NOTE NUMBER 232 P U B L I C P O L I C Y F O R T H E privatesector JUNE 2001 Incentive-Based Subsidies Andrés Gómez-Lobo This case study reviews Designing Output-Based Subsidies for Water Consumption Chile’s experience of using tax-funded subsidy payments, contingent on To guarantee adequate and affordable water and sanitation services for delivery of service by water companies. vulnerable households, Chile introduced an individual means-tested water consumption subsidy a decade ago. Although the public authorities T H E W O R L D B A N K G R O U P PRIVATE SECTOR AND INFRASTRUCTURE NETWORK Funding the subsidy through general taxation determine how the subsidy is applied, the mostly private companies deliver allows the regulator to set tariffs according to the service—under a scheme with built-in incentives to ensure cost-effective economic criteria. service delivery by the companies and low wastage by the customers. Targeting allows scarce resources to be directed to In the late 1980s Chile began an overhaul of the geted to individual customers rather than a tra- those who need subsidies legal, economic, and institutional structure of its ditional geographic or universal subsidy. most. Both characteristics water and sanitation sector. An important part of The subsidy program, introduced in the reduce the financial the reform was a new tariff setting methodology early 1990s, relies on the water companies to burden on the government as compared with the aimed at raising water prices to the true economic deliver the service. The government reimburses universal subsidy used in cost of the service. Before the reform water tariffs them for the subsidies on the basis of the actual the past. The write-up of covered less than 50 percent of this cost—and amount of water consumed by each beneficiary this case study has been only 20 percent in regions where production costs rather than a preestablished amount, a method funded by the Public- were high. Steady tariff increases in the 1990s used in some countries. With the most impor- Private Infrastructure doubled real charges for Empresa Metropolitana tant water companies having been privatized Advisory Facility, a de Obras Sanitarias (EMOS), the service provider since 1998, private companies now serve 73 per- multidonor technical in the Santiago metropolitan area and the coun- cent of urban clients. So the subsidy scheme is assistance program. try’s largest water company. Rates increased even essentially being implemented by private com- more for the smaller companies, especially those panies on behalf of the government. operating in high-cost regions. As a result of these large rate hikes, a new How the subsidy works mechanism was required to protect vulnerable By law, the subsidy can cover 25–85 percent of a households. Since connection is almost universal household’s water and sewerage bill for up to 20 in Chile’s urban areas, the affordability of con- cubic meters a month (though the limit now sumption—not of connection—was the main used is 15 cubic meters a month), with the client issue, and a consumption subsidy was the obvious paying the rest. All consumption above the limit solution. Chile chose a means-tested subsidy tar- is charged at the full tariff. I N C E N T I V E - B A S E D S U B S I D I E S DESIGNING OUTPUT-BASED SUBSIDIES FOR WATER CONSUMPTION Box Determining eligibility for subsidies To obtain a subsidy, a household must apply to its municipality, which determines its eligi- 1 An eligibility scoring system called CAS is the main tar- geting instrument used in Chile for distributing means- tested subsidies. It produces a score for each household bility mainly on the basis of a scoring system called CAS (box 1).2 Another important crite- rion is that households must not have payment wishing to be evaluated based on a personal interview arrears with the service provider. at its dwelling. The questionnaire used includes 50 ques- The municipality must award subsidies in the tions on general information, identification of household order of the applicants’ CAS scores. Subsidies members, living conditions, crowding conditions, health are normally renewed yearly for up to three years 2 conditions, comfort, occupation and income, ownership of durable goods, and other socioeconomic variables. Once before a household must reapply. But if a munic- the interview is conducted and the CAS score calculated, ipality has distributed all the subsidies assigned the score is valid for two years, and the household can to it and a new applicant has a lower CAS score use it to apply for many different subsidies. Besides the than the last beneficiary, the municipality must water subsidy, eligibility for pension payments, family withdraw the benefit from this last beneficiary subsidy, free health benefits, and other subsidies is and assign it to the more deserving applicant. determined on the basis of the CAS score. Many municipalities outsource the interviews to pri- Incentives vate survey companies, but still calculate the CAS score. The subsidy scheme has several incentive-based That lowers the risk of collusion between interviewers features. One centers on the fact that the and households, since interviewers do not know the amount of subsidy a beneficiary receives exact relationship between the households’ answers and depends on the level of consumption, and their CAS score. results from two aspects of the