Water & Wastewater Water and Sanitation a Division Utilities The World Bank Indicators 2nd Edition Guillermo Yepes Augusta Dianderas ( 1996 International Bank for Reconstruction and Development The World Bank 1818 H Street, NW Washington, DC 20433 USA The Water and Sanitation Division of the Transportation, Water, and Urban Development Depart- ment is addressing the sectoral challenges articulated in the World Bank's Water Resources Man- agement Policy Paper and in the World Development Reports of 1992 and 1994. These challenges involve implementing key management principles: * Water should be treated as both a social and economic good. * Water should be managed within a comprehensive framework, taking into account cross-sectoral considerations. * Water should be managed at the lowest appropriate level, employing a demand-based approach and facilitating participation of all stakeholders. * Institutional and policy reforms should be linked to incentives influencing decision making. The Division's work program is therefore selectively focused on three distinct themes: water resources management, performance of water and sanitation utilities, and service provision through non-formal institutions. This document is published informally by the World Bank. Copies are available free from the World Bank. Contact Ms. Mari Dhokai, Room S4-001, Telephone: (202) 473 3970, Fax: (202) 477 0164, Email: mdhokai@worldbank.org The World Bank does not accept responsibility for the views expressed herein, which are those of the author and should not be attributed to the World Bank or its affiliated organizations. The findings, interpretations, and conclusions are the results of research supported by the Bank. The designations employed and the presentation of the material are solely for the convenience of the reader and do not imply the expression of any legal opinion whatsoever on the part of the World Bank or its affili- ates concerning the delimitations of its boundaries or national affiliation. INDICATORS 2nd Edition WATER & WASTEWATER UTILITIES A, I& 'MEOW,5 INDICATORS WATER AND WASTEWATER UTILITIES INDEX Page No. Foreword .......................................................i Information About Cities And Utilities Cited ...................................................... ii SET I. OPERATIONAL INDICATORS Introduction ........................................................ A. WATER CONSUMPTION A. I Unit Consuption .2 A.2 Water Consumption & Metering .3 A.3 Distribution of Water Consumption .4 A.4 Consumption by Main Users Category .4 A.5 Ratio of Peak Day to Average Day .5 A.6 Water Price & Income Elasticities .5 B. WATER DISTRIBUTION SYSTEM B. I Length of Water Piped Systems ......................................................7 B.2 Storage Volume ......................................................8 B.3 Pipe Breaks ......................................................9 B.4 Pipe Breaks as a Function of Pipe Material ....................................................... 9 C. UNACCOUNTED FOR WATER C. I Water Losses ...................................................... 10 C.2 Composition of UFW ...................................................... 12 C.3 UFW Effective Reduction Programs ...................... ................................ 12 C.4 Sustainability of UFW Reduction Programs ...................................................... 13 D. WASTE WATER D. I Length of Sewer Systems ...................................................... 14 D.2 Infiltration Flows in Sewer Systems ...................................................... 14 E. WASTEWATER TREATMENT E. I Typical Composition of Untreated Municipal Wastewater . . ..................................... 15 E.2 Typical Constituent Removal Efficiencies ...................................................... 16 E.3 Removal of Microorganisms ...................................................... 16 F. PERSONNEL F. I Number of Staff ...................................................... 17 F.2 Staff Composition ...................................................... 18 F.3 Training Effort ...................................................... 18 G. MISCELLANEOUS INDICATORS G. 1 Vehicles/1000 Water Cornections .................. .................................... 18 G.2 Meter Reading ....................................................... 19 G.3 Meter Maintenance and Replacement Practices .19 Page No. SET II. FINANCIAL INDICATORS Introduction .........................................21 A. EFFICIENCY INDICATORS A. 1 Working Ratio ........................................ 29 A.2 Operating Ratio ........................................ 23 A.3 Accounts Receivable/Collection Period ........................................ 24 A.4 Percentage Contribution to Investment ........................................ 25 B. LEVERAGE INDICATORS B. 1 Debt Service Coverage Ratio ........................................ 26 B.2 Debt Equity Ratio ........................................ 27 C. LIQUIDITY INDICATOR C. l Current Ratio ........................................ 28 D. PROFITABILITY INDICATORS D. 1 Retum of Net Fixed Assets ........................................ 29 D.2 Return of Equity ........................................ 30 E. OPERATIONAL RATIOS E. 1 Personnel ........................................ 3 1 E. 1. 1 Personnel Costs ........................................ 3 1 E. 1.2 Staff Productivity Index ........................................ 31 E.2 Composition of Operational Costs ........................................ 33 E.3 Unit Operational Costs ........................................ 34 SET III. OVERVIEW OF TARIFF RATES AND STRUCTURES A. I Backgroud .35 A.2 Tariff Structure .35 A.3 Domestic Tariff .36 A.4 Average Charge & Incremental Cost .38 A.5 Rate Discrimination by Consumer Group .43 A.6 Water Billings. Consumption & Users .44 A.7 Conclusions & Hypothesis to be Tested .45 ANNEX: Personnel Costs as an Indicator for Water and Sanitation Utility Peiformance in Developing Countries (Infrast-ucture Note: W & S - I2). We hereby wish to extend our appreciation and thanks to Mr. Jorge A. Serraino for his assistance in the preparation of this report. FOREWORD Indicators can be a valuable tool to sector staff and practitioners working in the evaluation of operations and investments of water and sanitation utilities'. To make this job more manageable indicators on Water and Waste-Water services, mainly in urban areas, have been grouped into three sets: 1. Operational Indicators (first edition, April 1993) 2. Financial Indicators (first edition. June 1994), and 3. Overview of Tariff Rates and Structures (first edition, June 1994). In response to the heavy demand. this second edition has been updated and expanded with additional information collected since the three sets were first published. Indicators have been collected firomil a selected group of utilities firomii industrialized and developing countries. Indicators from the former group are believed to represent "acceptable" or "desirable" outcomes or best practice. General infonnation about the utilities cited is presented in Tables I and 2. Staff working in operations have day to day contact with utilities and therefore are in the best position to collect the infonnation requil-ed to keep this information Lup to date. It is only through your collaboration and that of practitioners that we will be able to keep these indicators current and to expand them. We appreciate the inputs and feedback received from staff in operations after the first edition was published and look forward to continue receivinig your comments. suggestion and additional data. Guillermo Yepes Augusta Dianderas Sectoral and Project Performance Indicators in Bank financed Water and Waste-Water Operationis. A First Edition Note, TWU Departimleit, ESD. April 1995. - i - INFORMATION ABOUT CITIES AND UTILITIES CITED All the companies in the sample are utilities responsible mainily for ulban centers with service areas covering a city, region or country. The population to be served by these utilities ranges from about 0.2 million to over 17 million. In terms of the service provided these utilities can be divided into three groups: water ( I I), waste water (4 - all in Korea) and water and waste water (19). All the public utilities have some degree of autonomy in the sense that they manage, at least, their own budget. However, their autonomy, regulatory system, sector policies and political forces that shape the behavior of these companies are not thoroughly documented in the source reports to allow a meaningful analysis. Therefore, information on these factors. imlpol-tant as they are, is not documented here. Table 1 Country Nanie of Utility Area Serviced Tyvje qf _____________________________ ~ ~~~~~~~~~Utility Latin America: Brazil Sao Paulo State Water Co. State of Sao Paulo Regional/Public Brazil Sta. Catarinia State Water Co. State of Sta. Catarina Regionatl/Public Brazil Minas Gerais State Watel Co. State of Minas Gerais Regional/Public Chile Obras Sanitarias of Valparaiso Metropolitan Valparaiso Municipal/Public Chile Empresa Municipal Obras Sanitarias Metropolitan Santiago Municipal/Public Colombia Water & Sewage Co. of Bogota Metropolitan Bogota Municipal/Public Costa Rica Inst. of Water & Sewage of C. Rica 70% of the Country National/Public Africa: Algeria Water Supply Co. of Oran Oran, Tiemcen, Ain &Mascara Regional/Public Algeria Water Supply Co. of Aninaba Annaba & El Tarf Areas Regional/Public Ghana Ghana\s Water & Sewerage Corp. 40% of the Country National/Public Morocco National Office of Potable Water 75% Urban Morocco National/Public Nigeria Katsina State Water Board 50% of Katsina State Regional/Public Nigeria Kaduna State Water Board 50% of Kaduna State Regional/Public Europe/Central Asia: Turkey Bursa's Water Supply & Sewrg. Authi. Metropolitan Bursa Municipal/Public Turkey Ankw-a's Water Supply & Servig. Auth. Metropolitan Ankara Municipal/Public South & East Asia: Pakistan Karachi Water- & Sewerage Board Metropolitan Karachi MuLnicipal/Public China Cbaliaocihiii Water Supply Co. Metropolitan Changchun Municipal/Public Korea Kwaungju Cosnfluct. Bureau (Se\ig. Div.) Metropolitan Kwangju Municipal/Public Korea Pusui City Goviit. (Sewerg. Division) Metropolitan Pusan M unicipal/Publ ic Korea Seoul Sewerage Division Metropolitan Seoul Mun1icipal/Public Korea Taejoin City Ciovint. (Sewerg. Division) Metropolitan Taejon Mun1icipal/Public Philippines Metrop. Waterwor-ks & Sewrs. Systemil Metropolitan Manila Municipal/Public High Income Countries: Belgium Compaglie hitelcomimilluniale Brussels area and surroundings Municipal Bruselloise des Eaux England Wessex Water Central/South Regiona;lPrivate France Banilieue Metro Area/Paris Private France Bordleaux City Private France Societe des Eaux de Marseille City Private Germany Hambuir City Public Japan Osaka City Public Japan Tokyo Capital City Public Singapore Public Utilities Boardl City/State Public Spain AgUIas .le Alicante City Mwlicipl/Pnivate Spain Aguas de Murcia City MtU1icipXl/Pivate Spain Aguas de Torrevieja City Mtuniciplx/Privatc jjt Table 2 Utility Year of Connections Service Population Population Information *000 / Provided '000 Served % Sao Paulo 1988 5080 W&S. 17,500 9o b1 S. Catarina 1990 529 W3 S. .3000 85b/ Minas 1990 1870 W &S. 7,600 96 b/ Valparaiso 1990 526 W & S. 760 97 b6 Santiago 1994 1811 :W S. 5.000 100 b/ Bogota 1990 1 2 . 