off exercise without periodic follow-up.
  • The number of handpumps within a district (density). The effect of handpump density on spare part supply chain viability may be more important on national or subregional scales than smaller geographic units.
  • Advance payment of water fees. This is likely because funds collected in advance and stored over time may be insufficient to pay for repairs when the need arises.

What should be done with this information?

  • Improve the way maintenance activities for handpumps are supported over the long-term.
  • Find ways for these elements to be sustained indefinitely: collection and storage of funds from users, active committees with genuine female representation, networks of skilled pump mechanics, and the viability of spare part supply chains
  • Assess alternative water point management models besides the community management model, such as private sector leasing.

Again, please note this is a simplified summary of Foster’s paper. Please read the full paper for more nuance.

2 Responses

Leave a Reply