program’s design. First, the subsidy is expressed as a per- centage of the household’s bill. It is therefore a Each year the Ministry of Planning price reduction per cubic meter consumed, and (Mideplan) determines, for each region, how no benefits are given if there is no consumption many subsidies are to be granted and how they or delivery of service. Second, the household are to be applied, following several general prin- must pay the full tariff for consumption above ciples: The subsidy is based on the willingness to the limit of 15 cubic meters a month. pay for water services among low-income house- This consumption limit reconciles the need holds. Only households that would be unable to to provide income support to low-income house- purchase what is considered to be a subsistence holds for basic water consumption with the need level of consumption should benefit. And the to preserve financial incentives for efficient subsidy should cover only the shortfall between resource use. In essence, the Chilean water sub- actual charges and willingness to pay.1 As a sidy can be thought of as a rising block tariff, crude proxy for willingness to pay, Mideplan where only means-tested households have access uses the benchmark set by the Pan-American to the lower priced initial consumption block.3 Health Organization—that no household The fact that the subsidy scheme requires should pay more than 5 percent of its monthly households to pay a fraction of the bill even income in water and sewerage charges. It is when their consumption does not exceed 15 unclear whether vulnerable households in cubic meters helps to maintain good payment Chile would be willing to pay more or less than habits among clients. It also preserves service this 5 percent. providers’ incentive to improve commercial The subsidy scheme is funded entirely from efficiency, since their income depends in part the central government’s budget. Using house- on the payment of this remaining charge. And hold survey information for each region and the additional eligibility requirement of not each company’s published tariffs, Mideplan can having payment arrears has led to an improve- determine how many households need a sub- ment in clients’ payment record. sidy and how large benefits need to be to meet Another incentive-based feature of the the benchmark for each region. scheme rests on the relationship between the government and the service provider. This rela- Financial control and procedures for the tionship is mediated by the subsidy law and its Figure subsidy scheme accompanying regulations.4 The law requires the companies to bill beneficiary clients net of the subsidy amount and then bill the municipality for the subsidies granted. The municipality is thus a 1 Presents an invoice to municipality Water company Validates invoice Pays water client of the service provider, meaning that it can and sends it Municipality company be charged interest for late payment and that the to governor service can be discontinued as a result of non- Aggregates invoices of all municipalities 3 payment (so that in the next payment period the Regional Transfers funds and presents regional governor to municipalities service provider can charge beneficiary house- invoice to Ministry holds the full amount of the bill). of the Interior This setup gives municipalities a strong Validates regional Ministry of incentive to transfer payments quickly to opera- invoices and presents the Interior consolidated invoice Undersecretariat tors. The central government funds for the pro- to Budget Office for Regional gram are earmarked, so municipalities do not Development stand to benefit financially from withholding Approves expenditure Ministry payment to the water companies. And the polit- of Finance and deposits funds Budget Office in each regional ical wrath that could arise if they failed to pay treasury’s account the service providers—and thus lost the benefit for households—is potentially costly. The financial flows and control of the pro- perhaps through the reduction of payment gram are concentrated in the Undersecretariat arrears by poor households. for Regional Development of the Ministry of the Interior (figure 1). The process requires that Take-up the company and municipality have synchro- In 1998 nearly 450,000 subsidies were distributed nized lists of beneficiary households and that nationally, benefiting almost 13 percent of house- the interior ministry verify that the regional holds by an average US$10 a month. The total invoice is consistent with the number and value cost was US$33.6 million. In some regions where of subsidies for the region approved in the incomes are low and water charges high, close to annual budget. The arrangement is clearly a third of households received the subsidy. On bureaucratic, and municipalities are often average, 52 percent of benefits in each region unable to pay the companies’ invoices on time. accrue to the three lowest income groups, and Some companies charge the municipalities only 23 percent leak to the five highest income interest for the payment delay. The municipali- groups (figure 2).5 Subsidies represent a larger ties must bear the interest and debt costs result- share of income for poorer households, nearly 8 ing from late payment, since there is no percent for the lowest income group. provision in the national water subsidy budget for these charges. Lessons Despite