5,000 94 b/ Costa Rica 1991 437 W&S. 1700 84b Oran 1992 204 Water 1,900 68 Annaba 1992 34 Water 600 73 Ghana 1988 197 w & S. 7,400 60N. Morocco 1990 601 Water 11,300 90 Katsuna 1W990 520 55 Kaduna 1990 43 Water 1,600 76 Bursa 1991 ~~~~307 W &S. 710 91 b/ Ankara 1988 832 W&S. 2,200 93b. Karachi 1989 102 3 W & S. 6.300 N.A. Changchun 1990 388 Water 1.600 90 Kwangju 1990 96 d/ Sewrge 1,100 91 d/ Pusan 1990 -287 d/ Sewrge 3.700 95 d/ Seoul 1990 ~~~~~1575 d/ Sewrge 8.700 72 d/ Taejon 1990 90 d/ Sewrge 1.000 87 dj Manila 1988 641 W&S. 5,000 87 d/ Brussels 1991 N.A. W& S. 1.000 N.A. Wessex 1991 N.A. W&S. N.A. N.A. Banlieue 1987 N.A. Water N.A. N.A. Bordeaux 1982 N.A. Water N.A. N.A. Marseille 1992 N.A. W & S. 1,00( N.A. Hamburg 1990 N.A. Water 1,901 N.A. Osaka 1990 N.A. Water 1.200 N.A. Tokyo 1990 N.A. b W S. 4,900 N.A. Singapore 1994 760 Water 21,800 100 Alicante 1990 17 &.600 10061/ Murcia l992 b5 W S 3510b Torrevieja 992 946 W&S. 200 loo b6 N.A: Not Available a/ Includes W and S connections for W & S utilities b/ Corresponds only to Water services c/ Includes 2.1 million inilabitants in rural areas (17% of the total population) d/ Corresponds to sewerage services only. iii WATER & WASTEWATER UTI LITI ES SET I OPERATIONAL INDICATORS U ... -~~ Introduction Operational indicators can be very useful in assessing the performance of water and waste-water utilities in the course of project formulation and supervision of Bank financed projects and in sector work. Invariable, any indicator portrays an incomplete picture of an utility as it often excludes other contributing factors of performance such as accountability of institutions and incentives, that are not readily captured or quantif'iable. In addition, utilities face different social, political and financial constraints. These factors and constraints need to be taken into account when evaluating the performance of an utility. It follows that indicators should not be used in a rigid prescriptive fashion, and judgment is required to interpret them or to set acceptable or desirable targets. The idea of a comprehensive and up to date list of indicators fiom a large number of utilities world wide is attractive but probably not realistic due to the costs involved in collecting this information. We also recognize the interest in con-elating indicators to other variables like city or utility size or to GNP. No attempt. however, has been made in this direction at this time as the data base is still small and, therefore, the conclusions reached forn such correlations could be spurious. Indicators should be used selectively. The use of too many is likely to dilute the power of all of them. Managers may become confused about priorities and burdened by paperwork and overwhelmed by detail. On the other hand, the use of too few may not adequately describe the utility's performance and progress in reaching its goals. The quality of the management information systems should be assessed before discussing with sector officials about which indicators are important and relevant and to whom and how often they should be reported to. If the management information systems are deficient or infomiation is not produced on time. it is important to develop a reliable system and the incentives to keep it relevant and up to date. Indicators are as good as the data base from which they are (derived. For instance, lack of metering of production or consumption casts doubts about the reliability of estimates on water consumption or water losses. There is also the danger of reducing performance evaluation to number-s and for utility managers and staff to play games with them. Therefore: * watch out for "creaming", e.g., managers tend to produce the numbers they are asked to deliver. * anticipate resistance. Hard information about efficiency and effectiveness can be threatening to insecure managers who doubt their ability to compete, * involve the utility managers in developing corrective measures. This is probably the best way to deal with resistance. Managers need to "own" the specific measures to be implemented and the indicators to be generated and to be convinced that they will help them improve the service they are in charge of, and * analyze the evolution of the indicators to assess progress or deteiioration of utilities' performance. Operational Indicators Water & Wastewater Utilities Page 2 Set I A. WATER CONSUMPTION A.1. UNIT CONSUMPTION Total water consumption, based on metered consumption. is reported as: * average daily consumption per person served (liters per capita per day, lpcd). * average consumption per connection per month (m'/month/connection, m/mi/c). Countries Water Consumption CountIy/City Year Ipcd m3/mn/c Algeria (average) 1990 46 N/A Brazil (average) 1989 151 1 25 a/ * Brasilia 1989 211 60 b/ * Sao Paulo 1988 237 38 c/ * Sta. Catarina 1990 143 22 * Minas 1990 154 25 Chile * Santiago 1994 204 34 d/ * Valparaiso 1992 N/A 23 China, Changchun 1 990 260 33 Colombia, Bogota 1992 167 30 Costa Rica 1991 208 29 1994 197 226 Cote d' Ivoire, Abidjan 1993 N/A 34 Senegal, Dakar 1993 N/A 36 Belgium, Brussels 1991 N/A 29 Canada (average) 1984 1 431 82 France, Paris, C. Banlieue 1987 256 i 75 e/ Japan, Tokyo 1990 I 355 57 Spain ' * Alicante 1987 267 16 * Murcia 1992 268 33 UK (average) I1990 136 1 18 USA (average) 1984 666 89 N/A = data not available Note: One connection serves more than one housinLg unit. a/ 1.3 units/water connectioni d/ 1.1 units/water connection b/ 2.3 ullits/water connectioni e/ 3.5 units/water coinection c/ 1.4 units/water connection Water & Wastewater Utilities Operational Indicators Set I Page 3 A.2. WATER CONSUMPTION AND METERING The effect of metering, as a proxy for price, on water consLImption is shown in the following graphs: Water Consumption vs Metering a) Canadian Utilities Consumption, liters/day/capita 700 \ * ~~~~~~~~Line of best fit 600 500 400 0 20 40 60 80 100 % Metering SoIure(: AWWA Water Util. Op. Data, 1985 b) Brazilian Utilities 80 ~Consumption (m3/m/c) 70 60 50 40 Line of best fit 30 20 10 20 40 60 80 1 00 '%c, Metering Sourcc: Catalogo Brasileiro de Engenharia Sanitaria e Ambiental. CABES. 1990.. Operational Indicators Water & ISWastewater Utilities Page4 Set I A.3. DISTRIBUTION OF WATER CONSUMPTION Distribution of water consumption as a function of the number of connections. Distribution of Residential Water Consumption % of Consumption 9( 8(J 70 60 50 40 30 10 0 ) I 0 2(1 30 4(0 50 60 70 80 9( 100 % of Connections Bogota Washington DC A.4. WATER CONSUMPTION BY MAIN USER CATEGORY San Jose S. Catarina Bogota Macao Minsk USER % % / % % % % % % % Conni Cons. Conn Cons. Con,ti Cons. Conii Cons. Con/l Coils. Residential 91 71 90 73 94 79 85 43 92 63 Commercial 6 8 8 11 5 9 14 50 --- --- Industrial 2 15 1 11 a/ 0.4 7 2 b/ 5 c/ 13 Official 1 6 1 5 0.2 5 1 7 6 19 TOTAL 100 100 100 100 100 100 100 100 100 100 a/ Industry has access to other supply sources [ground water]. Total industrial water consumptioni is not captured in the utility statistics. b/ Drinking water. Includes commercial uses. c/ Non-Potable water (technical) Water & Wastewater Utilities Operational Inidicators Set I Page A.5. RATIO OF PEAK DAY TO AVERAGE DAY (USA AVERAGE 1990) Ratio Peak Day/Average 1.81 1.82 1.77 1.5 ' ... 8 - - - 5 1.6 .6 1.5 ~ ~ ~ ~ ~ ~ ~~~~15 1.4 10- 25 25 - 50 50 - 100 > 100 Average (all cities) Population Served (000) Source: AWWA. Water Industry Database: Utility Profiles. 1992. A.6.a WATER PRICE & INCOME ELASTICITIES Investigator Price Elasticity In come Comments Elasticit Neiswiadomy & Increasing Iblock structures. Random sample Molina (1989) - 0.55 0.14 of 101 customers' monthly water use records from the city of Denton. Ohio. Neiswiadomy & Average price under an increasing block rate Molina (1993) - 0.63 0.64 structure. Uses data from the 1984 AWWA survey. USA Chi-Keung Woo Average price. Uses monthly consumption (1992) -0.38 0.28 data collected for Hong Kong during 1973 - 1984. Marginal and average price. Uses 1984 Neiswiadomy -0.11 0.44 AWWA survey of 430 utilities. Reported (1992) -0.28 0.25(a) results correspond to the North Central Region, USA. IWACO (1989) -0.29, - 0.33 0.40, 0.50 Monthly sales of metered domestic consumers in Bogor, Indonesia Average water price. Cross-sectional IWACO (1992) - 0.68 0.37 analysis of 100 households in Jakarta. Indonesia Martin (1992) -0.70. - 0.60 0.18, 0.27 Average and marginal prices. Cross- -0.49. - 0.32 0.04, 0.17 sectional analysis of 19,000 households in urban and suburban Columbia. USA. Rizaiza (1991) - 0.48 0.11 Average water price. Cross-sectional analysis of 400 households in Saudi Arabia. Hubbell (1977) - 0.48 0.36 Cross-sectional data for 230 households in Nairobi, Ken'a Note: (a) The variable is not statistically significant. Operationtal Indicators IW'ater & Wastewater LUtilities Page.6 Set I A.6.b WATER. SHORT RUN PRICE ELASTICITY FOR DOMESTIC USERS Investiaator Price Elasticity Comments Domestic use in Washington, Carver and Boland (1980) -0.1 D.C., USA, covering the period 1969 to 1974. Domestic use in Tucson, AZ., Agthee and Billings (1980) -0.18, -0.36 USA, for the period January 1974 through September 1977. Domestic use in Tucson, AZ., Martin et al. (1983) -0.26 USA, covering the period July 1976 through December 1979. Domestic use in Malmo, Hanke and de Mar6 (1982) -0.15 Sweden, covering the period 1971 - 1978.. Domestic use in Toowoonba, Gallagher et al. (1977) -0.26 Queensland, covering the period 1972/3 to 1976/7. Domestic use in France, Boistard (1993) -0.17 covering the period between 1985 and 1990. A.6.c WATER. PRICE ELASTICITY FOR INDUSTRIAL USERS Investigator Price Elasticity Comments For the average price and the Williams and Suh (1986) -0.74. -0.44 marginal price, USA. Paper and chemical plants, USA. Ziegler (1984) -(.98 Average price. Rees (1969) -0.96 Chemical water use, UK. Cross-sectional data for cotton, Gupta and Goldar (1991) -1.32 textile, paper, dairy, ball-bearing, and distillery, India (1983-84). Cross-sectional data for different Tate et al. -0.5 to - 1.2 industrial subsectors, Canada (1981-1986). Metaplanners (1992) -0.45 Steel and related industries, India. Source: Ramnesh Bathia et al ( 1994). Water & Wastewater Utilities Operational Indicators Set) Page 7 B. WATER DISTRIBUTION SYSTEM B.1. LENGTH OF WATER PIPED SYSTEMS Length of the water distribution pipe system as a function of: * the number of people served [meters/person], * number of coinections [meters/connection]. CountylCity Year Unit Length Unit Length (mtslperson) (mtlconnection) Brazil (average) 1989 2.3 12.5 a/ * Brasilia 1989 1.8 17.1 b/ * Sa6 Paulo 1993 2.6 10.7 Chile * Santiago 1994 1.6 8.7 * Valparaiso 1992 N/A 11.3 Colombia, Bogota 1992 1.4 8.0 Costa Rica (average) 1990 N/A 11.6 Togo (country total) 1990 9. 