Skip to main content


8 arguments about self supply: to have, or avoid

Published on: 17/02/2012

A summary of some recent discussions within IRC about self supply as an approach to rural water supply, based around 8 arguments.

At the IRC International Water and Sanitation Centre, self supply is increasingly being recognised as a useful service delivery model alongside other water supply models. We’ve been arguing the merits and drawbacks of self supply a lot over recent months, including at a recent internal seminar on ‘household-led investments in WASH’ and on the margins of the Rural Water Supply Network forum in Kampala at the end of 2011. Some of our flagship projects are developing knowledge in this area: the water services that last initiative has picked up self supply at an international, if not yet national, level; WASHCOST is reporting fascinating data on the level of investments by households including on purchases of soap; and WASHTECH is critically examining technology introduction in Africa including several technologies associated with self supply. Our Ethiopian country programme is supporting the Ministry of Water and Energy and other partners to develop a Self Supply Acceleration Programme and with MoWE we are currently co-leading a Self Supply Working Group in Ethiopia to help make this happen. And we are an active partner in the internationally-focused Self Supply Working Group of the Rural Water Supply Network. Why this interest, and what have we been arguing about?

1.    Community water supply is better than self supply (or vice versa). This is an argument I have been trying to avoid, but we have often had anyway. You can obviously point to the poorer water quality of traditional family wells and argue that self supply is a health hazard compared to protected community water points (you could conclude this from the results of our research in Ethiopia and other similar surveys, see Sutton et al., 2011, available below). But the latter don’t always score very well either in water quality surveys, and as reflected by the high reported levels of non-functionality of handpumps they are often not working anyway. And is it fair to compare the performance of community rural water supply which has (at least in theory) a support structure backed by government to ensure standards are kept, with self supply, which has often been ignored and not subject to much effort to encourage improvements? We are recognising now that self supply is complementary to other service delivery models like community water supply in rural areas (and other options in urban areas, more about that later). What country has reached universal coverage without self supply? It is common in the USA, Ireland, Poland to name just a few examples, to this day. A big problem we have as a sector is that we often think households use a single water source and when we collect data, we often (e.g. in the household surveys used by the JMP) only think about ’drinking water’. Community water supply is generally a good way of providing a fairly small amount of water of generally decent quality but quite far away from the home. Self supply often does something different. Many sources are used for drinking, but the bulk water provided for small-scale productive uses (irrigation, livestock, crop processing), washing and cleaning is often more important.

2.    All subsidies are to be avoided: the consensus is generally that household-led investments like family wells, rainwater harvesting and construction of latrines do not need subsidies but rather good low cost technologies and products that are made available through well-developed supply chains. But isn’t that also partly because where subsidies have been deployed it has often been in a poorly conceived manner? Rather than debating the merits of subsidies versus no subsidies, perhaps it is better to keep subsidies as an option and to consider more whether a subsidy is needed to help grow (rather than distort) a market, and whether it is scaleable and sustainable. At the RWSN forum, experiments with subsidies for rainwater storage bags in Uganda were discussed, where vouchers were used to avoid breaking the links between the suppliers and consumers. This has happened in Ethiopia when government has purchased rope pumps in bulk and distributed them. Then the lucky owners don’t know who they can go to for assistance when it breaks down.

3.    Separating water, sanitation and hygiene makes sense. Because of the different service delivery models, sanitation with its household focus, hygiene with its personal individual nature and community water supply may well be best to go their different ways, with coordination where possible to make sure they all happen. But when we are talking about a self supply approach to water, it makes much more sense to closely learn, link to or integrate with sanitation and hygiene and their own household-driven investment and behaviour-focused approaches. At the implementation level, there is room for integration as well as linking and learning from closely related approaches like self supply, Community-Led Total Sanitation (CLTS) and introducing Household Water Treatment and Safe Storage (HWTS). It makes much more sense than trying to integrate community rural water supply and household-level sanitation and hygiene work which we have often tried but struggled to do. Of course there are differences. It perhaps makes much more sense in our African focus countries where self supply in water is focused on encouraging new investments, just like in latrines, and encouraging behaviour changes to fetch and store water safely. Perhaps in India, self supply is much more about regulating groundwater resources abstraction and that doesn’t link so well, unless the need is to regulate and control e.g. where on-site sanitation has reached high levels of penetration too.

