Increasing block tariffs seek a cross-subsidy mechanism between the water network users, based on the common assumption of a low water price elasticity. In Manaus, the capital city of the Brazilian state of Amazonas, where most of the 1.6 million dwellers are supplied through a municipal water network, a substantial consumption drop followed the tariff increase of 2004. This drop questions the cross-subsidy capacity of the current structure. We see this 31.51% tariff increase as a natural experiment applied to the whole network user population of Manaus, and this allows us to measure the impact on monthly consumption of metered households, using month-on-month differences between years 2003 and 2004.
Keywords: Brazil, Impact evaluation, Natural experiment, Price-elasticity, Tariff, Water demand