1 c/ 74.6 c/ Philippines * Cabanatuan 1994 N/A 6.8 Romania, Bucarest 1994 1.3 d/ N/A Belgium, Brussels 1991 N/A 9.8 France, Marseille 1992 2.6 N/A Germany, Hamburg 1990 2.9 N/A Japan (average) 1990 4.1 11.8 * Osaka 1990 3.9 N/A * Tokyo 1990 N/A 9.4 Singapore 1991 1.6 5.5 Spain * Alicante 1992 N/A 7.6 * Murcia 1992 3.1 12.6 USA (average) 1984 4.9 24.0 1990 6.4 N/A Notes: a/ 1.3 units/connection b/ 2.3 units/coninection c/ Significant number of standposts (56% of population served). d/ 7%7v of population is served by standposts. Operational Indicators Water & Wastewater Utilities 13-e 8 _ Set I B.2. STORAGE VOLUME a) Storage volume in the distribution system expressed as: * m'/person served. * m'/water connection. .ountiy/City Year Storage Volume Storage Volume Country/City Year m3 eso mleonnec m3/person m/cne Chile, * Santiago 1990 n.d. 0.7 * Valparaiso 1992 0.5 1.9 Colombia, Bogota 1991 0.2 0.9 Mexico, Monterrey 1987 0.4 2.3 Belgium, Brussels 1991 n.d. 1.0 Canada (average) 1984 0.6 3.9 France, Bordeaux 1982 n.d. 0.8 Singapore 1990 0.4 1.2 Spain, * Murcia 1992 n.d. 0.7 * Torrevieja 1992 0.5 1.9 USA (average) 1984 0.6 3.0 b) Storage volume as a function of population served USA 1/. Population Served m3/person ('000) 10 - 25 1.08 25 - 50 0.87 50 - 100 0.82 100 - 500 0.80 500 - 1000 0.71 > 1000 0.55 1/ AWWA - 1984 water utility operating data. Water & Wastewater Utilities Operational Indicators Set I Page 9 B.3. PIPE BREAKS Number of pipe breaks per year per 100 kms of pipes in the water system. A higher number is indicative of problems due to materials. installation. age. soil conditions, traffic and of inadequate maintenance. Country/City Year Pipe Breaks Breaksi OOkmlyr. Chile, Santiago 1994 31" Colombia. Bogota 1994 187 Belarus * Minsk 1993 70 * Gomel 1993 25 Belgium, Brussels 1991 21 Singapore 1990 17 USA (average) 1990 17 * Denver, Colorado 76-83 7 * Oakland, California, EBMUD 73-82 16 Note: a/ Down from 39ih 1991. B.4. PIPE BREAKS AS A FUNCTION OF PIPE MATERIAL Infonration on different types of pipes materials. It is useful when designing, strategries to reduce physical water losses. Pipe BreaksilOOkmlyr. Pipe Material __ __ _ __ ___ Densver EBMUD Bogota Santiago A.C. 3.7 10.3 294 38 Cast Iron 7.5 2.6 --- 23 Concrete 0.9 --- Ductile Iron 1.8 --- --- Galv. Iron 35.5 5.6 --- --- PVC --- --- 78 8 Steel 0.4 --- --- 6 Other Materials --- --- 58 --- Average 6.8 16.8 187 31 Source: a/ Guiding Manual. Rehabilitation Criteria for Water mains. AWWA. 1986. Operational Indicators Water & Wastewater Utilities Page 10 Set I C. UNACCOUNTED FOR WATER (UFW) C.1. WATER LOSSES A major concern about operations of a water utility is the level of UFW. UFW reflects the difference between the volume of water delivered to the distribution system and the water sold. The level of UFW is considered a good proxy for the overall efficiency of operations of a water utility. UFW includes physical losses [pipe breaks and overflows] and commercial losses [meter under-registration. illegal use including fraudulent or unregistered connections and legal. but usually not metered uses like fire fighting]. Unaccounted for water (UFW) is expressed as: . a percentage of net water production (delivered to the distribution system, % UFW). . as m/day/kmn of water distribution pipe system network (ml/day/km d.s.). The average rate of UFW in the developing countries of this sample is 37%. more than twice what is considered acceptable in industrialized countries (less than 20%). The highest rate is found in Bursa. Turkey, with 62% and the lowest in Abidjan. Ivory Coast with 17%. Caution should be used in interpreting UFW data, however, as some reported UFW r atios are not more than gross estimates since full metering is not in place and utilities often do not adlhere to the definition given above. Waler & Wastewater Utilities Operational Indicators Set I Page 11 UNACCOUNTED FOR WATER (UFW) Count.ylCity Year Water Losses Water Losses Country/City Year % UFW mnldaylkm d.s. Brazil (average) 1989 39 42 * Brasilia 1989 19 27 j * Sao Paulo Metrop. Area 1992 40 a/ 70 * S. Catarina 1990 45 n.d. * Minas 1990 25 n.d. Chile * Valparaiso 1990 41 n.d. * Santiago 1990 28 52 1994 22 44 Colombia, Bogota 1991 40 135 Costa Rica 1991 45 n.d. Ivory Coast, Abidjan 1993 17 n.d. Algeria, Annaba 1992 35 n.d. Gambia, Banjol 1993 27 n.d. Guinea, Conakry 1993 53 n.d. Senegal, Dakar 1993 29 n.d. Ghana 1988 49 n.d. Morocco 1990 32 n.d. Nigeria * Katsina 1990 44 n.d. * Kaduna 1990 41 n.d. Togo 1990 22 7 Turkey * Bursa 1991 62 n.d. * Ankara 1988 45 n.d. Pakistan, Karachi 1989 40 n.d. China, Changchun 1990 40 n.d. Philippines, Manila 1988 59 n.d. Thailand. Bangkok 1990 33 73 France, Bordeau 1982 15 n.d. Canada (average) 1984 15 16 Japan (average) 1990 11 13 * Tokyo 1990 15 35 Macao 1991 11 n.d Singapore 1994 6 9 Spain, Murcia 1993 25 h/ 22 USA (averag-.e) 1984 12 17 Notes: a/ LJp fhom 25% 'k in 1988. b/ Down from 45%: in 1989. Operational Indicators Water & Waste water Utilities Page 12 Set)I ... ..... ..--- ------- C.2. COMPOSITION OF UFW UFW is broken down by its two main components: physical [lpipe leaks and storage tank overflows] and comm-ercial [meter under-registrationi, illegal connections. etc.]. A good uniderstanding, of the relative weights of these com-ponents is a sine-qiua-non condition for the development of a sound program- to reduce UFW. Composition of UFW (%) Country/City Year Physical Commercial Total Singapore 1989 4 7 11 Spain, Barcelona 1988 1 1 12 23 Colombia, Bogota 1991 14 2)6 40 Costa Rica, San Jose 1990 2 1 2 5 46 C.3 UFW EFFECTIVE REDUCTION PROGRAMS linformation about four highly successful UFW reductionl pr-ogramis is presented hiere. These pr-ograms share one commonl approach: the initial effort was directed towards reducing commt-ercial losses: users were idenitified, the commnnercial system (meter reading and billing) wvas revamped, defective mieters were replaced and the number of metered connections was substantially increased. ReduiCtion of leaks was also par-t of the UFW reduction program but secondary to the reduction of commercial losses. a) Macao & Murcia UFW, Percent 45 40~~~~3 30 ------- -- - - - - - - - - - - - - - - - t 24 ~~~~~~~24 2012 I'WI 0 I 2 3 4 5 6 7 8 U Murcia U Macao Year of Program Note: Mur-cia: Year 0 = 1988 Mvacao: Year 0 = 1982 Water & Waste water Utilities Operational Indicators Set I Page 13 b) Singapore & Santiago UFW, Percent 30 29 28 276 25 2 20 .6 10 7- 7a 6 10:: ~~~~~~~*~~~' ~~6.7 6.4 6.0 5 0' 1989 1990 1991 1992 1993 1994 * Singapore U Santiago Year of Program C.4. SUSTAINABILITY OF UFW REDUCTION PROGRAMS UFW levels can easily deteriorate when a tight control on operations and maintenance and of the commercial system hard and software sub-systems that affect productivity levels, is not maintained. In this particular case, the deterioration is due to a relaxation of policies and accountability. Evolution of UFW a) Sao Paulo, Brazil, 1977-92 U FW . Percent 40) 20 1"77 1979 1981 1983 1985 19 89 19)9 19 92 Y ear Soarce: SA B ESP. Contract 085/92- C Rep.l. 03/93. Operational Indicators Water & Waste water Utilities Page 14 _ _Set I b) Bogota, Colombia, 1975-89 UFW, Percent 50 45 40 35 30 25- 20 15 - 10) CD Nl- co CD CD C\ CO LoIf CD N- co a) *n - N- X- N- D OD C OD CD aD CD CD X CD ou~~~~~r co co co a 0 a OD co 00 00 OD Souirce: Yepes Guillermo, Infrastructure Maintenance in LAC. The Costs of Maintenance Neglect and Options for Improvement. Vol. 3. June 1992. D. WASTE WATER COLLECTION SYSTEMS D.1. Length of Sewer Systems Length of the sewerage distribution system as a fuction of: * the number of people served [meters of pipes/person], - number of connections [nmeters/connection]. Unit lengrth: Country/City ] Year | mts/person mtslconnec Brazil (average) 1989 1.6 11.1 a/ * Brasilia 1989 1.2 11.9 b/ Chile * Valparaiso 1992 n.d. 9.8 * Santiago 1990 1.4 7.5 Colombia, Bogota 1992 0.9 6.0 France, Bordeaux 1982 n.d. 10.1 U.K ,Wessex. 