Coimbatore, like many of the rapidly expanding urban regions in India is characterised by intermittent water supply, where households meet their water requirements by accessing multiple sources, both formal and informal. However, there is little information on the various water sources and the actual quantity of water used by household levels in Coimbatore. We examined the patterns of water sources used and the drivers of water use among households in Coimbatore investigating both household-level decision making and intrahousehold bargaining.

Analysis of residential factors influencing water demand in developing countries is much more complex owing to households’ access to and use of multiple water sources, both formal and informal. Additionally, water supply is often limited, so it is hard to observe actual unconstrained “demand”. Currently, there is very little empirical literature on estimating household water in such situations; because of the complexity of multiple source use and constrained supply. The Coimbatore City Municipal Corporation (CCMC) uses a combination of imported surface water, “sweet water”, and local groundwater, “salt water”, to meet the water needs of the city. The supply of multiple qualities of water by the city corporation at different prices, in addition to self-supply and private tankers greatly complicates the situation.

Therefore the objectives of the study were to focus on the factors which influence household water use. Our starting hypothesis was that a household’s decision about which source(s) to use and how much water to draw from each source, depends on both the household-level factors (supply side and demand side) and intra-household factors. On the supply side it is the frequency of water supply which can determine their choice of source mix. On the demand side, the socio-economic status of the household and the water infrastructure (overhead tanks and sumps) household already has will drive the choice of source mix.

To understand household drivers of water use, a primary survey was conducted among households after an extensive pilot survey. The survey was designed to gather data on households’ socio-economic characteristics; sources of water used; the cost of accessing water and the quantity of water drawn from each source. Additionally, metering experiments were also conducted to validate the data from the primary survey. In addition to the household survey, we also examined intra-household and community dynamics via a targeted study of five slums. The study yielded both methodological and empirical insights.