this problem, the fact that companies The introduction of the subsidy—and especially receive a reimbursement for services and subsi- the targeting results achieved—have been key to dies already delivered has several benefits. The Chile’s ability to raise water tariffs to levels reflect- arrangement gives the companies full incentives ing costs without compromising its social and for providing efficient and reliable service. The distributional goals. And the costs to the govern- subsidies accrue to households, not companies, ment of doing so have been low. The cost of the and the amount of resources distributed is inde- subsidy in 1998, US$33.6 million, was well below pendent of the service provider’s operational the cost of the previous universal subsidy scheme. efficiency. Companies should be indifferent Before the reforms in 1988 the water and sewer- with respect to the subsidy scheme and receive age sector had a financial deficit of 2 percent of no financial benefit from the program, except assets. But in 1998 this situation was reversed. I N C E N T I V E - B A S E D S U B S I D I E S DESIGNING OUTPUT-BASED SUBSIDIES FOR WATER CONSUMPTION Distribution of water subsidies across Chile, some municipalities still lack sufficient Figure income deciles, November 1998 capacity to adequately administer and control the 2 subsidy scheme. For countries with less institu- Share of total subsidy expenditure tional capacity such a complex system may not be Average subsidy as a share of household income viable. These countries could adopt simpler tar- Income decile Percent geting mechanisms, for example, a scheme based viewpoint 0 5 10 15 20 on a geographic poverty map, like that used in 1 Colombia. Moreover, a scheme using a connec- is an open forum to 2 tion rather than a consumption subsidy will encourage dissemination of 3 require less institutional capacity, since house- public policy innovations for holds’ eligibility must be evaluated only once. 4 private sector – led and Third, an individual means-tested subsidy market-based solutions for 5 may be expensive to apply. Chile uses the same development. The views 6 targeting instrument to distribute several wel- published are those of the 7 fare benefits, lowering the administrative costs authors and should not be 8 significantly. Applying such a scheme for only attributed to the World one subsidy program may be too expensive. Bank or any other affiliated 9 Again, however, the administrative costs for a organizations. Nor do any of 10 connection subsidy are much lower than those the conclusions represent for a consumption subsidy. official policy of the World Note: Income deciles are based on per capita household income. Data include only households with shared or own water connections. Bank or of its Executive Source: Mideplan (Ministry of Planning), “Encuesta de Caracterización Socioeconomica Nacional” (Santiago, 1998). Directors or the countries they represent. Companies reported a surplus of close to 4 per- Notes cent of assets and net profits of US$107 million, 1. This is not to say that no further benefits should To order additional copies more than three times the cost of the subsidy accrue to low-income households for purely distribu- contact Suzanne Smith, scheme (excluding administrative costs). tional reasons, only that the best way to meet such goals managing editor, Room I9-017, Despite the successes of the subsidy program, may be through general welfare programs, not a sectoral The World Bank, several issues need to be considered if such a consumption subsidy. 1818 H Street, NW, scheme is to be replicated in other countries. 2. CAS comes from Comités de Asistencia Social Washington, DC 20433. First, metering is a prerequisite for this type of Comunal (Communal Social Assistance Committees), output-based consumption subsidy. For coun- which promoted the idea of using a standard measure for Telephone: tries with low coverage of meters among low- allocating social resources among the poor. 001 202 458 7281 income households, such a scheme may not be 3. A consumption-based subsidy requires that each Fax: viable. In these countries, however, the most household have a meter. This is not much of an issue in 001 202 522 3181 pressing social issue usually is not ensuring that Chile, where metering is almost universal in urban areas. Email: water is affordable, but increasing the number 4. These laws and regulations are available in ssmith7@worldbank.org of connections. For that purpose, a means- Spanish at www.siss.cl. tested subsidy analogous to the one in Chile 5. By comparison, in Colombia, the only other Latin would be easier to apply, since metering is not American country with an important formal water subsidy Printed on recycled paper required. Service providers could charge new scheme, 37 percent of subsidies reach the five highest clients the true cost of connection—perhaps income groups. providing some credit by permitting payment in installments—minus a subsidy to eligible house- Andrés Gómez-Lobo (agomezlo@econ.facea.uchile.cl), Univer- holds. The government could then reimburse sity of Chile. the service providers on the basis of the number of eligible households connected. Second, the means-tested targeting used in Chile requires a certain amount of institutional capacity, especially at the municipal level. Even in This Note is available online: www.worldbank.org/html/fpd/notes/