1991 n.d. 5.2 Notes: a/ 1.7 units/sewerage connection. b/ 2.4 units/sewerage connection. D.2. Infiltration Flows in Sewer Systems USA, EPA guidelines.... less than 500 gallons/day/ini-dia per mile (465 liters/day/cm-dia per km) Water & Wastewater Utilities Operational Indicators Set I Page 15 E. WASTE WATER TREATMENT E.1. Typical Composition of Untreated Municipal Wastewater 1 Concentration Range bi: U.S. Constituent al Strong Medium Weak Average c/ Solids (total) 1,200 720 350 Dissolved, total d/ 850 500 250 * Fixed 525 300 145 * Volatile 325 200 105 Suspended 350 220 1(0 192 * Fixed 75 55 20 * Volatile 275 165 80 Settleable solids, ml/L 20 10 5 Biochemical oxygen demand, 5-day20C 400 220 110 181 Total organic carbon 290 160 80 102 Chemical oxygen demand 1.100 500 250 417 Nitrogen (total) 85 40 20 34 * Organic 1 35 15 8 13 * Ammonia 50 25 1 2 20 * Nitrite 0 0 0 0 * Nitrate 0 0 006 Phosphorus 15 8 4 9.4 * Organic 5 3 1 2.6 * Inorganic 10 5 3 6.8 Chlorides d/ 100 50 30 Alkalinity (as CaCO3) d/ 200 100 50 211 Grease 150 100 50 Total coliform bacteria 107-1(9 107-108 106-107 22x106 e/ (no./ 100 mL) Fecal coliform bacteria --- --- --- 8xl06 (no./100 mL) l Viruses, pfu/100 mL b/ - 500 Notes: a/ Values are expressecl in mg/L. except as rioted. d/ Values should be increased by amoLint in domestic water b/ After Metcalf & Eddy, lic., 199 1. supply. c/ CulIp et al.. 1979. e/ Geldreich, 1978. pfu= Plaque-fonning units/IOnL. Source: Water Reuse. Assessment Repoit Project 92 WRE-1. Enviro-nmeint Research Founldationi 1994. Operational Indicators Water & Wastewater Utilities Page 16 Set I E.2. Typical Constituent Removal Efficiencies for Primary and Secondary Treatment A verage Percent Removal Constituent Primary Secondary Treat!ent Treatment Activated Sludge Trickling Filter BOD 42 89 69 COD 38 72 58 TSS 53 81 63 NH3--N 18 63 Phosphorus 27 45 Oil and grease 65 86 --- Arsenic 34 83 Cadmium 38 28 --- Chromium 44 55 5 Copper 49 70 19 Iron 43 65 56 Lead 52 60 46 Manganese 20 58 40 Mercury I1 30 16 Selenium 0 13 0 Silver 55 7 Zinic 36 75 5 Color 15 55 56 Foaming agents 27 Turbidity 3 1 TOC 34 _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ Source: Water Reuse. Assessmient Repoil ProJect 9' WRE-I. Environment Research Foundation 1994. E.3. Removal of Microorganisms Expected Removal of Excreted Microorganisms in Various Wastewater Systems Treatment Removal (log 10 units) Process ' Bacteria Helminths Viruses Cysts Primary sedimentation Plain (0-I 0-2 (-1 0-1 Chemically assisted( 1 2 10 (h1 (_-I 0-1 Activated sludge(c' 0-2 0-2 (-I 0-1 Biofiltration(d) 0-2 0-2 ()-I 0-1 Aaerated lagoon(dl 1-2 1-2 0-1 Oxidation ditch 1-2 0-2 1-2 0-1 Disinfectione' 1 6ih) 0-1 0-4 0-3 Waste stabilization pondsr' 16h) 14 h) 1-4 1-4 Effluent stoage r 1-4 1-4 Notes: (a) Conventionial filtration is not included among the processes in the originial table. (b) Furthier reseatch is needed to confiriml performance. (c) Including secondary sedimentation. (d) Including settlitig pond. (e) Chlorinationi or ozonation. (f) Perforniance depends on number of ponds in series and other environmental factors. (gI) Performanice depends on retenitioni tine, which varies with demand. (h) With goocd designi and proper- operation, the recommended guidelinies are achievable. (i) A log 10 removal represenits a 90 percent reduction: 2 log 10 unlits represents 99 petcent removal. Source: E.P.A. ManItLal Guidelines for Water Reuse, Sept. 19922. Water & Wastewater Utilities Operational I ndicators Set I Page 17 F. PERSONNEL F.1. NUMBER OF STAFF Number of staff as a function of: * staff per thousand water connections (W/000) or per thousand water plus sewerage connections ([W + S]/000). * thousands of im of water sold per year per staff (m'[000]/staff). * kns. of pipes in the water supply system per staff (km/staff). * personis served [thousands] per staff (PS [000]/st). Staff Ratios Country! City Year W/000 W+S/00 00OOM3staff kinlstaff 000 PS/st 0 Belarus * Minsk 1993 n.a. n.d. 56 n.d. 0.7 * Gomel 1993 n.a. n.d. 20 n.d. 0.3 Belgium, Brussels 1 992 3.2 n.d. 105 3.3 3.0 Brazil (average) 1989 6.5 5.0 47 1.9 a/ 0.8 * Brasilia 1989 13.5 7.1 54 1.3 b/ 0.7 * Sao Paulo 1993 5.1 3.1 n.d. 2.1 0.8 Canada (average) 1984 2.0 n.d. 424 n.d. 1.7 Chile, Santiago 1990 2.1 1.1 191 4.1 2.5 Colombia. Bogota 1994 3.6 1.8 106 1.1 1.7 France, C. Banlieue 1987 4.5 n.a. 200 n.d. 2.2 Guinea (average) 1993 15.0 n.d. 8 n.d. n.d. Ivory Coast (average) 1995 I 4.8 n.d. 22 n.d. n.d. Japan (average) 1990 1.7 n.a. n.d. 7.0 1.7 Macao 1991 2.2 n.a. 148 n.d. n.d. Mexico, Monterrey 1987 4.1 2.2 86 2.2 1.5 Romania, Bucharest 1994 n.a. n.d. 75 n.d. 0.5 Senegal (average) 1993 8.6 n.d. 13 n.d. n.d. Spain * Alicante 1987 1.1 0.6 170 i.d. 1.9 * Murcia 1992 2.5 n.d. 165 4.9 1.6 Togo 1990 2 22.4 n.d. 26 3.3 0.5 Turkey, Bursa 1992 4.6 n.d. 40 0.4 0.9 USA (average) I 1990 1 2.7 n.d. 370 8.6 1.5 n.d. = data not available: na.a. = not applicable. Notes: a/ 1.3 water Lllits anic 1.7 seWerage units per conilection. b/ 2.3 water unlits and 2.4 sewerage UllitS per connectioni. Operationtal Indicators Water & Wastewater Utilities Page 18 Set I F.2. STAFF COMPOSITION CategorylLevel Country/City, Percentage of Labor Force Management Professional Clerical Blue Collar Brazil (range) 0.1 - 0.4 7 - 9 17 - 44 46 - 76 * Brasilia 0.2 12 24 64 Chile, Santiago 3 18 37 42 Macao 25 75 F.3. TRAINING EFFORT Country/City Year Training Chile, Esval-Valparaiso 1992 41% staff trained/yr 1.2 days/staff/yr France, C. Banlieue 1987 4% of salaries G. MISCELLANEOUS INDICATORS G.1. VEHICLES/1000 WATER CONNECTIONS This indicator includes all types of vehicles used in the operation and maintenance of the utility's system. When the utility also provides sewerage services the subscript (w + s) is added next to the figure. ComntrylCity Year Vehiclesl/ 000 connec Spain, * Muicia 1992 0.9 (w) * Alicante 1992 0.6 (w) Chile, Valparaiso 1992 0.4 (w+s) Washington, WSSC 1992 0.6 (w+s) _~~~~~~ __ Water & Wastewater Utilities Operational Indicators Set I Page 19 G.2. METER READING Number of consumiiption meters read per day per reader. Count,ylCity Meters read per day/reader Spain, Murcia (1992) 215 France (1986) 80 - 00 G.3. METER MAINTENANCE & REPLACEMENT PRACTICES Country Meter Replacement in Years Macao (1991) 15 Singapore (1991) 7 (domestic) 4 (large) Country Meter Testing Replacement Diameter (") Years Diameter (") Years USA (1984)a/ 5/8 -- 3/4 9 5/8 -- 3/4 17 (average) 1/ 1 7 1 16 4 4 4 13 6 3 6 12 8 3 8 11 1/. Meter costs are coming down. Therefore, in many cases it is most cost-effective to replace meters than to repair them. Source: a/ AWWA. opus cit. INDICATORS 2nd EDITION WATER & WASTEWATER UTILITIES SET 11 FINANCIAL INDICATORS 'I> Water & Wastewater Utilities Finanicial Indicators Set 2 Page 21 INTRODUCTION Only a selected group of financial indicators is presenited in thlis report. since the objective is not to overwhelm the reader with information that, in most cases, is unlikely to be of relevance. The technique of ratio analysis is a useful tool to analyze a utility's financial position. Ratio indicators presented here provide information about efficiency and operational performance, credit worthilness and liquidity and profitability. As such they provide insight into areas that merit further investigation but they do not. in themselves, provide definitive answers on the financial conditioni of a given utility. Some ratio indicators, such as contributioni to investmenit and rate of retunn, can be very volatile fi-om one year to the next. Therefore, to present a more realistic picture. they have been calculated as an average over an arbitrary three year period. Otherwise, the indicators reflect a one year perfonniance. based on information available between 1988 and 1994. The utilities represenited in this paper are from Latil America, Africa, EuL-ope/Cential Asia, andl South and East Asia. Sources of the data collected are recent Staff Appraisal Reports. Project Coompletion Reports and the utilities' annual financial reports. The sainple was selected taking into consideration diversity in geographical location, type and size of the utilities and service provided by them. Not surprisingly. complete information needed to calculate all ratios was not available for all the utilities. Therefore, some of the tables and graphs present infomiation only of a selected group of utilities. For the purpose of comparison. infomiation of W&S utilities in industrialized countries is presented in certain graphs and tables. Unfortunately, pertinent financial information available from these utilities is often consolidated thus making it impossible to obtain many of W&S financial indicators reported here for other utilities. All monetary values are expressed in US dollars. The exchange rates used for conversion are the average annual exchange rates for each coulitry as reported by the Intemnational Finance Statistics of the IMF. Finiancial Indicalors Waler & Wastewaler Utilities Page 22 Set 2 A. EFFICIENCY INDICATORS A.1. WORKING RATIO (WR) The WR is the ratio of operating costs to operating revenues. Operating costs in this ratio exclude depreciation and interest payments (but no debt service paymenits), a key difference with the Operating Ratio (OR) that includes these costs. Operating revenues remain the same flor both ratios. They include revenutes from water and sewerage tariffs, connection fees, well abstraction fees and re-connection fees. Working Ratio and Operating Ratio 3 2.5 2 1.5 1 0.5 0 0.5 1 1.5 2 2.5 0.47 0.19 Pusan 0.56 0.25 Kwangju 0.55 0.29 Morocco 1.06 0.38 Taejon 0.52 0.40 I Manila 0.49 0.43 Sao Paulo 0.75 0.47 Seoul 0.74 0.53 Santiago 0.71 0.58 Ankara 0.89 0.65 Valparaiso 0.73 0.66 Bogota 0.71 0.67 S.Catarina 0.76 0.68 I Minas Gerais 0.92 0.73 Costa Rica 1.21 0.73 Ghana 0.78 0.78 Karachi 1.21 0.81 Bursa 1.01 0.86 Annaba 0.93 0.86 Oran 0.95 _0.89 Changchun 2.47 1.43 2.30 Kaduna | 2.55 2 Katsina *Operating Ratio *Working Ratio Wlater & Waste water Utilities Financial Indicators Set 2 Page 23 Soulid finanicial management requires the WR to be well below 1. About 30% of the utilities have a WR lower than 0.50. and two utilities (9%) have a ratio larger than 1. The four Korean sewerage companies are among the utilities that have a WR lower than 0.5. Otherwise, it does not seem to be any signiilicant difference in working ratios between companies that provide W&S services and those that provide only one of these services. Caution should be used in interpreting this ratio when there is evidence that utilities are cuttinsg down on maintenance costs which would improve the WR but could lead to critical situationIs in the future. A.2. OPERATING RATIO (OR) The OR is the ratio of operating costs to operating revenues. In this case, operational costs include all the expenses together with depreciation and interest costs (but no debt ser-vice payments). Sound financial management requires that this ratio should also be less than 1. Nine utilities (41 %') have an OR less than 0.75, 32% an OR between 0.75 and less than I and 27% an OR greater than 1. The latter utilities, as in the case of Ghana and Nigeria. must rely on -ovemimient subsidies to cover their operational expenses. The same cautionary note made to the WR applies to the OR. In addition. caution should be exercised when assets are not revalued and therefore depreciation charges do not give a realistic value or whenl revaluation of assets is not consistently applied. There does not seem to be any sigiificant difference in ORs between companies that provide W&S services and those that provide only one of these services. Interestingly enoughi most of the utilities that have an OR larger than 1. also have an average UFW ratio larger than 40%c. Financial IndicalorsV Water & Wastewater Utilities Page 24 Set 2 A.3 ACCOUNTS RECEIVABLE/COLLECTION PERIOD (CP) This indicator, expressed in montlh equivalent of sales, is the ratio between the year-end accounts receivable and operating revenues, multiplied by 12. Of the 22 utilities with information on accouLnts receivable, 41 % have collection periods of less than 2 months, 20% between 2 and 4 montlhs, and 30% more thani 4 months. When the CP is increasing the company' cash flow can be in jeopardy. This is specially of concemn in countries where inflation is high. where no charges are levied against late payment or when these charges do not reflect the financial cost of borrowing money. Poor collection efficiency is mostly blamed otn consumers, and in some cases in particular- on public sector agencies. However, the water utility may also be at fault for delayed and faulty billings, inadequate responses to consumer's queries on billings, and a lukewann effort to collect overdue accounts. A common factor found among the utilities with poor collection efficiency is the lack of a clear policy to promote and entforce prompt payment (like disconnecting the service to consumers witlh ari-ears of more thani 2 to 3 mloniths). Accounts Receivable/Collection Period (Months equivalent) Kwangju 0.8 Changchun 0.8 Pusa 1.1 Minas 1.1 Seoul 1.2 Brussels 1.2 Taejon 1.4 SCatarn. 