Body of Knowledge on Infrastructure Regulation

Social Pricing and Rural Issues

What do regulators need to do differently to tackle the needs of poor consumers?

Do higher income customers benefit more from subsidies than do poorer customers?

Response Rating:

  • Currently 3/5 Stars.
  • 1
  • 2
  • 3
  • 4
  • 5

3 stars with 2 votes

Click above to rate this response.

Quick Feedback Print this Section

[Response by Sophie Trémolet and Diane Binder, June 2009]

Many governments in developing countries have encouraged subsidies as a social welfare tool to give poor people access to basic public services. Subsidies are generally defined as the difference between the incremental cost of service and the capacity to pay of users1: in many cases, poor people cannot afford the services purchased by the comparatively richer customers, as the overall costs of providing the service are high relative to what customers can afford. In these situations, it might be necessary to provide subsidies to ensure affordable prices to the poor. However necessary, subsidies can at the same time lead to inefficient resource use and the creation of financially weak service providers. Besides, direct subsidies have often been ill-targeted and have benefited middle-class users rather than the poor, while service coverage and quality remain insufficient2.

Whilst subsidies can play a role in the extension of service to poor consumers, there is a growing consensus that subsidy schemes should be more carefully designed. Research has shown that poor consumers rarely benefit from broadly-based subsidy schemes, as they are plagued with several problems:

  • "Errors of exclusion3 are the most frequent. Poor consumers are most of the time not connected to formal utilities and receive water services from other sources, i.e. from small independent service providers (SSIPs). As a result, those consumers are not eligible to benefit from subsidies when such subsidies are passed through tariffs. Errors of exclusion also happen in the case of IBT (Increasing Block Tariffs) when poor consumers actually consume above the first block, and therefore are not eligible to receive the related subsidies. In addition, regulators may require that consumers have secure land tenure titles in order to receive subsidies, which precludes the very poor living in informal settlements.
  • "Errors of inclusion4 also happen. This happens, for example, in the case of IBT when all customers benefit from a low first block, set at a level below average costs of service provision. Surveys show that the case for subsidy should be assessed rather than assumed. In many cases, the willingness to pay of end-users is no lower than the proposed prices. Consumers who may have been willing to pay the standard price, whether they belong to the middle class or they are considered to be poor, may still receive subsidies.
  • Subsidy schemes may continue beyond their usefulness, because policy makers neglect to re-evaluate the schemes, the needs of the poor change, or lobbying groups advocate for further subsidies5.

Possible strategies to address the poor effectiveness of subsidy targeting include the following:

  • Thoroughly assess the amount of subsidies needed: a subsidy that covers the full cost of service would for instance eliminate incentives for efficient use of water.
  • Subsidize access rather than consumption: quantity-targeted subsidies, such as in increasing block tariffs or volume-differentiated tariffs, perform poorly compared to other transfer mechanisms such as connection subsidy mechanisms. Econometrics analysis even shows that they provide smaller subsidies for the poor6. While consumption subsidies are affected by metering conditions and rate of coverage, and are based on the tenuous assumption that low-volume consumers are poor consumers, connection subsidies mitigate most exclusion risks, encourage poor people to obtain access and encourage efficient usage, as prices reflect marginal costs. This is the case only if all subsidies are transferred through connection charges, not if tariffs are left at current levels, which is often the case. However, connection subsidies are not a panacea, since they raise an issue of fairness for consumers who had previously contracted unsubsidized connections.
  • Make sure that subsidy delivery mechanisms are triggered by household indication of demand.
  • Use an administrative selection method (instead of quantity consumed) through proxies such as geographical area, income distribution, household size, etc
  • Use the service-level-targeted subsidy, where subsidies are granted for lower levels of service, presumably unattractive to the wealthier consumers, which enables a self-selection process for accessing the subsidy;
  • Estimate errors of inclusion or exclusion of existing and proposed subsidy schemes to assess options correctly.

Resources

Footnotes

  1. See Chapter 6 Section D. Back to Content
  2. See Chapter 5 Section C. Back to Content
  3. When members of the target group are not captured by the eligibility criteria (Halpern, Foster,& al.) Back to Content
  4. When relatively well-off people find themselves benefiting from subsidies (Halpern, Foster,& al.) Back to Content
  5. See Chapter 5 Section C. Back to Content
  6. Halpern, Foster and al. Back to Content