4.    Piped water supply has economies of scale and self supply is an expensive solution: this should be the case in urban areas where people are close together and few people have suitable land for constructing their own well, but even in towns and cities a study from IIED reports self supply to be thriving. Self supply is common in urban and peri-urban areas where alternative services are not provided or cannot be relied upon. In rural areas, family wells are likely to be a low cost option even on a per capita basis when compared with community waterpoints as they are often also shared by their owners with neighbours. But that’s not really the point anyway. Because self supply is paid for by households and community water supply is paid for by governments or donors, the economies of scale argument is not really appropriate.

5.    Self supply is an abrogation of responsibility by governments. Instead of serving their citizens with proper services, self supply is a cop out, and governments will not be upholding their commitments to realise the human rights of their citizens. This assumes that community water supply (or other options) is something that the government does or makes possible, which is not the case with self supply. That is often true. But leaving households to do self supply without any sensible support, market development or encouragement (even subsidies if they are smart) is not the idea. Households can be encouraged to do self supply faster, better, safer. And in both countries with low levels of coverage (as a stop-gap measure that will later be replaced by other supplies) and high-levels (in the scattered last 10% or 20% that are too costly to serve through communal systems) there seems to be a place for self supply alongside other service delivery models. Self supply can be part of a proper service.

6.    Self supply is not sustainable. Just look at the groundwater overabstraction problems in India. Yes, it can be unsustainable. But it is important not to look at all the world as being in same situation, or to fail to differentiate within countries. Even in Ethiopia groundwater resources are being over-abstracted in some local areas where intensive agriculture is supported by mechanised pumping, but generally resources are underdeveloped. And some countries have regulated and controlled groundwater abstraction. Regulation is needed at some point e.g. when pumping becomes mechanised on a large-scale. And as well as managing water resources, a regulatory regime is needed to minimise water quality risks and to make self supply as safe as possible. Restaurants should be more tightly regulated (like community water supplies) compared to household kitchens (like family wells or other self supply sources) as Sally Sutton has pointed out.

7.    Self supply is a cheap inferior solution for the rural poor, or is inequitable because only rural elites can afford it. If self supply is a cheap inferior solution then people won’t invest in it. End of story. The fact that many thousands of poor rural households do invest in wells, lifting devices and other options suggests it is the best choice available (it also says something about access to or satisfaction with the alternatives) or one they are willing to pay to supplement their water access. Providing services and products that people can afford and find attractive is obviously a pre-requisite for self supply. Self supply can surely be inequitable. In rural areas, better off farmers that drill large numbers of borewells are impacting on the village water supplies relied upon by people with no land (and the old elites that have land in the command area of tanks storing surface water). Recent research on family wells in Ethiopia showed that well ownerships was not correlated strongly with wealth but that initiative seemed rather more important. It is often also the culture to share, and family wells may provide access to neighbours as well as owners. Surely equity could be improved too through better access to low-cost options and more micro-credit (in the areas we researched in Ethiopia micro-finance banks did generally not lend for family wells as they did not see these as productive assets). A combination of community water supply and self supply builds in some more equity, and will have other benefits too including more climate resilience in some areas.

8.    Piped water supplies are the future in Africa. The numbers point to lots more household connections and some think that the future in Africa will all be about piped water supplies. That is certainly where a lot of the interest (and money) is going. But the fact that self supply is thriving in urban and peri-urban areas suggests that this may not happen quickly and certainly not in entirety. In fact, self supply and piped supplies are similar in providing bulk water, often of lower quality than possible, on the doorstep. How will the future look? More piped water supplies and household connections (increasing rapidly), less communal water points (fading slowly?) and more self supply (increasing steadily?).

So self supply is a very good idea and part of the future, not the past? I wouldn’t yet say that is the viewpoint of most IRC colleagues, but it is certainly generating more interest, and we will surely have many more arguments about it. We’ll keep reporting on how efforts to accelerate self supply develop in Ethiopia especially.


At IRC we have strong opinions and we value honest and frank discussion, so you won't be surprised to hear that not all the opinions on this site represent our official policy.