1.5 Sev.Trent 1.9 Sao Paulo 2.7 Bursa 2.8 Valpars. 2.9 Santiago 3 Bogota 3.3 C. Rica 4 Ankara 5.1 Manila 6.3 Karachi~~~~~~~~~MNMMM 7.6X Ghana 8. Annabal.9i Oran 9.9 Morocco _10.8| -- -T 1 1 - R 10 '~~~~~~~~~~~~~~~~~~~~~~11 W!ater & IVaste water Utilities Financial Indicators Set 2 Page 25 A.4. PERCENTAGE CONTRIBUTION TO INVESTMENT (CTI) The percentage contribution to investment shows the proportion of capital expenditures financed by the net internal cash gener-ated by the utility. This ratio is often calculated on a yearly basis and therefore depends oni the annual cash flow of the utility. As a consequence the CTI ratio. calculated on a yearly basis, can vary widely. Thus, to present a more balanced picture, the CTI has been calculated as the cash contribution over an arbitrary three year period. To provide a sense of the magnitude of investmiients, this ratio is contrasted with the relative value of new investments to fixed assets over the same period. In general, the data suggest, and not surprisingly, that utilities with relatively large investments have lower CTI ratios. For the utilities with infonnation available, the overall average contribution to investment rate is 40% and the investment over net fix assets is 13%. Twenty five percent have a negative or 0 CTI. 35% have CTI of less than 30%, 25% between 30 and 50%. and 15% CTIs larger thanl 50%. Percentage Contribution to Investment -40 -20 0 20 40 60 80 100 120 140 Santiago; _ 8.8: 124.3 i Bursa, 90.3 Annabai _-7.2 61.7 Valpars. 8 37.6 Manilaj 1 34.8 Ankara i 33.4 - ~~~~~31.2 Minas Q_ 3. Seoul _3 0 S.Catarin.; _16. 29.4 KwangjuC 22.3 Pusan i 44! Taejonj 17 Bogota 13 C. Rica! 11.2 Changchun; _ 35.9 Katsina! 15.8 Kaduna! 0 s 1.3 Morocco ml 20 Sao Paulo! -15.8 14.4 Sev.Trent -2417.0 Oran! 1.7 * Net Cash genr./Invs. * Inv./Avg.Net F. Asst Financial Indicators Water & Wastewater Utilities Page 26 Set 2 B. LEVERAGE INDICATORS B.1 DEBT SERVICE COVERAGE RATIO (DSC) The debt-service-coverage ratio measures the extent to which internal cash geineration covers total debt service.. As with the CTI indicator, the DSC has been calculated as the average of the last 3 years of information available. Thirty three percent of the utilities in the sample have a DSC less thani 1; that is, their cash generation is not adequate to cover debt service obligations. 28% have a DSC between I and 2 and 38% a DSC larger than 2. Pusan Bursa 997 16.6 Valparaiso 7.35 Ankara Santiaqo = Taejon 2.84 Kwanqiu Seoul Annaba Minas 1.8 Manila - Boqota C. Rica _ S.Catarina 1.2 Morocco * Oran Sao Paulo * 0.7 Katsina P Kaduna P -1 0 1 3 5 7 9 11 13 15 17 The last two utilities in the graph show a negative ratio. Water & Wastewater U'tilities Finanicial Indicators Set 2 Page 27 B.2 DEBT EQUITY RATIO (DER) This ratio is defined as Total Debt/Equity. The average DER for the sample of utilities is 0.40 which is considered quite acceptable: 29% of the utilities show a debt situation that is highly leveraged (over 0.50 and up to 1.50). Two water utilities in industrialized countries, Severn Trent in England (private) and the water utility of Brussels (public) both report a DER of 0.25. Not surprisingly, when comparing Graphs 3, 4. and 5 we find that utilities with the highest debt service coverage ratio also have the lowest debt equity ratio. DER is also affected by the revaluation of fixed assets and therefore caution should be exercised wlhen using this ratio. Bursa *0.05 Valparaiso Taeion SantiaaO 0.06 Chanachun Pusan Seoul 0.14 Ghana Kwanaiu Sev.Trent Brussels Morocco 0.27 Annaba Ankara C. Rica 0.34 Manila Boaota Met. Australia Karachi Sao Paulo 0.70 S.Catarina Oran Minas Kaduna. 1.5 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 Financial Indicators Water & Wastewater Utilities Page 28 Set 2 C. LIQUIDITIY INDICATOR C.1. CURRENT RATIO (CR) This ratio is computed by clividing curTent assets by current liabilities. Current assets include cash, accounts receivable and inventories. The CR measures the short-run paying ability of the utility. From the data we obseive that 75% of the utilities have a curTent ratio of less than one. e.g.. most do seem to have short-tenr liquidity problems. However, this observation should be taken with caution as this ratio does not provide infonration on utility's capacity to pay and collect its bills promptly. Such is the case of Ghania, for examliple. whichi has a CR of almost 2 and a collection period of 8 monthis. Bursa I1n.6 Seoul Pusan Taejon 4. 4 Valparaiso KwangjuI::: Santiago I 1.9 Ghana S.Catarina Changchun 1.5 Annaba Manila Karachi 1.2 Ankara S.Paulo St. Costa Rica 0.9 Sev.Trent 0.82 Oran Morocco Bogota 0.3 Minas _ _ 0 2 4 6 8 10 12 Water & Wastewater Utilities Finanzcial Indicators Set 2 Page 29 D. PROFITABILITY INDICATORS The ratios examined thus far provide some information about the operations of the utility. Profitability ratios on the other hand show the combined effects of liquidity, asset management and debt management on operating results. D.1. RETURN ON NET FIXED ASSETS (RR) This indicator measures the productivity of fixed assets in use, expressed as the ratio between net operating income and net fixed assets. The RR has been calculated as the average over a three year period. The median RR of the 18 utilities is 3%: 29% of the utilities report a negative RR. another 29% a RR between 0 and 5%, 27% a RR between 5 and 10% and 13% a RR higher than 10%. However, we should be cautious when comparing this ratio since it runs into the probleni mentioned before related to the revaluation of fixed assets. In addition, it is not uncommon to observe that it would be in the interest of an utility to undervalue its fixed assets in order to meet or show a higher RR target. On the other hand, it is also quite common to find in balance sheets an account "work in progress" which very often includes a large portion of works completed but which are not included in the RR calculation. For instance, EAAB-Bogota reports 17% of its fixed assets as work in progress, while EMOS-Santiago reports 5%. S.Catarina i 20.0 Ankara ' -!11.9 Sev.Trent 1 1.0 Oran - 8.0 Manila 8.0 Sao Paulo 7.0 Santiaqo 7= 0 Annaba 7.0 Pusan I= .3 Bogota Minas 5.0 Morocco O 2.9 Seoul 2.6 Kwanqiu _2.2 C. Rica _1.6 Brussels 1.5 *Valparaiso l *Taeion *Chancchun Bursa -1.6 Ghana -2.4 Kaduna -11 Katsina _ -25 -20 -15 -10 -5 0 5 10 15 20 Percentaqe * Slightly negative. Financial Indicators WVater & Waste water Utilities Page3O Set2~~~~------ --- I,.,~1~-.~-,,~-.-"-,,-..".---"--,~-,,.--",-,- --------- e D.2. RETURN ON EQUITY (RE) Return on equity shows the return to the owners, expressed as the relationship between net inicome (net income after interest paymients) and equity (total assets minus liabilities). As for the RR the RE has been calculated as thie average for three years. The RE also suffers from the problems associated with the revaluation of fixed assets which would lower the value of equity which in turn would increase the RE. The mediani RE is 0%; 58% of the uitilities show a negative RE. 8% a RE between 0 and 3% and 25% a RE between 3 and 10%. Only two utilities shows a RE larger than 10%. The RR anid REs behave, in general, in a similar way. However, five utilities (Kar-achi, Sao Paulo, Minas. Orani and Morocco.) shiow completely opposite reSuIltS in both ratios. This is explained by its low contribution to investment ratio and thus a high debt to equity ratio. Sev.Trent j Ankara 1. S.Catarina 8 Pusan 6. Manila 6.5 Boqota 6.3 Santiaqo~ Annaba Seoul Kwangiu 100.5 *Valparaiso *Chanqchun 'C. Rica *Taejonp Morocco d1o3 Bursa 16 Oran - Ghana -3.3 - Sao Paulo -7 Katsina -10.2 Karachi -14.2 I Minas -21 J Kaduna -2 i__ -30 -25 -20 -15 -10 -5 0 5 10 15 Per-centage Sligrhtlv negative. Water & Wastewater Ltilities Financial Indicators Set 2 Page 31 E. OPERATIONAL RATIOS E.1. PERSONNEL E.1. 1 PERSONNEL COSTS Personnel costs are expressed as a ratio to total operating costs (depreciation and debt service excluded). Depreciation and debt service are excluded due to lack of unifonnity in treating revaluation of fixed assets and to facilitate comparison of utilities with and without debt service obligations. As indicated in Infrastructure note W5 - 12 (Annex 2), staff productivity index (See E.1.2 below) and personnel costs related to operational costs should be examined simultaneously. Personnel Costs vs. Operating Costs Percent 70 60 50 - 40 30 20 Germally Fiance G.Britain Spain Japan C'oloniibia Bazil C. Rica Mexico Tunkia lTurkey Thailand Nepal E.1.2 STAFF PRODUCTIVITY INDEX (SPI) This ratio is an important measure of the efficiency of a water and/or sewerage utility. It relates the number of staff with the number of connections. Sixty percent of the utilities with W&S services have a SPI of 4 or less ([w + sl connections). 