“We just take the programmes as they fall upon us, with their conditions. One donor uses a per capita threshold of 150 US$/capita and wants us to follow one approach, and we will do that. Another uses a threshold of 250 US$/day, but with another approach, and a different degree of community contribution to capital costs. Sometimes we even work in one municipality with various approaches, and we have to explain to a community why they need to provide a contribution to a water project, whereas the neighbouring community doesn’t need to pay for it”. These words, expressed by the deputy-minister of FHIS, the government agency responsible for implementing water and sanitation projects in Honduras, reflect the very real problems of lack of coordination between donors, who come with different intervention approaches and financial conditions, or of the limited capacity of the government to enforce some kind of standardization among donors who put their money through its implementing agencies. But they are also a reflection of how costless the water and sanitation sector is being managed.
If you would buy something, let’s say a bread, it would be nothing but logic that you would have some idea of what it would cost beforehand. The price may vary from one baker to another, and even by the type of bread, but you would have some idea of the order of magnitude. This is not exactly happening in the case of investments in water supply and sanitation.
Obviously, the cost of a water supply system is more complex than the one of bread, as it depends on many factors: the population size and density, the location of the village, the specific design, etc. But, in a country with thousands of rural water supply schemes, most of which are straightforward gravity-fed piped schemes, there should be some idea of the cost ranges for different types of settlements. Uganda is one of the countries where such unit costs are monitored annually, but this is not the case in Honduras.
Besides, it gets more blurred when one also starts taking into account the costs of developing the system. FHIS develops some systems by using contractors, others by sub-contracting NGOs, and again others by delegating part of the procurement to communities themselves. Each of these approaches have their own overhead costs. Then, there are all kinds of complex cost drivers. For example, in some of the systems, communities need to contribute to the capital costs. It may take time for the community to raise these funds. Until all funds are there, contractors cannot do their job and FHIS may have to pay heavy penalties to the contractors because of time delays, eventually raising the costs so much that it offsets the small amount contributed by ommunities. So, in other projects, communities do not have to provide their contribution so as to keep costs down. Surprisingly, there is no insight into how different intervention methodologies influence the costs of implementation.
And this just refers to the costs of developing new systems. Data on all kinds of other costs to run the sector are also absent. In one of my previous blog posts I mentioned the various structures for monitoring and supporting community-based management. In addition, there have been efforts to decentralise regulatory functions to municipalities who are supposed to hire technicians for this. Whereas these structures may well result in improved performance of community-based service providers, there is not yet an idea of how much this whole structure costs, and whether that is then a feasible cost to bear at sector level.
How come that a sector operates with hardly any insight in the costs of developing water systems and running them? Is providing access to water so priceless that a sector can be run without considering costs? The WASHCost project has produced some reports on how cost data are used, or not, in sector decision making. For Honduras, my personal hypothesis is that it partially lies in the fact that the sector is budget-driven: an agency gets a budget to run a donor-funded programme, starts executing that and then sees how far the budget it gets them, rather than defining what is needed, costing it and then seeing what a reasonable budget would be to execute it. A second reason lies probably in the fact that the sector is more interested in efficacy than in efficiency. As long as we get the job done, and get an X number of water systems developed, it is fine. We are not so interested whether a contractor-driven approach is more efficient than the one with community procurement, or the other way round. A third reason – and now I am speculating – is that there are so few economists or financial specialists around in the sector.
I am therefore very glad that one of the areas of our work in Honduras for the next few years will be focussing on costs, following the approaches set out by the WASHCost project elsewhere. In a series of meetings I have had this week to define IRC’s programme in Honduras, the realization transpired that this would very much needed.
If the government is to lead a process of standardization of intervention approaches, it should be based on a good understanding of the costs of these approaches. Then it will have some data to assess which of the approaches are most cost-effective and what realistic cost ranges are for different settlement types. It may even help identifying where there is room to reduce costs and improve efficiency. It may also gain further insight into how these costs can eventually be financed and where there are gaps in financing the sector. I hope that in 4 years from now the vice-minister of FHIS can then tell his donors (hopefully including his own Minister of Finance, and not just external donor ) that access to water and sanitation is priceless, but that investing in it is no longer costless.