Published on: 25/10/2011
Determining life-cycle costs or aggregate costs to ensure the delivery of adequate, equitable and sustainable services to a population in a specified area is not easy. Numerous factors influence costs and the costs of a water supply system can vary considerably from place to place, and over time. This is further complicated by the fact that the life-cycle costs of an adequate, equitable and sustainable service are not completely dependent on the hardware component of a supply system. Software costs related to design, capacity building, institutional development, establishment of micro-credit systems and so on are equally important, but often rendered invisible or not relevant enough. With little attention paid to software costs, cost data reviewed -- even in this study -- is severely limited to hardware and engineering costs. Notwithstanding these caveats, it is possible to offer generalisations regarding the costs of services delivery. These are (not in order of importance):
- CapEx of RWH systems is relatively high when compared to systems that do not require storage tanks, but are relatively low when compared with, for example, groundwater-based piped water supply. However the most expensive supply systems in terms of CapEx, are not necessarily the most expensive when consideration is given to: 1) the number of users and uses of the system; 2) unit CapEx per capita, per m3 of storage and/or per m3 of water supplied; and, 3) annualised CapEx that takes into account the expected lifespan of the system.
- Typically, CapEx per m3 of storage for RWH systems using jars and tanks falls within the range of US$ 40-200 PPP2008, whilst CapEx per m3 of storage for sand dams is more likely to be in the range US$ 10-30 PPP2008.
- OpEx of RWH systems is relatively low when compared to boreholes and piped schemes. OpEx is also low when compared to CapEx. However when annualised CapEx is considered, OpEx is typically within the range 0-20% of annualised CapEx.
- CapEx of RWH systems in Africa is approximately double than that in Asia and Latin America.
- When the systems are designed and used according to a design specification aimed at meeting a household’s primary needs, a typical borehole and hand pump system has lower CapEx per capita (around US$ 30-50 per capita PPP2008) than a typical RWH system (around US$ 50-100 per capita PPP2008). This point however is most likely to be incorrect, if there are a limited number of users of a borehole and/or a hand pump system.
- RAIN estimates that total post-construction support costs amount to around 10% of total costs. Intuitively this seems appropriate but clearly this percentage will increase for work in remote areas as a result of travel costs and increased staff inputs connected to time spent for travelling. 38 • Village-level averages of service levels, costs and expenditure mask many household-level realities (e.g. the low levels of service received by many poor households and/or households living towards the tail-end of water supply networks, or the high levels of variability in household expenditure on maintaining or improving service levels).
Cost comparisons, whether or not they are based on life-cycle costs or annualised, tend to focus on the costs water supply infrastructure. The result being that limited attention is given to the ultimate purpose of a water supply system -- that is, to provide a range of services. Based on the findings presented in this report, a number of interesting generalisations can be made regarding the influence of life-cycle costs on water services provided by water supply systems:
- Household level data from the WASHCost Project in India shows that households in rural villages are willing and able to make significant capital and recurring investments in their water services. In most cases, the primary aim is to compensate for the inadequate or unreliable services provided by the public water supply. Most common one-off investments in India are in storage tanks, pipes and fittings, illegal connections to the public supply system and small electric pumps that are used to extract water from the public supply system.
- In South Africa, Gibson (2010) reveals the potential benefits of allocating expenditure towards direct technical support, in terms of the levels of service provided by public supply systems. The study illustrates that, in terms of direct support costs, RWH systems that are privately owned and relatively easy to operate have significant cost advantage over more complex public systems. However this will not be the case when sand or check dams are used as a source of public supply.
- Also in South Africa, Moriarty and Butterworth (2003) show that higher levels of service tend to cost disproportionately more than lower levels of service.
Finally, a recent economic evaluation of water buffering concluded with a recommendation that a database should be established for consolidating cost information related to 3R measures (i.e. Rainfall, Retention and Reuse), and for making this information readily available (Tuinhof, van den Ham and Lasage, 2011). This study strongly supports this recommendation and suggests for broadening the scope of this database to encompass all life-cycle costs components (especially post-construction costs) and, as important, the information needed to annualise CapEx (hardware and software) in order to help calculate unit costs per capita and per m3 of water harvested and supplied/used.