20% between 4 and 7 and 20% more than 7. The SPI for utilities in some African utilities, that only provide water services is extremely high (over 30 [wj conniection). The four sewerage utilities of Korea. on the other hand, have very low SPI's of under 2 ([s] connection). As a guideline, it would appear that a SPI of less than 4 could be considered adequate but still with room for improvement. In some cities, particularly in Eastern Europe. residential consumers live in large apartment buildings where consumption in apartments is not individually metered. In similar circumstances, the practice in Brazil is to report simultaneously the number of apartments (economifus) and the number of connections. In these cases this SPI index may not be particularly meaningful. Alternative indicators, to handle this situation, can be found in Set I, Section F. 1. Financial Indicators Water & WVastewater Utilities Page 32 Set 2 It is also important to remember that a reduction in the SPI ratio cannot necessarily be interpreted as an increase in efficiency. To complete the analysis of staff productivity, as mentioned previously, expenditures on personnel also need to be examinled (personnel costs as a % of operational costs). There have been cases of utilities with staff/connection ratios decreasing while staff costs, in proportion to operating costs. are incr-easinig as slhowin in Graplh 1. In addition, it is also important to examine the staff composition which might show important imbalances or inadequate number of qualified middle-level managers and technical staff. Not surprisingly, utilities with large personnel costs show a low contribution to investments and a low debt service coverage ratio (Set It A.4 and B. I). such as in Sao Paulo and Costa Rica. Number Staff /000 Conn.11' % Personnel Costs/ Oper. Costsh Water & Sanitation Utilities: 32.0 Ghana 36.6 13.7 Manila 47.8 C. Rica 6.1 64.3 Ankara 3.5 36.0 _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ 6 9 . 0 S. Paulo 4.1 71.3 S.Catarina 6.0 BLirsa 2. 71.1~ I f ~~~~~~~~~~~~~~~76.11 Bogota, . BW70t ~2.0K Valparaiso 1.8 46.1 SantiaeoI ' L1.1 I 54.5 Onlv Water Utilities: 38.0 Kaduna3 17.4 33.0 t i Katsinza 26.3 30.8w 64.1 ' Annaba _ 64.1 Oran 11.0 55.2 Chanigcehunl 1 8.0 _ 11.1 I l Only Sewerage Utilities: Pusan 1.7 35.9 Seoul 1.6 36.0 Tae.jon 1.4 * 24.4 Kwangju 1.7 U 56.7 10 5 0 20 40 60 80 Notes: a/ Iicludes water and sewerage connections tor W&S Utilities bt/ Opelatiolnal Costs eXclUde Depreciationi and Debt Service. WVater & Waste water Utilities Financial Indicators Set 2 Page 33 Graphic I Number of Staff/'000 Connections vs. Personnel Costs/m3 Sold Bogota (1982-1991) 140 Personnel Costs/rn sold Staff/lOQO water connections 6 120 _ _ _ 5 100 ~^ ~ - - -~ 4 80 3 60 2 40 1 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 Year Personnel Costs 0 Staff/'000 water connections E.2 COMPOSITION OF OPERATIONAL COSTS The two main categories of operating costs are often personnel and fuel/energy conisumptioni. Other operating cost components include chemnicals, mainitenance and miscellaneous. Depreciation charges are not included. Bogota S.Catarinaom Bursa Sao Paulo Costa Rica Kwangju Minas Manila Valparaiso Ghana Ankara Pusan _ Katsina Tae jon Kaduna Changchu___ i 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% * Personnel * Energy 0 Other Finiancial Indicators Water & Waste water Utilities P 34 ~~~~~~~~~~~~~~~~~~~Set 2 Range of Operational Costs Low Value Average Value High Value Personnel 11.1 % 46.0 % 73.4 % Energy 5.3 % 18.4 % 44.0 % Other 11.5 % 35.6 % 6-3.6% No information is available for industrialized countries for comlparison purposes. BecaulSe operating, costs are utility specific and given the paucity of data at this time, any further elaboration at this point is not warranted. E.3 UNIT OPERATIONAL COST (UOC) Unit operationial cost (operational costs"' / nm' produced) var-ies fromn US$0.05 per cubic mieter in Karachi and Changchun to US$0.31 per cubic myeter in Minas Gerais. About 28% of the utlities have an UOC below 10 cents per cubic meter. 44% an UOC between 10 and 20 cents per cubic meter. and 28% an UOC of 20 cents or higher. Contrary to what could be expected. we do not see evidence, in this sample, thlat large utilities have lower UOCs. This could be explained, in part, by specific site conditions and by wide variationis of domestic costs anid price levels in the different countries. We believe, however, that major differenices in UOC can be explained by the quality of ser-vice provided by the utilities. US$ Unit Operation Cost IliX M3 Produced 0.31 0.30- 0.26 0.2 5 -0.24 0.20 -0.20 0.15 -0.14 0.14 0.11 ~~0.12 0.12 0.10.12 0.10 - ~0.11 0.110.1 0.09 0.090.0 0.05 - ~~~~~~~~~~~~~~~~~0.05 0.05 0.05 0. 0 B . Excluding depreciation. INDICATORS 2nd EDITION WATER & WASTEWATER UTI LITI ES SET III OVERVIEW OF TARIFF RATES AND STRUCTURES Water & Wastewater Utilities Overview of Tariff Rates and Structures Sel 3 Page 35 OVERVIEW OF TARIFF RATES AND STRUCTURES A. 1 BACKGROUND An overview of water and sewerage tariff rates and structures, from a sample of 9 cities' believed to be representative of practices in developing countries. is presented in this section. This overview provides some insights into how rates are structured and offers some recommendations (based on hypotheses that need further testing), into the problems such structures may be causing and pitfalls to avoid. This overview is thus a complement to the set on1 financial ratios but is not intended to be a thorough discussion of pricing issues. An understanding of the implications of tariff structures and of the possible distortions that they may be causino is an important step in the design of an strategy and action plan to improve the financial position of a utility based on sound economic principles. A. 2 TARIFF STRUCTURE Satisfactory tariff levels must provide adequate funds to meet operations, debt service and capital expansion requirements. Tariffs should also encourage efficiency in the use of resources; and many would argue. on fairness grounds. that the tariff structure should make these services affordable by the poor. Reconciliation of all these objectives remains an elusive task. All tariff rates, except Singapore and Ankara. reviewed hele have two components: a fixed charge and a volumetric charge (related to consumptioll). The first is often intended to cover the fixed costs of the utility and the second the variable ones. In addition, most tariff rates are progressive. e.g., the volumetric charge increases as consumption increases. Rate progressivity is often the result of the decision to provide a cross-subsidy from some, presumably wealthy, groups to other groups with a more limited capacity to pay (the poor, schools. hospitaLls, etc.). More recently, rate progressivity has found its defenders among those interested in promoting water conservation. Whatever the argument. progressivity introduces economic distortions in the use of water, which need to be but are often poorly understood. In addition, low rates often discourage utilities from reacling the poor or reducinCg UFW. Most utilities in the sample have also flat charges, independent of consumption. either by design or fiat (when meters are not installed or operative) to deal with non-metered consumption. None of the utilities in this sample fully meters consumption. In all documented cases here, sewerage charges are a fixed percentage of the water charges. These charges range from a low 18% in San Jose to 100% in Santa Catarina. The Ankata. Bogota. Ghana. Manila, San Jose, Santa Cataritna, Sao Paulo. Singapore and Seoul. Sources for the data on water tarifts are Staff Appraisal Reports and fronm the Asian Development Bank publication "Water Utilities Data Book" (for Singapore. and Seouil. Overview of Tariff Rates and Structures Water & WVaste water Utilities Page 36 Set 3 average lies around 40% to 50%2 . In some cases, like Ankara, the tariff charged for water also covers sewerage services. A. 3 DOMESTIC TARIFF All the utilities in this sample have a progressive tariff structure (normally 3-4 block rates up to about 60 m' of consumption per month per connection-m3/m/c). Rate progressivity applies also to non-residential users. As shown in Graphs A, B, C and D, we find that: there seems to be little consistency in how consumption block intervals are established; * all utilities offer a subsidized base line consumption. intended to benefit some residential users. This base line is in the range of 5 to 20 m3/m/c; * relatively high fixed charges imposed on base line consumption often negate the subsidy intended (Graphs F, G. and H). Fixed-charges are also applied to non-residential users, but most often they are set at higher levels; * if 2Om3 per month is taken as a reasonable household consumption in developing countries, household payment for this volume ranges from US$2 to US$12 per month. As a way of comparison, households in some European cities pay for this consumption (1991): US$2.60 in Milan, US$5.40 in Rome. US$14.40 in Paris. and US$29.60 in Brussels (Table 3): for unmetered connections. the utility often makes an estimate of the monthly consumption based on past consumption patterns or other criteria and charges a corresponding, de facto, fixed charge. - It should be noted that the outcome is that incremental costs of seweraee. which are 1.5 to 3 times those of water. are even less well covered bv user charges than water supply costs. Water & Wastewater U'tilities Overview of Tariff Rates and Structures Set 3 Page 37 Table 3 Comparison of annual water charges in industrialized countries for a family of four in a house and consuming 200 m3/year [137 liters/capita/dayl (prices in 1991 US dollars) Charge Charge Charge Country/City US$/m Country/City US$/m3 Country/City US$/m3 Austria Germany Luxembourg * Linz 0.64 * Berlin 1.00 * Luxembourg 1.34 * Salzburg 0.99 * Dusseldorf 1.56 * Vienna 1.07 * Frankfurt 1.66 Netherlands * Gelsenwasser 1.95 * Amsterdam 0.94 Belgium * Hamburg 1.59 * The Hague 1.17 * Antwerp 0.68 * Munich 1.05 * Utrecht 0.57 * Brussels 1.48 * Stuttgart 1 .79 * Liege 0.93 Spam Hungary * Madrid 0.84 Denmark * Budapest 0.21 * Barcelona 0.90 * Aarhus 0.72 * Miskolc 0.72 * Seville 0.51 * Copenhagen 0.60 * Pecs 0)8 * Alicante 0.46 * Odense 065* Murcia 0.98 Italy Finland * Bologna 0.49 Switzerland * Helsinki 0.94 * Milan 0.13 * Beme 0.63 * Tampere 1.01 * Naples 0.62 * Geneva 2.12 * RomeO.7 Zuih16 France * Turin 0.27 1.63 * Banlieue/Paris 1.46 0.25 United Kingdom * Lyon 1.52 Japan * Bristol 1.23 * Marseille 1.52 * Nagoya 041 * Cardiff 1.72 * Nice 1.51 * Osaka 0* London 0.88 * Paris 0.72 * Sapporo 0.25 * Manchester 1.29 * Tokyo 0.57 * Newcastle 1 .37 * Yokohama 0.45 Upon Tyne 0_0.41 L _ _ __ Low Value (Milan) US$ 0.13 /m3 Average US$ 0.96 /m3 High Value (Geneva) US$ 2.12 /m' Notes: a/ Average b/ Metered Consumption Overview of Tariff Rates and Strutetures Water & Wastewater U.tilities Page 38 Set 3 A. 4 AVERAGE CHARGES AND AVERAGE INCREMENTAL COST Information oni rate structuLre, average charges (AR) paid by consumers and average incremental cost (AIC) for water services provided by the utilities of Ankara, Sao Paulo and Manila. is presented in graphs E. F and G. It is worth noting, in these graphs. that all these utilities are subsidizing a large segment of domestic users (fiinancial subsidy: AR less average charge paid by consumner; economic subsidy: AIC less average paid by consumiier). Although not shown, small commercial and industrial consumers are also often subsidized. A. 1 Domestic Water Tariff Structure Tariff Rate US$/m3 1.4- Sta. Catarina (9) () 8 jS~~~~~~~~~~ao Paulo ('89) 0).6_ ; . 