Water utilities commonly use complex, nonlinear tariff structures to balance multiple tariff objectives. When these tariffs change, how will customers respond? Do customers respond to the marginal volumetric prices embedded in each block, or do they respond to an average price? Because empirical demand estimation relies heavily on the answer to this question, it has been discussed in the water, electricity, and tax literatures for over 50 years. To optimize water consumption in an economically rational way, consumers must have knowledge of the tariff structure and their consumption. The former is challenging because of nonlinear tariffs and inadequate tariff information provided on bills; the latter is challenging because consumption is observed only once and with a lag (at the end of the period of consumption). A large number of empirical studies show that, when asked, consumers have poor knowledge about tariff structures, marginal prices, and (often) their water consumption. Several studies since 2010 have used methods with cleaner causal identification, namely regression discontinuity approaches that exploit natural experiments across changes in kinks in the tariff structure, changes in utility service area borders, changes in billing periods, or a combination. Three studies found clear evidence that consumers respond to average volumetric price. Two studies found evidence that consumers react to marginal prices, although in both studies the change in price may have been especially salient. One study did not explicitly rule out an average price response. Only one study examined responsiveness to average total price, which includes the fixed, nonvolumetric component of the bill. There are five messages for water professionals. First, inattention to complex tariff schedules and marginal prices should not be confused with inattention to all prices: customers do react to changes in prices, and prices should remain an important tool for managing scarcity and increasing economic efficiency. Second, there is substantial evidence that most customers do not understand complex tariffs and likely do not respond to changes in marginal price. Third, most studies have failed to clearly distinguish between average total price and average volumetric price, highlighting the importance of fixed charges in consumer perception. Fourth, evidence as of late 2020 pointed toward consumers’ responding to average volumetric price, but it may be that this simply better approximates average total price than marginal or expected marginal prices; no studies have explicitly tested this. Finally, although information treatments can likely increase customers’ understanding of complex tariffs (and hence marginal price), it is likely a better use of resources to simplify tariffs and pair increased volumetric charges with enhanced customer assistance programs to help poor customers, rather than relying on increasing block tariffs.
Joseph Cook and Daniel Brent
Dale Whittington and Céline Nauges
The design of municipal water tariffs requires balancing multiple criteria such as financial self-sufficiency for the service provider, equity among customers, and economic efficiency for society. Globally, various forms of water tariffs are in use (e.g., tariffs based on fixed or volumetric charges, single and two-part tariffs, and increasing or decreasing block tariffs) but increasing block tariffs (IBTs) have become popular worldwide over the last few decades for two main reasons. Apart from the fact that IBTs incentivize households to save water by charging large volumes at a higher price, there is a widespread belief that IBTs are pro-poor. The latter would be the consequence of providing all households with a minimum amount of water at a low (subsidized) price while large water users pay higher prices. However cross-subsidization between wealthy and poor households will occur only if poor households’ consumption falls in the low (subsidized) block and if rich households consume in the higher block and pay a price that is above the average cost of supply. These two conditions are rarely met in reality and IBTs often fail to allocate subsidies to the poor effectively. There are a few examples of water utilities making adjustments to the tariff to take into account that poor households with large families are likely to be adversely affected by IBTs. However, the provision of a minimum amount of water for free (as in South Africa), the design of household-specific low-cost water allowances (as in California), or tariffs being adjusted based on household size do not usually improve the targeting of subsidies to the poorest households. The widespread use of IBTs is difficult to rationalize, in particular while knowing that the use of a (simple) uniform volumetric tariff where water provision is charged at its full cost could improve social welfare by removing price distortions and would be easier for households to understand than IBTs. This simple tariff could be combined with some consumer assistance programs to help the poorest households pay their bills.