Incentives for improved performance – How can a regulator develop incentive to discourage energy / water losses?

[Response by Sophie Trémolet and Diane Binder, November 2010]

In the energy and water sectors, incentive-based regulation reflects the level of losses that the regulator allows the distribution company to recover in tariffs. Losses include technical losses, which arise because of the design and operation of the distribution system, and non-technical losses that result from theft, absence of metering, under-billing and poor collections.

The level of losses very much depends on managerial efforts. Whilst reducing technical losses reduces the costs of power purchase / water production, a reduction in non-technical (or commercial) losses can increase revenues. A key question is how loss reductions can be encouraged – potential tools and actions for meeting this objective are listed below:

Setting loss-reduction targets in a multi-year tariff setting regime . An adjustment for losses can be made in the tariff-setting formula through a “grossing-up” mechanism, which allows the distribution company to charge paying customers for electricity losses[1]. The tariff-setting regime would then include a required rate of improvement, which should remain fixed so that investors are encouraged to take the actions needed to bring down losses. Loss-reduction targets should not be measured annually so as to avoid disputes that would reduce the regulatory certainty (Bakovic et al., 2003).

For example, in the water sector in Senegal, loss reduction was a key objective of the reform ahead of substantial investments to reinforce water supply capacities in Dakar. Performance targets were introduced in the afterimage contract of the private operator SDE, and SDE focused on improving operating performance, notably the unaccounted-for-water within water treatment plants (5% of losses allowed), network leakage reduction (with the objective of bringing down technical losses to 15% over a 5 year period) and bill collection (with a target set to 95% of recovery rate the first year, and an improvement of 1 percentage point each year thereafter). In order to focus SDE’s attention on losses reduction, financial incentives were attached to reaching the targets. The operator’s remuneration was made up of two parts: the bid price multiplied by the volume of water which would be billed if SDE attained its technical efficiency and collection targets, and the average tariff applied to the difference between the actual amount of water billed and the target amount. The second component of SDE’s remuneration formula thus acted as a direct financial incentive to improve performance, as it could be positive if SDE exceeded the target or negative if it failed to meet them. This system was effective at bringing about a reduction in losses although the technical efficiency target was set at an over-ambitious level as was realized soon (partly because the technical yield assumed for the base year in 1998 was overvalued at 73% with an associated target of reaching 80% after 2 years). In many affermage contracts, however, such added incentives are usually not incorporated because the fact that the operator is remunerated based on each cubic meter of water sold is consider a sufficient incentive.

Introducing private sector participation (PSP). In several instances, PSP has acted as a catalyst for drastic loss reductions, such as in Latin America for example. Technical and non-technical losses were estimated to be as high as 30 to 40 percent prior to privatization. Most private electricity distribution companies in Latin America have had considerable success in reducing losses. In Chile for example, overall losses were reduced by more than 50 percent in seven years. In Argentina, similar reductions were achieved in even less time. Key success factors included that private companies actually bore the operating risk (thereby having a de facto incentive of reducing costs and increasing revenues) and were given full control over the management of their labor force.

Punishing theft by law . In India, a strict Anti-Theft Law was passed in 2000, providing for a minimum mandatory punishment of 3 to 60 months imprisonment for theft of electricity, mandatory financial penalties, a prohibition from receiving electricity for 2 years and the establishment of special courts to quickly try cases under the new law. Before the law went into effect, citizens were given the opportunity to pay back their bills and “regularize” their status. Over an 18-months period, billings increased by a third and revenues increased by 44% (while average tariffs increased by 15%).

Nevertheless, there are limits in creating specific incentives targeting loss reductions:

  • The effectiveness of managerial incentives . Loss reduction is related to managerial incentives to avoid theft and increase efforts in bill collections. Whether the firm is private or state-owned impacts the efficiency of such incentives. For example, if the firm does not have the ability to hire and fire employees, it may be less successful at reducing losses because some employees may be able to sabotage efforts to reduce losses if they currently profit from such losses in the form of corruption.
  • The extent of political support for collecting from non-paying customers, disconnecting illegal connections and implementing a “law and order” policy. Non-paying customers are often well-off and politically connected (and sometimes comprise public institutions such as army barracks and ministerial buildings) and may apply pressure on the company for maintaining the status-quo.
  • The type of contract between the company and the public authorities . For example in the case of Senegal, added incentives had limited impact, as SDE’s ability to reduce leakage was dependent on investments to be carried out by the publicly-owned asset-holding company, SONES. This momentarily threatened the viability of the contract, as SDE asked for the payment of financial compensation from SONES given that its revenues were directly dependent on the volumes of water sold and on its own technical efficiency performance, both being affected by SONES’s investment delays. However, a conciliation process led to a settlement on the amount of the compensation that SONES would pay to SDE.



  1. For example, in many developing countries, the regulator (or the contract) may include a formula that assumes that a company will have to buy 150 units of electricity for every 100 billed to its customers.