0.4 - -- ().2_ _ J w San Jo-----se ('9 2 ) O 1 .-. L i , 10 20 30 40 50 60 70 Monthly Consumption m3/c A. 2 Household's Monthlv Expenditure on Water Expenditure in US$ per month 50 Sta. Catarina ('91) 40 Sao Paulo ('89) 30 20 10 0 10 20 30 40 50 60 70 Monthly consumption m3/c Water & Wastewater Utilities v'erview of Tariff Rates and Structures Set 3 Page 39 B. 1 Domestic Water Tariff Structure Tariff Rate US$/m3 0.8 ~~~~~~~~Ankara ('891) 0.6 Ghana ('88) 0.4 0.2 Manila ('88) 0 0 1 0 20 30 40 50 60 70 Monthiv Consumption m3/c B. 2 Household's Monthly Expenditure on Water 60 - Expenditure in US$ per month 50 - ------- / Ankara ('89) _, ~Ghana ('88)_ 20 -- - -- -- -- -- -- - - - - - - - - - ----------- - -- -- -- - - -------------- - -- - -- -- --- 0 0 10 20 30 40 50 60 70 8() Monthly Consumption m3/c Overview of Tariff Rates and Structures Wt ater & Wastewater Utilities Page 40 Set 3 C.1 Domestic Water Tariff Structure Tariff Rate US$/m3 0.7 ......................... 0.6 Singapore ('91) 0.5 0.4 Seoul ('91) 0.3 -... . 0.1 0 ( 10 20 30 40 50 60 70 Monthly Consumption m3/c C. 2 Household's Monthly Expenditure on Water 40 Expenditure in US$ per month 30 Singapore ('91) 20 10 0 10 20 30 40 50 60 70 Monthly Consumption m3/c Water & Wastewater Utilities Overview of Tariff Rates and Structures _Set 3 _ _Page 411 D. Household Payment for Water Consumption US$ per month 1 2 I 0 8 6 4 0 Seoul Manila S.Paulo Singapore Ghana S. Jose S.Catarina Ankara U 5m' * 10 m3 20 m3/month E. Ankara's Tariff Structure and Average Incremental Cost (1989) US$/m3 1 -__ Actual Tariff 0.8 0.6 * Average Price 0.4 + Avg. Increm. Cost 0.2 0 I I I 0 10 20 30 40 50 60 70 Monthly Household Consumption m3 Source: Staff Appraisal Report November 1989. Overview of Tariff Rates ard Structures 4ater & Wt'astewater ltilities Page 42 Set 3 Sao Paulo's Tariff Structure and Average Incremental Cost (1989) US$/mn3 -+ Average Price Manila's TActual Tariff 0.5- -- - - - - - - ~~ Average Prwe~Av. ierm os 01 _ 0q I III 0 10 20 30 40 50 60 70 Monthly Household Consumption mY Source: Stalff Appraisal Report May 1 s 99. G. Manila's Tariff Structure and Average Incremental Cost (1988) 0.6- 0.5- 0.4- \+Average Ptice * ~~~~Actual Tariff- 0.2--?= _H 0.1 i Z + ~~~~~~Avg. Increm Cost 0 10 '20 30 40 50 60 70 Monthly Household Cnu ti mA Source: Staff Appraisal Report AugusNt 1989. Water & Wastewater Utilities Overview of Taniff Rates and Structures Set 3 Pa~ge A. 5 RATE DISCRIMINATION BY CONSUMER GROUP Tariff structures, in this sample of utilities, are often designed in such way that industrial, commercial, domestic and other users are subject to different rates for the same consumption (rate discrimination). This practice seems to be rooted in the belief that industrial and other users, presumably wealthy ones, should subsidize all or some of the domestic consumers and on occasion special groups such as schools, churches and the public sector. This cross subsidy element is often substantial as shown in the following table: Rate Discrimination Ratio of Average User Rate to Average Rate User Utility Sanl Jose al Sta. Catarina Bogota Bursa Minsk Domestic 0.76 0.75 0.86 b/ 0.86 0.05 Commercial 1.79 2.51 1.63 1.32 Industrial 1.82 2.85 1.51 1.32 3.84 c/, 2.03 d/ Pub. Sector 0.76 4.11 1.06 1.32 2.63 Other 0.76 0.69 0.34 Average Rate 1.00 1.00 1.00 1.00 1.00 a!. San Jose industrial rates differentiate on the basis of water being part of the final product (e.g., soda water and beer). b/. Average of 6 categories for domestic consumers. The corresponding ratio within each category varies from about 0.32 to 2.71 and the relative fixed charge from I to 100. c/. Drinking water dI. Non-treated water No explicit rationale for price discrimination or progressivity in rate structure was provided in any of the utilities' reports. The implicit rationale seems to be "charge whatever it is thought the market can bear", without analysis of the consequences on the price discrimination on the behavior of different consumer groups, and of the distortions created. Overview of Tariff Rates and Structlures Water & Waste water LUtilities Page 44 Set 3 H. Price Discrimination Average Tariff Among Consumer Groups US$/in3 1.2 0.4 0 Manila Bogota San Jose S. Catarina Ankara Barcelona 1988 1992 1992 1991 1989 1992 -Avg Tariff U Avg Dom.Trf. M Avg Cornm.Trf. M Avg Ind.Trf. A. 6 WATER BILLINGS, CONSUMPTION AND USERS. Consumption pattems in all users groups are often accentuated by discriminatolry pricing practices as a resl]t of cross-subsidies. The following table presents infonrmationl for San Jose. Manila and Bogota. These differences in consumption and billings must be fully uniderstood when analyzing the impact of tariff reforms and programs to correct large tariff distortions. Water Consumption and Billing (Percentages) San Jose Manila Bogota Type of User %LU %Q %$ %_L/ 1-Q %$ %U %Q %$ Domestic 91 71 55 90 61 42 95 76 69 Commercial 6 8 13 9 32 47 5 8 14 Industrial 2 15 27 1 7 11 0.5 6 11 Public Sector I 6 5 --- --- --- 9.3 5 6 Other --- --- 0.( 4 0.3 Total 100 100 100 100 100 100 100 100 100 Totals miay not ad(d because of rounding. U = nuiber of usels: Q = total consuiliption; $ = total billing. Suggestedl reading: a) Nieswianidonilv. Michael L. "Estiniatini Urbhan Residential Water Demiiand: Ef'fects o' Price StnictUre Conservation and EducIatioll." Water Resouices Research. Vol. 'S. March 1992. hi) K. L. Kollar andL P. 1\acAuley. "Water Reqo irereriets ftii 1Idustrial Dcvelopnieit. ' Jounial AkWXVA, January 198(0. Water & Wastewater Utilities O'ernview of Tariff Rates and Structures Set 3 Page 45 A. 7 CONCLUSIONS AND HYPOTHESES TO BE TESTED: Based on this limited sample, some prelimiiary conclusiolns and hypotheses can be drawn, that need further testing: Issue Conclusion/Hypothesis 1. Tariff Structure a) to reach the same level of revenues, for a given consumption volume, it is preferable, on efficiency grounds, to opt for relatively highier volumetric charges and relatively lower fixed charges. Volumetric charges require metering; b) water and sewerage rates shoLIld be separated. Otherwise some users are charged for one service that they do not receive often (sewerage) and end up subsidizing those who receive both: c) cumbersomiie classification of users, based on notional ideas of capacity to pay, are difficult and expensive to maintain up to date ancl are pronie to abuses and corruption; d) consumption estimates based on a proxy sLcic as area of construction, imputed or actual property values or size of industry are prone to substantial errors that are best colTected by metering consumption. 2. Fixed Charges High fixed charges can have two undesirable etfects: a) they do not promote water conservation: and b) they can be regressive e.g., they often penalize consumers that use less water and, in particular, low income groups. 3. Price Discrimination Price discrimination can also have undesirable effects: a) industry, in particular. may look for alternative sources thus reducing the commercial base of the utility- b) perceptions of unduly high price discriminationi can lead to fi-aud: and c) low charges remove incentives for utilities to provide services in low income areas. They also remove incentives to reduce UFW. Water & Wastewater Utilities ANNEX 2nd Edition Page I PERSONNEL COSTS AS AN INDICATOR FOR W&S UTILITY PERFORMANCE IN DEVELOPING COUNTRIES' Karin Kemper, Guillermo Yepes and Mike Gamn Personnel costs in many Water & Sanitation utilities in developing countries constitute a larger cost factor than usually recognized. draining resources from maintenance and other necessary operating expenses and imposing costs on consumers. Although most developing countries have invested RATIO OF PERSONNEL COSTS TO OPERATING considerable amouLits of money in W&S, new COSTS investments are still essential in the sector due to rapid population growth, large unmet needs, and a sizeable When the finanicial statements of sector companies in backlog of postponed maintenance expenditures. developing countries are examined, it becomes clear that: However, internal generation of funds by W&S utilities (a) the impact of personnel costs on the cost structure in has been disappointing])' low and consequently the many companies is very high, and (b) the ratio of pressure on national and local governments to provide personnel costs to operating costs varies widely, even funds for investments, and often for operations. is large. from company to company in the same country. It is becoming obvious that needed levels of investment The ratio of personnel costs to operating costs (PC/OC) cannot be sustained over the long run if business for selected W&S utilities in Great Britain. France. continues as usual. Germany. Japan and Spain and selected developing countries is presented in Figure 1. In spite of the Personnel costs are not usually expected to play a relatively high unit labor costs in the group of predominant role in W&S utilities since other operating industrialized countries, PC/OC ratios are, on average, costs. e.g. for energy, chemicals, and maintenance, weigh under 40 percent. Yet, as illustrated, some utilities in heavily. (Operating costs, as treated here, do not include developing countries have PC/OC ratios in excess of 60 depreciation or interest payments). There is evidence, percent. The average for Brazil and Colombia, for however, that personnel costs in a number of developing example. is around 55 percent, considerably higher than countries are disproportionately high in comparison to the average for the industrialized countries (Figure I). other operating costs and should receive closer attention when assessing company performance. Figure 1: Personnel Costs Pcrsonniel Cosls/Operatin, Costs '.' 70 If the ratio of personnel costs to other operating costs is too high. the sustainability of previous investments may 60 be undercut. The need for external funds increases. 50 draining more resources than necessary from other 40 sectors, and consumers may have to be charged higher prices if maintenanice and other necessary operating 30 expenses are to be met. Thus, although there is a substantial need to expand W&S services in developing 20 countries. the inefficient allocation of resources within the lo companies may not even permit mainitenianice of current 0 levels of service. let alone allow investmenits from GeCn-n. B, C p k, - bil C Ri', Tl Thand internally generated funds. Selected Counitries Published in Infrastructure Notes. January 1994 - Transportation. Water and Urban Development Department, World Bank. W & S No. WS 12. ANNEX Water & Wastewater Utilities Page 2 2nd. Edition The personilel cost figures for the industrialized countries ratio. Thus, the problem would be mechanistic, and if usually include pension costs. However, in many of the personnel costs were perceived as too high. the solution developing countries, reported personinel costs do not would be to reduce staff numbers. include all costs. In some documents cases, W&S companies provide additional fiinge benefits to their staff The staff/connection indicators alone, however, do not such as free water and sanitation services, subsidized give a satisfactory picture of the situation. An analysis of housing and housinig loans. educational benefits for family a Colombian sample of 16 W&S utilities shows that in 75 members, etc.. These costs are not always identified in percent of the cases the staff/conn1ection indicator was the financial statements or easily quantified. It is improving (decreasing) or stationary while the ratio of reasonable then. to assume that the personinel costs and personnel costs to operating costs was increasing. In one related ratios presented here for developing countries are Colombian company, the staff/connection ratio for the last underestimated. 10 years has been declining steadily and has reached levels very few companies in the region can emulate. RELATIVE PRICES However, personnel costs (in real terms) and their impact on the cost of water delivered have been increasing at an One explanation for the larger share of personnel costs in alarming rate to the point that the company is now near developing countries compared to industrialized countries insolvency. As a result the company has also reduced might be that labor in the former is relatively cheaper and allocations for maintenance, and investment plans are in therefore substituted for capital. This would increase, the abeyance. relative share of personnel costs in these utilities' accounts. However, this would not explain wide This is not an isolated case. Average salary levels and intracountry differences. There exist substanitial their impact on the cost per mndelivered are -rowinlg very differences between companies withini countr-ies. For fast, in real terms, in many W&S companies in instance in Ecuador, personnel costs in two companies (leveloping countries. In a sample of Bank-financed constitLite only 39 percenit of total operating costs, which projects implemented between 1965 and 1987. the ratio of corresponds well to the average observed in industrialized salaries to the cost of water sold grew in real terms (over a countries while two other companies exhibit personnel 6 to 10 years period) at unsustainable rates. costs over 60 percent. Similar differences can be observed in Colombia and Brazil. Thus even if it seems Figure 2 illustrates that the staff/1000 connections reasonable to expect that utilities in developing countries indicator does not tell the whole story. As can be seen will use somewhat more labor than those in industrialized from the sample of 32 Asian and Pacific cities, there is no countr-ies this would only explain parts of the picture. di-ect connection between staff/1000 connections and the ratio of personnel costs to operating costs. For example, ECONOMIES OF SCALE in Singapore the staff/1000 connections ratio is a favourable 2.4, but personnel costs constitute 49 percent Another explanationi for the large differences between of operating costs. And in Bombay. there are 61 staff per companies might be their size. Due to economies of 1000 connections, but personnel costs amount to only 32 scale, larger companies could be expected to have lower percent of operating costs. personnel costs per unit of service ($mn sold). One study (Yepes. 1990) observed economies of scale in a Brazilian Figure 2: Asian and Pacific Cities sample. and also data from the UJS (1970, 1976) indicate that there are economies of scale. However. the * Pers.Costs/Oper.Costs * No.of Staff/lOOOConn. Colombian sample illustrates that economies of scale n. cannot in all cases be counted on. Economies of scale are n50 l hX observable when measured as staff/10)000 water 4( 40 connections ratio. but when measured against personnel 30 cost/mi sold, these economies do not exist. 2I) 20 WHAT DOES THE "NUMBER OF STAFF PER 1000 0 0 CONNECTIONS" TELL US? W&S Utilities in Asian and Pacific Cities Number of staff per 1000 water connections or 1000 The examples above give an indication that the frequently water & sewerage connections are frequently used as Llsed staff/1000 coniiections ratio in itself may be indicators of company performanice. When related to misleading and that an improvement in that ratio cannot personnel costs. the picture would be presumed to be necessarily be interpreted as an increase in efficiency. fairly straiogltforward. i.e. companies with high personinlel Personliiel costs in these cases appear not to be costs would be assumed to have a higher staff/connection automatically linked to staff numbers or economies of Water & Wastewater Utilities ANNVEX 2nd Edition Page 3 scale, so that it must be assumed that other factors play a combination of indicators and a determiniationl of the role. trade-offs, between their implications are necessary if A recent survey carried out by the Unionl Africaine Des company performanice is to be adequately analysed. To Distributeurs D'Eau in West Africa arnoli its members complemenit the picture, the expenditure on personinel also suggests that inadequate iiformationi on personinel costs is needs to be examined. Personnel costs can. tor example. an additional problem in W&S utilities. While most of be related to othel operatino costs andl to the cost of water the companiies were able to provi(le all kinds of data soldl. relating to turinover, energy consLimptioll, number of personinel etc.. only three of the 14 respondin- companies In this Note, only technical explanations for high provided data on personiniel costs. personinel costs, suci as relative prices and economiiies of scale, are examined. As could be seen. althouhll these certainly provide answers in some cases. they do not tell A LOOK AT STAFFING STRUCTURES the whole story. Complemenitary explanations are provi(le(l by institutionial economiiics. This evolving field The data lead to the assumption that in many W&S of economilic theory uses, inter- alia, public choice theory. utilities averace salaries and wages are hioher than one as well as conicepts such as the Soft Budget Constraint. would expect. But this nee(d not be the case for the and principal-agenit relations for the analysis and overall workforce. Ther-e is evidence from sonie studies explanation of company and sector performance. that salaries of lower-level workers in public utilities are incentive structures and stakeholders' behavior. substantially higher than the average in the private sector. while wages at managemiienlt level are lower. The latter Concerninig the perfor-mlianice of the utilities discLissed in seems to be especially true for smaller- utilities. w*hich this Note, explanations for highli personinlel costs may for consequenitly have difficulties in attracting qualified example be that the various actors (mitinanaers. employees. managers. Bank reports menition these problems politicians. etc.) are subject to incenitive structuLes that concerning the W&S sectors in counitries as differenit as lead to over/underpayments of certain employees, to Turkey and Nigeria. demilands for higher than appropriate salaries and to overstaffing. This is indicated in sector reports for sonie Furthermore. the repor-t on the W&S sector in T'urkey countries. For future analyses, it would thus be of initer-est points out that the combined effect of a reductioll of UFW to look at incenitive structures in the W&S sector by by 20 percenlt and an increase in productivity of personnel employingC the analytical framework of institutionial by 21) percenlt in urban areas on savinas and revenues economiiics. This would give additional guidance in policv would add about US$180 millioni a year to the cash formiulationi for specific settings. generationi potential of local aiencies. That WLould imply that a significant part of the cash gener-ation effort recommenided at the local level could be achieved without a major effect on existing tariff levels. Similar TO LEARN MORE: calculations for Costa Rica aii( Colombia also point to poteiitial savings. It is thus evident that the opportunity OstronII Elinor. Larry Schroeder. and SUsaIn )Wvnne. 1993. cost of high personiniel expenditures is significant. listin/i//I IIjies and Sleaiale Develn(II. Boulder: Wesiview Press. CONCLUSIONS AND POLICY IMPLICATIONS Triche. Thelma. 1992. Prixah Secfor Partcipation in lnftia. ci rla/nre Serl, l ics in Ni.,,, eim -The (llrban Wato Sup p/, To suLm111marize, personinel costs constitute a sizeable part Ssi tor." TWtUWS Backiround ! Paper, uLnpublished. of overall operating costs in mianiy companiies in a wide variety of developinig countr-ies. Thlis need not be the case World Bank. 1993. "Colnnibia -Review ,f ICVate'r and, since countries with generally higlher- wages and salaries, SeweragNe Sector Ia.ifolio/o>." Report No. 11560. Washineton. such as Britain. Spaii and Japan. man1age to keep D.C. personnel costs at a relatively low level. Excessive personinel costs draini funds from other necessary Yepes. Guillermo. 1990. MMana"'enlient an/d O1erlaianall expendituires for e.g. mainltenianice and investilents for P/inraMes of Mi//ici ipa ald Re,c,'imial Water and Sewer ae expansion. This Note thus iniplies that wheni analysing- Coc,anie.i ini Latili Atucricia ciiiald t Caribbean. W xvortld Baik. W&S utilities' perforim1anice. personnel costs oug7ht to be InfraStrCIure & Urban Developinent Dept.. DiSCuSsion Paper, Report INU 61. Washingitoni. D.C. taken into accounlt because theii- weight might have serious imiplicationis for a companlys t'inancial sustainability: routinely used technical indicators. such as staff/!1000 connection1s mighlt be misleading because theil relation to finanicial performance is not clear: and a -